T H E T I ME L E S S W A Y O F B U I L D I N G

The Timeless Way of Building is the first in a series of

books whi ch describe an enti rel y ne w atti tude to archi t ec-
ture and pl anni ng. T h e books are i ntended to provi de a
compl ete wor ki ng al ternati ve to our present ideas about ar-
chi tecture, bui l di ng, and pl a nni ng a n al ternati ve whi c h
wi l l , we hope, gradual l y repl ace current ideas and practices.
volume I THE TIMELESS WAY OF BUILDING
volume 2 A PATTERN LANGUAGE
volume 3 THE OREGON EXPERIMENT
Center for Environmental Structure
BERKELEY, CALIFORNIA
THE
TIMELESS WAY
OF BUILDING
Christopher Alexander
with love and thanks
to Ingrid, Sara, and Peter
NE W Y OR K
O X F O R D U N I V E R S I T Y P RES S
1
979
Library of Congress Catal ogi ng in Publication Data
Alexander, Christopher.
The timeless way of building.
i . Architecture. 2. Pattern perception. I. Ti t l e.
NA2500. A45 yzo'.i 76-42650
ISBN - 13 978 -0-19-502402-9
printing, last digit: 30 29 28
Copyright 1979 by Christopher Alexander
Printed in the United States of America
on acid free paper
To you, mind of no mind, in whom
the timeless way was born.
ON READING THIS BOOK
Wh a t lies in this book is perhaps more i mportant as a
whol e than in its details. If you only have an hour to
spend on it, it makes much more sense to read the whol e
book roughl y in that hour, than to read only the first t wo
chapters in detail. For this reason, I have arranged each
chapter in such a way that you can read the whol e chapter
in a couple of minutes, simply by readi ng the headlines
whi ch are in italics. If you read the begi nni ng and end of
every chapter, and the italic headlines that lie bet ween
them, t urni ng the pages almost as fast as you can, you
wi l l be able to get the overal l structure of the book in
less than an hour.
Th e n , if you want to go into detail, you wi l l know
where to go, but al ways in the context of the whol e.
CONTENTS
T H E T I ME L E S S W A Y
1. The timeless way 3
T H E Q U A L I T Y
2. The quality without a name 19
3. Being alive 41
4. Patterns of events 55
5. Patterns of sface 75
6. Patterns which are alive 101
7. The multiflicity of living fatterns
S. The quality itself 137
T H E G A T E
9. The flower and the seed 157
zo. Our -pattern languages 167
vii
C O N T E N T S
ii. Our -pattern languages (conl.) 193
72. The creative -power of language 21 1
1 j. The breakdown of language 225
14. Patterns which can be shared 243
75. The reality of patterns 277
16. The structure of a language 305
1 j. The evolution of a common language
for a town 325
T H E W A Y
18. The genetic power of language 351
19. Differentiating space 365
20. One pattern at a lime 385
27. Shaping one building 403
22. Shaping a group of buildings 427
23. The process of construction 455
24. The process of repair 475
25. The slow emergence of a town 493
26. Its ageless character 51 1
T H E K E R NE L OF T H E WA Y
27. The kernel of the way 531
vi u
DETAILED TABLE OF CONTENTS
T H E T I MEL ES S WA Y
A bui l di ng or a t own wi l l onl y be al i ve to t he e x-
t ent that it is g o v e r n e d by t he t i mel ess wa y .
1. It is a process which brings order out of nothing
but ourselves; it cannot be attained, but it will happen
of its own accord, if we will only let it.
T H E Q U A L I T Y
T o seek t he t i mel ess wa y we must first know t he
qual i t y wi t hout a name.
2. There is a central quality which is the root criterion
of life and spirit in a man, a town, a building, or a
wilderness. This quality is objective and precise, but
it cannot be named.
3. The search which we make for this quality, in our
IX
D E T A I L E D T A B L E OF C O N T E N T S
own lives, is the central search of any person, and the
crux of any individual person's story. It is the search
for those moments and situations when we are most
alive.
4. In order to define this quality in buildings and in
towns, we must begin by understanding that every
place is given its character by certain patterns of
events that keep on happening there.
5. These patterns of events are always interlocked
with certain geometric patterns in the space. Indeed,
as we shall see, each building and each town is ulti-
mately made out of these patterns in the space, and
out of nothing else: they are the atoms and the mole-
cules from which a building or a town is made.
6. The specific patterns out of which a building or a
town is made may be alive or dead. To the extent they
are alive, they let our inner forces loose, and set us
free; but when they are dead, they keep us locked in
inner conflict.
7. The more living patterns there are in a place
a room, a building, or a townthe more it comes to
life as an entirety, the more it glows, the more it has
that self-maintaining fire which is the quality without
a name.
8. And when a building has this fire, then it becomes
a part of nature. Tike ocean waves, or blades of grass,
its parts are governed by the endless play of repetition
x
D E T A I L E D T A B L E OF C O N T E N T S
and varietycreated in the presence of the fact that all
things pass. This is the quality itself.
T H E G A T E
To reach the quality without a name we must then
build a l i vi ng pattern l anguage as a gate.
g. This quality in buildings and in towns cannot be
made, but only generated, indirectly, by the ordinary
actions of the people, just as a flower cannot be made,
but only generated from the seed.
10. The people can shape buildings for themselves,
and have done it for centuries, by using languages
which I call pattern languages. A pattern language
gives each person who uses it the power to create an
infinite variety of new and unique buildings, just as
his ordinary language gives him the power to create
an infinite variety of sentences.
11. These pattern languages are not confined to vil-
lages and farm society. All acts of building are gov-
erned by a pattern language of some sort, and the
patterns in the world are there, entirely because they
are created by the pattern languages which people
use.
XII
D E T A I L E D T A B L E OF C O N T E N T S
12. And, beyond that, it is not just the shape of
towns and buildings which comes from pattern lan-
guagesit is their quality as well. Even the life and
beauty of the most awe-inspiring great religious build-
ings came from the languages their builders used.
/ j . But in our time the languages have broken down.
Since they are no longer shared, the processes which
keep them deep have broken down; and it is therefore
virtually impossible for anybody, in our time, to make
a building live.
14. To work our way towards a shared and living lan-
guage once again, we must first learn how to discover
patterns which are deep, and capable of generating
life.
15. We may then gradually improve these patterns
which we share, by testing them against experience:
we can determine, very simply, whether these pat-
terns make our surroundings live, or not, by recog-
nizing how they make us feel.
16. Once we have understood how to discover indi-
vidual patterns which are alive, we may then make a
language for ourselves for any building task we face.
The structure of the language is created by the net-
work of connections among individual patterns: and
the language lives, or not, as a totality, to the degree
these patterns form a whole.
XII
D E T A I L E D T A B L E OF C O N T E N T S
77. Then finally, from separate languages for (lifer-
ent building tasks, we can create a larger structure
still, a structure of structures, evolving constantly,
which is the common language for a town. This is
the gate.
T H E W A Y
Once we have built the gate, we can pass through it
to the practice of the timeless way.
18. Now we shall begin to see in detail how the rich
and complex order of a town can grow from thousands
of creative acts. For once we have a common pattern
language in our town, we shall all have the power to
make our streets and buildings live, through our most
ordinary acts. The language, like a seed, is the genetic
system which gives our millions of small acts the
power to form a whole.
19. Within this process, every individual act of build-
ing is a process in which space gets differentiated. It
is not a process of addition, in which preformed parts
are combined to create a whole, but a process of un-
folding, like the evolution of an embryo, in which
the whole precedes the parts, and actually gives birth
to tJiem, by splitting.
20. The process of unfolding goes step by step, one
pattern at a time. Each step brings fust one pattern to
xui
D E T A I L E D T A B L E OF C O N T E N T S
life; and the intensity of the result defends on the in-
tensity of each one of these individual stefs.
21. From a sequence of these individual fat terns,
whole buildings with the character of nature will
form themselves within your thoughts, as easily as
sentences.
22. In the same way, groufs of feofle can conceive
their larger fublic buildings, on the ground, by fol-
lowing a common fattem language, almost as if they
had a single mind.
23. Once the buildings are conceived like this, they
can be built, directly, from a few simfle marks made
in the groundagain within a common language, but
directly, and without the use of drawings.
24. Next, several acts of building, each one done to
refair and magnify the froduct of the previous acts,
will slowly generate a larger and more comflex whole
than any single act can generate.
25. Finally, within the framework of a common lan-
guage, millions of individual acts of building will
together generate a town which is alive, and whole,
and unfredictable, without control. This is the slow
emergence of the quality without a name, as if from
nothing.
xiv
D E T A I L E D T A B L E OF C O N T E N T S
26. And as the whole emerges, we shall see it take
that ageless character which gives the timeless way its
name. This character is a specific, morphological char-
acter, sharp and precise, which must come into being
any time a building or a town becomes alive: it is the
physical embodiment, in buildings, of the quality with-
out a name.
T H E K E R NE L OF T H E WA Y
And yet the timeless way is not complete, and wi l l
not f ul l y generate the quality without a name, until
we leave the gate behind.
27. Indeed this ageless character has nothing, in the
end, to do with languages. The language, and the
processes which stem from it, merely release the fun-
damental order which is native to us. They do not
teach us, they only remind us of what we know al-
ready, and of what we shall discover time and time
again, when we give up our ideas and opinions, and
do exactly what emerges from ourselves.
xv
T H E
T I M E L E S S
W A Y
A building or a town wi l l only be
alive to the extent that it is governed
by the timeless way.
CHAPTER I
THE TIMELESS WAY
It is a process which brings order out of
nothing but ourselves; it cannot be
attained, but it will happen of its own
accord, if we will only let it.
There is one timeless way of building.
It is thousands of years old, and the same today as
it has always been.
The great traditional buildings of the fast, the
villages and tents and temples in which man feels at
home, have always been made by people who were
very close to the center of this way. It is not possible
to make great buildings, or great towns, beautiful
places, places where you feel yourself, places where
you feel alive, except by following this way. And, as
you will see, this way will lead anyone who looks for
it to buildings which are themselves as ancient in
their form, as the trees and hills, and as our faces are.
It is a process through whi ch the order of a building or a
town grows out directly from the inner nature of the
people, and the animals, and plants, and matter whi ch are
in it.
It is a process whi ch al l ows the life inside a person, or a
famil y, or a t own, to flourish, openly, in freedom, so
vividly that it gives birth, of its own accord, to the natural
order whi ch is needed to sustain this life.
It is so powerful and fundamental that with its help
you can make any building in the world as beautiful
as any place that you have ever seen.
Once you understand this way, you will be able to make
your room al i ve; you will be able to design a house to-
gether with your f ami l y; a garden for your chil dren;
7
T HE TI MELESS WAY
places where you can wor k; beautiful terraces where you
can sit and dream.
It is so powerful, that with its help hundreds of people
together can create a town, which is alive and vibrant,
peaceful and relaxed, a town as beautiful as any town
in history.
Wi t hout the help of architects or planners, if you are
worki ng in the timeless way, a t own will grow under
your hands, as steady as the fl owers in your garden.
And there is no other way in which a building or a
town which lives can possibly be made.
Thi s does not mean that all ways of maki ng buildings
are identical. It means that at the core of all successful
acts of building and at the core of all successful processes
of growt h, even though there are a million di fferent
versions of these acts and processes, there is one f unda-
mental invariant feature, which is responsible for their
success. Al t hough this way has taken on a thousand dif-
ferent forms at di fferent times, in di fferent places, still,
there is an unavoidable, invariant core to all of them.
Look at the buildings in the photographs which start
this chapter.
The y are alive. Th e y have that sleepy, awkwar d grace
which comes from perfect ease.
And the Al hambra, some tiny gothic church, an old
8
T HE TI MELESS WAY
Ne w Engl and house, an Al pi ne hill village, an ancient
Zen temple, a seat by a mountain stream, a courtyard
filled with blue and yel l ow tiles among the earth. Wh a t is
it they have in common? Th e y are beautiful, ordered,
harmoni ousyes, all these things. But especially, and
what strikes to the heart, they live.
Each one of us wants to be able to bring a building or
fart of a town to life like this.
It is a fundamental human instinct, as much a part of
our desire as the desire for children. It is, quite simply,
the desire to make a part of nature, to complete a worl d
which is already made of mountains, streams, snowdrops,
and stones, with something made by us, as much a part of
nature, and a part of our immediate surroundings.
Each one of us has, somewhere in his heart, the
dream to make a living world, a universe.
Those of us who have been trained as architects have this
desire perhaps at the very center of our lives: that one
day, somewhere, somehow, we shall build one building
which is wonderf ul , beautiful, breathtaking, a place where
people can walk and dream for centuries.
In some f orm, every person has some version of this
dream: whoever you are, you may have the dream of one
day building a most beautiful house for your family, a
garden, a fountain, a fishpond, a big room with soft light,
flowers outside and the smell of new grass.
In some less clear fashion, anyone who is concerned
9
T HE TI MELESS WAY
with towns has this same dream, perhaps, for an entire
t own.
And there is a way that a building or a town can
actually be brought to life like this.
Ther e is a definable sequence of activities which are at
the heart of all acts of building, and it is possible to spec-
i fy, precisely, under what conditions these activities will
generate a building which is alive. Al l this can be made
so explicit that anyone can do it.
And just so, the process by whi ch a group of indepen-
dent people make part of a t own alive can equally be
made precise. Agai n, there is a definable sequence of ac-
tivities, more complex in this case, which are at the heart
of all collective building processes, and it is possible to
specify exactly when these processes will bri ng things to
life. And, once again, these processes can be made so ex-
plicit, and so clear, that any group of people can make
use of them.
This one way of building has always existed.
It is behind the building of traditional villages in Af ri ca,
and India, and Japan. It was behind the building of the
great religious buildings: the mosques of Islam, the mon-
asteries of the middle ages, and the temples of Japan. It
was behind the building of the simple benches, and clois-
ters and arcades of Engl i sh country t owns; of the moun-
tain huts of Nor way and Austri a; the roof tiles on the
10
T HE TI MELESS WAY
walls of castles and palaces; the bridges of the Italian
middle ages; the cathedral of Pisa.
I n an unconscious f orm, this way has been behind al-
most all ways of building for thousands of years.
But it has become possible to identify it, only now, by
going to a level of analysis which is deef enough to
show what is invariant in all the different versions of
this way.
Thi s hinges on a form of representation which reveals all
possible construction processes, as versions of one deeper
process.
First, we have a way of looking at the ultimate con-
stituents of the envi ronment: the ultimate "t hi ngs" which
a building or a t own is made of. As we shall see, in
chapters 4 and 5, every building, every town, is made of
certain entities which I call patterns: and once we under-
stand buildings in terms of their patterns, we have a way
of looking at them, whi ch makes all buildings, all parts of
a town similar, all members of the same class of physical
structures.
Second, we have a way of understanding the genera-
tive processes whi ch give rise to these patterns: in short,
the source f rom whi ch the ultimate constituents of build-
ing come. As we shall see in chapters 10, 1 1 , and 12,
these patterns al ways come from certain combinatory
processes, whi ch are di fferent in the specific patterns
whi ch they generate, but al ways similar in their overall
structure, and in the way they work. Th e y are essentially
11
T HE TI MELESS WAY
like l anguages. And again, in terms of these pattern l an-
guages, all the di fferent ways of building, although dif-
ferent in detail, become similar in general outline.
At this level of analysis, we can compare many dif-
ferent building processes.
The n, once we see their differences clearly, it becomes
possible to define the difference between those processes
which make buildings live, and those which make them
dead.
And it turns out that, invariant, behind all processes
which allow us to make buildings live, there is a single
common process.
Thi s single process is operational and precise. It is not
merely a vague idea, or a class of processes which we can
understand: it is concrete enough and specific enough, so
that it functions practically. It gives us the power to make
towns and buildings live, as concretely as a match gives
us the power to make a flame. It is a method or a disci-
pline, whi ch teaches us precisely what we have to do to
make our buildings live.
But though this method is precise, it cannot be used
mechanically.
Th e fact is, that even when we have seen deep into the
processes by which it is possible to make a building or a
12
T HE TI MELESS WAY
town alive, in the end, it turns out that this knowl edge
only brings us back to that part of ourselves which is for-
gotten.
Al t hough the process is precise, and can be defined in
exact scientific terms, finally it becomes valuable, not so
much because it shows us things whi ch we don' t know,
but instead, because it shows us what we know already,
only daren' t admit because it seems so childish, and so
primitive.
Indeed it turns out, in the endy that what this method
does is simply free us from all method.
T h e more we learn to use this method, the more we find
that what it does is not so much to teach us processes
we did not know before, but rather opens up a process
in us, which was part of us already.
We find out that we already know how to make
buildings live, but that the power has been f rozen in us:
that we have it, but are afraid to use it: that we are crip-
pled by our fears; and crippled by the methods and the
images which we use to overcome these fears.
And what happens fi nal l y, is that we learn to overcome
our fears, and reach that portion of our selves which
knows exactly how to make a building live, instinctively.
But we learn too, that this capacity in us is not accessible,
until we first go through the discipline which teaches us to
let go of our fears.
And that is why the timeless way is, in the end, a
timeless one.
T HE TI MELESS WAY
It is not an external method, whi ch can be imposed on
things. It is instead a process whi ch lies deep in us: and
only needs to be released.
The power to make buildings beautiful lies in each
of us already.
It is a core so simple, and so deep, that we are born with
it. Thi s is no metaphor. I mean it literally. Imagi ne the
greatest possible beauty and harmony in the wor l dt he
most beautiful place that you have ever seen or dreamt
of. Y o u have the power to create it, at this very moment ,
just as you are.
And this power we have is so firmly rooted and co-
herent in every one of us that once it is liberated, it will
al l ow us, by our individual, unconnected acts, to make a
town, without the slightest need for plans, because, like
every living process, it is a process whi ch builds order out
of nothing.
But as things are, we have so far beset ourselves with
rules, and concepts, and ideas of what must be done
to make a building or a town alive, that we have be-
come afraid of what will happen naturally, and con-
vinced that we must work within a "system" and with
"methods" since without them our surroundings will
come tumbling down in chaos.
We are afraid, perhaps, that without images and methods,
chaos will break loose; worse still, that unless we use im-
14
T HE TI MELESS WAY
ages of some kind, ourselves, our own creation will itself
be chaos. And why are we afraid of that? Is it because
people will laugh at us, if we make chaos? Or is it, per-
haps, that we are most afraid of all that if we do make
chaos, when we hope to create art, we will ourselves be
chaos, hol l ow, nothing?
Thi s is why it is so easy for others to play on our fears.
Th e y can persuade us that we must have more method,
and more system, because we are afraid of our own chaos.
Wi t hout method and more method, we are afraid the
chaos which is in us will reveal itself. And yet these
methods only make things worse.
The thoughts and jears which feed these methods
are illusions.
It is the fears which these illusions have created in us,
that make places which are dead and lifeless and artificial.
Andgr eat es t irony of al l i t is the very methods we in-
vent to free us from our fears which are themselves the
chains whose grip on us creates our difficulties.
For the fact is, that this seeming chaos whi ch is in us is
a rich, rolling, swelling, dying, lilting, singing, l aughi ng,
shouting, cryi ng, sleeping order. If we will only let this
order guide our acts of building, the buildings that we
make, the towns we help to make, will be the forests and
the meadows of the human heart.
To purge ourselves of these illusions, to become free
of all the artificial images of order which distort the
15
T H E T I ME L E S S WA Y
nature that is in us, we must first learn a discipline
which teaches us the true relationship between our-
selves and our surroundings.
Then, once this discipline has done its work, and
pricked the bubbles of illusion which we cling to now,
we will be ready to give up the discipline, and act as
nature does.
This is the timeless way of building: learning the
disciplineand shedding it.
1 6
T H E Q U A L I T Y
To seek the timeless way we must
first know the quality without a
name.
CHAPTER 2
THE QUALITY WITHOUT A NAME
There is a central quality which is the
root criterion of life and spirit in a man,
a town, a building, or a wilderness. This
quality is obfective and precise, but it
cannot be named.
We have been taught that there is no objective differ-
ence between good buildings and bad, good towns
and bad.
The fact is that the diference between a good build-
ing and a bad building, between a good town and a
bad town, is an objective matter. It is the diference
between health and sickness, wholeness and divided-
ness, s elf -maintenance and selj-destruction. In a world
which is healthy, whole, alive, and self-maintaining,
people themselves can be alive and self-creating. In a
world which is unwhole and self-destroying, people
cannot be alive: they will inevitably themselves be
self-destroying, and miserable.
But it is easy to understand why people believe so
firmly that there is no single, solid basis for the
difference between good building and bad.
It happens because the single central quality which
makes the difference cannot be named.
Th e first place I think of, when I try to tell someone
about this quality, is a corner of an Engl i sh country gar-
den, where a peach tree gr ows against a wal l .
T h e wal l runs east to west ; the peach tree grows flat
against its southern side. Th e sun shines on the tree and
as it war ms the bricks behind the tree, the war m bricks
themselves war m the peaches on the tree. It has a slightly
dozy quality. T h e tree, careful l y tied to gr ow flat against
the wal l ; warmi ng the bricks; the peaches growi ng in the
sun; the wild grass gr owi ng around the roots of the tree,
in the angle where the earth and roots and wal l all meet.
25
T HE QUALI TY
Thi s quality is the most fundamental quality there is in
anythi ng.
It is never twice the same, because it always takes its
shape jrom the particular place in which it occurs.
In one place it is calm, in another it is stormy; in one per-
son it is ti dy; in another it is careless; in one house it is
l i ght; in another it is dark; in one room it is soft and
quiet; in another it is yel l ow. In one family it is a love of
picnics; in another danci ng; in another playing poker; in
another group of people it is not family life at all.
It is a subtle kind of freedom from inner contradic-
tions.
A system has this quality when it is at one with itself;
it lacks it when it is divided.
It has it when it is true to its own inner forces; lacks
it when it is untrue to its own inner forces.
It has it when it is at peace with itself; and lacks it
when it is at war wi th itself.
You already know this quality. T h e feeling for it is the
most primitive feeling which an animal or a man can
have. Th e feeling for it is as primitive as the feeling for
our own wel l -bei ng, for our own health, as primitive as
the intuition which tells us when something is false or
true.
But to grasp it fully you must overcome the prejudice
26
T HE QUALI TY WI T HOUT A NAME
of physics whi ch tells us that all tilings are equally alive
and real.
In physics and chemistry there is no sense in which
one system can be more at one with itself than an-
other.
And no sense at all in whi ch what a system "ought to be"
grows naturally from " what it is." Ta ke , for example,
the atoms whi ch a physicist deals wi th. An atom is so sim-
ple that there is never any question whether it is true to
its own nature. At oms are all true to their own natures;
they are all equally real ; they simply exist. An atom
cannot be more true to itself, or less true to itself. And
because physics has concentrated on very simple systems,
like atoms, we have been led to believe that what some-
thing "i s, " is an entirely separate question from what it
"ought to be" ; and that science and ethics can' t be mixed.
But the view of the world which physics teaches,
powerful and wonderful as it is, is limited by this
very blindness.
In the worl d of complex systems it is not so. Most men
are not ful l y true to their own inner natures or fully
" r eal . " I n fact, for many people, the effort to become
true to themselves is the central problem of life. Wh e n
you meet a person who is true to himself, you feel at once
that he is "more real " than other people are. At the hu-
27
T HE QUALI TY
man level of complexity, then, there is a distinction be-
tween systems whi ch are true to their "i nner nat ure, "
and those whi ch aren' t. Not all of us are equally true to
our inner nature, or equally real, or equally whol e.
And exactly the same is true in those l arger systems,
outside us, which we call our worl d. Not all parts of the
worl d are equally true to themselves, equally real, equally
whol e. In the worl d of physics, any system whi ch is self-
destroying simply ceases to exist. But in the worl d of
compl ex systems this is not so.
Indeed, this subtle and Complex freedom from inner
contradictions is just the very quality which makes
things live.
In the worl d of living things, every system can be more
real or less real, more true to itself or less true to itself.
It cannot become more true to itself by copying any
externally imposed criterion of what it ought to be. But it
is possible to define a process whi ch will tell you how the
system can become more true to itself, in short what it
"ought to be, " only according to what it is.
Thi s oneness, or the lack of it, is the fundamental
quality for any thing. Whet her it is in a poem, or a man,
or in a building full of people, or in a forest, or a city,
everythi ng that matters stems from it. It embodies every-
thing.
Yet still this quality cannot be named.
28
T HE QUALI TY WI T HOUT A NAME
Th e fact that this quality cannot be named does not
mean that it is vague or imprecise. It is impossible to
name because it is unerringly precise. Wor ds fail to cap-
ture it because it is much more precise than any word.
T h e quality itself is sharp, exact, wi th no looseness in it
whatsoever. But each word you choose to capture it has
f uz z y edges and extensions which blur the central mean-
ing of the quality.
I shall try to show you now, why words can never cap-
ture it, by circling round it, through the medium of half
a dozen words.
The word which we most often use to talk about the
quality without a name is the word "alive."
Ther e is a sense in whi ch the distinction between some-
thing alive and somethi ng lifeless is much more general,
and far more prof ound, than the distinction between
living things and nonl iving things, or between life and
death. Thi ngs whi ch are living may be lifeless; nonliving
tilings may be alive. A man who is wal ki ng and talking
can be alive; or he can be lifeless. Beethoven' s last quar-
tets are alive; so are the waves at the ocean shore; so is a
candle flame; a tiger may be more alive, because more in
tune with its own inner forces, than a man.
A wel l -made fire is alive. The r e is a worl d of di ffer-
ence between a fire which is a pile of burni ng logs, and a
fire whi ch is made by someone who really understands a
fire. He places each log exactly to make the air between
29
T HE QUALI TY
the logs just right. He doesn't stir the logs with a poker,
but while they are burni ng, grasps each one, and places it
again, perhaps only an inch from where it was before.
Th e logs are so exactly placed that they form channels
for the draft. Wa v e s of liquid yel l ow flame run up the
logs when the draft blows. Each l og gl ows with full in-
tensity. T h e fire, watched, burns so intensely and so
steadily, that when it dies, finally, it burns to nothi ng;
when the last gl ow dies, there is nothing but a little dust
l eft in the fireplace.
But the very beauty of the word "alive" is just its
weakness.
Th e overwhel mi ng thing that stays with you is that the
fire lives. And yet this is a metaphor. Li teral l y, we know
that plants and animals are alive, and fire and music are
not alive. If we are pressed to explain why we call one fire
alive and another dead, then we are at a loss. T h e meta-
phor makes us believe that we have found a word to
grasp the quality wi thout a name. But we can only use
the word to name the quality, when we already under-
stand the quality.
Another word we often use to talk about the quality
without a name is "whole."
A thing is whol e accordi ng to how free it is of inner con-
tradictions. Whe n it is at war with itself, and gives rise to
forces whi ch act to tear it down, it is unwhol e. T h e more
46
T HE QUALI TY WI T HOUT A NAME
free it is of its own inner contradictions, the more whole
and healthy and whol ehearted it becomes.
Compare the trees al ong a wild and wi ndbl own lake,
with an eroded gul l y. These trees and branches are so
made that when the wi nd blows they all bend, and all
the forces in the system, even the violent forces of the
wi nd, are still in balance when the trees are bent; and be-
cause they are in balance, they do no harm, they do no
violence. Th e configuration of the bending trees makes
them self-maintaining.
But think about a piece of land that is very steep, and
where erosion is taking place. Ther e aren' t enough tree
roots to hold the earth together, let's say; the rain falls,
in torrents, and carries the earth down streams which
form gullies; again, the earth is still not bound together
because there aren' t enough plants there; the wind
bl ows; the erosion goes f urt her; next time the water
comes, it runs in the very same gullies, and deepens
t hem; and widens them. Th e configuration of this sys-
tem is such that the forces which it gives birth to, which
arise in it, in the l ong run act to destroy the system. Th e
system is sel f-destroying; it does not have the capacity to
contain the forces whi ch arise within it.
Th e system of the trees and wi nd is whol e; the system
of the gul l y and the rain is unwhol e.
But the word "whole" is too enclosed.
It suggests closure, containment, finiteness. Wh e n you
call a thing whol e, it makes you think that it is whole
3
1
T HE QUALI TY
unto itself, and isolated from the worl d around it. But a
l ung is whol e, only so l ong as it is breathing oxygen from
the air outside the organi sm; a person is whol e only so
l ong as he is a member of some human group; a town is
whole only so l ong as it is in balance with the surround-
ing countryside.
Th e word carries a subtle hint of self-containment.
And sel f-containment al ways undermines the quality
which has no name. For this reason, the word " whol e "
can never perfectly describe this quality.
Another facet of the quality which has no name is
caught by the word "comfortable."
T h e word "comf ort abl e" is more profound than people
usually realize. T h e mystery of genuine comf ort goes far
beyond the simple idea that the word first seems to mean.
Places which are comfortable are comfortable because
they have no inner contradictions, because there is no lit-
tle restlessness disturbing them.
Imagi ne yourself on a wi nter afternoon with a pot of
tea, a book, a reading light, and t wo or three huge pillows
to lean back against. No w make yourself comfortabl e.
Not in some way which you can show to other people,
and say how much you like it. I mean so that you really
like it, for yourself.
Yo u put the tea where you can reach it: but in a place
where you can' t possibly knock it over. Yo u pull the light
down, to shine on the book, but not too brightly, and so
that you can' t see the naked bulb. Yo u put the cushions
32
T HE QUALI TY WI T HOUT A NAME
behind you, and place them, careful l y, one by one, just
where you want them, to support your back, your neck,
your ar m: so that you are supported just comfortabl y,
just as you want to sip your tea, and read, and dream.
Wh e n you take the trouble to do all that, and you do
it careful l y, with much attention, then it may begin to
have the quality whi ch has no name.
Yet the word "comfortable" is easy to misuse, and
has too many other meanings.
Ther e are kinds of comf ort whi ch stultify and deaden
too. It is too easy to use the word for situations which
have no life in them because they are too sheltered.
A family with too much money, a bed whi ch is too
soft, a room which al ways has an even temperature, a
covered path on whi ch you never have to wal k out in the
rain, these are all "comf ort abl e" in a more stupid sense,
and so distort the central meani ng of the word.
A word which overcomes the lack of openness in the
words "whole" and "comfortable " is the word "free."
The quality without a name is never calculated, never
perfect; that subtle balance of forces only happens when
the ideas and images are left behind; and created with
abandon.
Thi nk of a truck, filled wi th bags of cement. If the
bags are stacked perfectly, in lines, it may be careful , and
intelligent, and quite precise. But it will not begin to
33
T HE QUALI TY
have the quality wi thout a name, until there is a certain
freedom there: the men who piled the bags, runni ng, and
throwi ng them, forgetti ng themselves, throwi ng them-
selves into it, lost, wi l d. . . .
And a steel mill too can have this quality because its
freedom and its wildness show there, blazing in the night.
And yety of course, this freedom can be too theatrical:
a pose, a form, a manner.
A building which has a " f r e e " f o r ma shape without
roots in the forces or materials it is made of i s like a
man whose gestures have no roots in his own nature. Its
shape is borrowed, artificial, forced, contrived, made to
copy outside images, not generated by the forces inside.
Tha t kind of so-called freedom is opposite to the qual-
ity which has no name.
A word which helps restore the balance is the word
"exact."
T h e word " exact " helps to counterbalance the impression
of other words like "comf ort abl e" and " f r ee. " These
words suggest that the quality without a name is some-
how inexact. And it is true that it is loose and fluid and
relaxed. But it is never inexact. T h e forces in a situation
are real forces. The r e is no getti ng round them. If the
adaptation to the forces is not perfectly exact, there can
be no comfort, and no freedom, because the small forces
which have been left out will al ways work to make the
system fail.
34
T HE QUALI TY WI T HOUT A NAME
Suppose that I am tryi ng to make a table for the black-
birds in my garden. In wi nter, when the snow is on the
ground, and the blackbirds are short of food, I will put
food out for them on the table. So I build the table; and
dream about the clusters of blackbirds which will come
flocking to the table in the snow.
But it is not so easy to build a table that will really
work. T h e birds follow their own l aws; and if I don' t
understand them, they just won' t come. If I put the table
too l ow, the birds won' t fly down to it, because they
don' t like to swoop too close to the ground. If it is too
high in the air, or too exposed, the wind won' t let them
settle on it. If it is near a laundry line, bl owi ng in the
wi nd, they will be frightened by the movi ng line. Most
of the places where I put the table actually don' t work.
I slowly learn that blackbirds have a million subtle
forces gui di ng them in their behavior. If I don' t under-
stand these forces, there is simply nothing I can do to
make the table come to life. So l ong as the placing of the
table is inexact, my image of the blackbirds flocked around
the table eating, is just wishful thinking. T o make the ta-
ble live, I must take these forces seriously, and place the
table in a position which is perfectly exact.
And} yet, of course, the word "exact" does not de-
scribe it property.
It has no sense of freedom in it; and it is too reminiscent
of those other things whi ch are exact in an entirely dif-
ferent sense.
Usual l y, when we say something is exact, we mean
35
T HE QUALI TY
that it fits some abstract image perfectl y. If I cut a square
of cardboard, and make it perfectly exact, it means that I
have made the cardboard perfectly square: its sides are
exactly equal: and its angles are exactly ninety degrees. I
have matched an image perfectly.
T h e meani ng of the word " exac t " which I use here is
almost opposite. A thing which has the quality without a
name never fits any image exactly. Wh a t is exact is its
adaptation to the forces which are in it. But this exact-
ness requires that it be loose and fluid in its f orm.
A word which goes much deeper than the word
"exact" is "egoless."
Wh e n a place is lifeless or unreal, there is almost al ways a
mastermind behind it. It is so filled with the will of its
maker that there is no room for its own nature.
Thi nk, by contrast, of the decoration on an old bench
smal l hearts carved in it; simple holes, cut out while it
was being put t oget hert hese can be egoless.
Th e y are not carved accordi ng to some plan. The y
are carefree, carved into it, wherever there seems to be a
gap. It is not in the least contri ved; there is no ef f ort in
the decoration; it does not seek to express the personality
of the man who carved it. It is so natural, that it almost
seems as though the bench itself cried out for it: and the
carver simply did what was required.
And yet, although the old bench and its carving may
be egoless, this word is also not quite right.
36
T HE QUALI TY WI T HOUT A NAME
It does not mean, for instance, that the man who made it
left his own person out of it. It was part of his person that
he liked the bench, and wanted to carve hearts in it. Per-
haps he made it for his favorite girl.
It is perfectly possible to make a thing which has the
quality which has no name, and still let it reflect your
personality. Yo ur person, and the likes and dislikes whi ch
are part of you, are themselves forces in your garden, and
your garden must reflect those forces just as it reflects the
other forces which make leaves gr ow and birds sing.
But if you use the word " e g o" to mean the center of a
person's character, then the idea of maki ng something
egoless can sound as though you want the person to efface
himself completely. Th a t is not what the word means at
al l ; and yet because of it, the word is not quite right.
A last word which can hel-p to catch the quality with-
out a name is the word "eternal."
Al l things and people and places whi ch have the quality
without a name, reach into the realm of the eternal.
Some are eternal in almost a literal sense: they are so
strong, so balanced, so strongly self-maintaining, that
they are not easily disturbed, almost imperishable. Others
reach the quality for no more than an instant, and then
fall back into the lesser state, where inner contradictions
rule.
Th e word "et ernal " describes them both. For the in-
stant that they have this quality, they reach into the realm
of eternal truth. At that moment when they are free
37
T HE QUALI TY
from inner contradictions, they take their place among
the order of things whi ch stand outside of time.
I once saw a simple -fish pond in a Japanese village
which was perhaps eternal.
A farmer made it for his farm. T h e pond was a simple
rectangle, about 6 feet wide, and 8 feet l ong; opening
off a little irrigation stream. At one end, a bush of flow-
ers hung over the water. At the other end, under the
water, was a circle of wood, its top perhaps 12 inches be-
l ow the surface of the water. In the pond there were
eight great ancient carp, each maybe 18 inches l ong,
orange, gol d, purple, and black: the oldest one had been
there eighty years. Th e eight fish swam, sl owl y, sl owl y,
in ci rcl esof t en within the wooden circle. Th e whol e
worl d was in that pond. Ever y day the f armer sat by it
for a f ew minutes. I was there only one day and I sat by
it all afternoon. Even now, I cannot think of it without
tears. Those ancient fish had been swi mmi ng, sl owl y, in
that pond for eighty years. It was so true to the nature
of the fish, and fl owers, and the water, and the farmers,
that it had sustained itself for all that time, endlessly re-
peating, al ways different. The r e is no degree of whol e-
ness or reality which can be reached beyond that simple
pond.
And yet, like all the other words, this word confuses
more than it explains.
38
T HE QUALI TY WI T HOUT A NAME
It hints at a religious quality. T h e hint is accurate. And
yet it makes it seem as though the quality whi ch that pond
has is a mysterious one. It is not mysterious. It is above
all ordinary. Wh a t makes it eternal is its ordinariness.
Th e word "et ernal " cannot capture that.
And so you see, in spite of every effort to give this
quality a name, there is no single name which cap-
tures it.
Imagi ne the quality wi thout a name as a point, and each
of the words which we have tried as an ellipse. Each el-
lipse includes this point. But each ellipse also covers many
other meanings, whi ch are distant from this point.
Since every word is al ways an ellipse like thi sthen
every word will al ways be too broad, too vague, too large
in scope to refer only and exactly to the quality whi ch is
the point. No word can ever catch the quality without a
name because the quality is too particular, and words too
broad. And yet it is the most important quality there is, in
anyone, or anythi ng.
It is not only simple beauty of form and color. Man
can make that without making nature. It is not only
fitness to purpose. Man can make that too, without
making nature. And it is not only the spiritual
quality of beautiful music or of a quiet mosque, that
comes from faith. Man can make that too, without
making nature.
39
T H E Q U A L I T Y
The quality which has no name includes these
simpler sweeter qualities. But it is so ordinary as
well, that it somehow reminds us of the passing of
our life.
It is a slightly bitter quality.
40
CHAPTER 3
BEING ALIVE
The search which we make for this
quality, in our own lives, is the central
search of any -person, and the crux of any
individual person's story. It is the search
for those moments and situations when
we are most alive.
' v . r
We know, nowy what the quality without a name is
like, in feeling and in character. But so far, con-
cretely, we have not. seen this quality in any system
larger than a tree, a fond, a bench. Yet it can be in
anythingin buildings, animals, plants, cities, streets,
the wildernessand in ourselves. We shall begin to
understand it concretely, in all these larger pieces
of the world, only when we first understand it in
ourselves.
It is, for instance, the wild smile of the gypsies
dancing in the road.
T h e broad brim of the big hat, like arms spread wide,
open to the worl d, confi dent, huge, . . . T h e embrace
of the child's arms about the grass. . . . It is the solid
and entrenched repose of the old man lighting a ciga-
rette: hands on his knees, solid, resting, waiting, listen-
ing.
In our lives, this quality without a name is the most
precious thing we ever have.
And I am free to the extent I have this quality in me.
One man is free at that one instant when you see in him
a certain smile and you know he is himself, and perfectly
at home within himself. Imagi ne him especially, perhaps,
weari ng a great wide hat, his arm flung out in an expan-
sive gesture, singing perhaps and for one instant utterly
oblivious to everythi ng but what is in him and around him
at that second.
47
T HE QUALI TY
This wild, freedom, this -passion, comes into our lives
in the instant we let go.
It is when all our forces can move freely in us. In nature,
this quality is almost automatic, because there are no
images to interfere with natural processes of maki ng
things. But in all of our creations, the possibility occurs
that images can interfere with the natural, necessary
order of a thing. And, most of all, this way that images
distort the things we make, is familiar in ourselves. For
we ourselves are, like our works, the products of our
own creation. And we are only free, and have the quality
without a name in us, when we give up the images whi ch
guide our lives.
Ye t each of us faces the fear of letting go. T h e fear
of being just exactly what one is, of letting the forces flow
f reel y; of letting the configuration of one's person adjust
truly to these forces.
Our letting go is stifled, all the time, so l ong as we
have ideas and opinions about ourselves, whi ch make us
hug too tightly to our images of how to live, and bottle
up these forces.
So l ong as we are still bottled up, like this, there is a
tightness about the mouth, a nervous tension in the eyes,
a stiffness and a brittleness in the way we wal k, the
way we move.
And yet, until one does let go, it is impossible to be
alive. T h e stereotypes are restricted; there are very dif-
ferent configurations. T h e infinite variety of actual
48
BEI NG ALI VE
people, with their vastly and utterly different forces, re-
quire a huge creation, to find the resolution of the per-
son: and in finding this resolution truly, one must above
all be free of the stereotypes.
The great film, I k i r u t o livedescribes it in the
life of an old man.
He has sat for thirty years behind a counter, preventing
things from happening. And then he finds out that he is
to die of cancer of the stomach, in six months. He tries
to l ive; he seeks enj oyment ; it doesn' t amount to much.
And finally, against all obstacles, he helps to make a park
in a dirty slum of Toky o. He has lost his fear, because he
knows that he is goi ng to die; he works and works and
works, there is no stopping him, because he is no l onger
afraid of anyone, or anythi ng. He has no l onger anything
to lose, and so in this short time gains everythi ng; and
then dies, in the snow, swi ngi ng on a child' s swi ng in the
park whi ch he has made, and singing.
Each of us lives most fully "on the wire " in the face
of death, daring to do the very thing which fear fre-
vents us from.
A few years ago a family of high wire artists had a
terrible fall from the high wi re, in the middle of their
performance. Al l of them were killed or maimed, ex-
cept the father, who escaped with broken legs. But even
49
T HE QUALI TY
after losing his children in the fall, a f ew months later
he was back to work, in the circus, on the wire again.
Someone asked him in an interview, how he could bring
himself to do it, after such a terrible accident. He an-
swered: " O n the wi re, that's living . . . all the rest
is wai t i ng. "
Of course for most of us it is not quite so literal.
T h e fear whi ch prevents us from being ourselves, from
being that one person unique in all the worl d, from com-
ing to l i f et hat may mean nothi ng greater than the
fear of gi vi ng up the image of a certain job, an image
of a certain kind of family life.
One man can be as free in lighting up a cigarette, as
that old man dancing on the wi re. Another traveling wi th
the gypsies. A handkerchief around your head; a horse-
drawn yel l ow caravan, pulled up in a f i el d; a rabbit stew,
simmering and bubbling on the fire otuside the caravan;
licking and sucking your fi ngers as you eat spoonfuls of
the stew.
It has above all to do with the elements.
Th e wi nd, the soft rain; sitting on the back of an old
truck movi ng clothes and baskets of possessions while the
gentle rain is falling, l aughi ng, crouching under a shawl
to keep from getti ng wet , but getti ng wet. Eat i ng a loaf
of bread, torn in pieces, hunks of cheese cut crudely with
a hatchet which is lying in the corner; red flowers
50
BEI NG ALI VE
glistening in the rain along the roadside; bangi ng on the
wi ndow of the truck to shout some j oke.
Not hi ng to keep, nothi ng to lose. No possessions, no
security, no concern about possessions, and no concern
about security: in this mood it is possible to do exactly
what makes sense, and nothing else: there are no hidden
fears, no morals, no rules, no undercurrent of constraint,
no subtle sense of concern for the form of what the
people round about you are doing, and above all no con-
cern for what you are yourself, no subtle fear of other
people's ridicule, no subtle train of fears which can con-
nect the smallest triviality with bankruptcy and loss of
love and loss of friends and death, no ties, no suits, no
outward elements of majesty at all. Onl y the laughter
and the rain.
And it happens when our inner forces are resolved.
And when a person's forces are resolved, it makes us
feel at home, because we know, by some sixth sense, that
there are no other unexpected forces lurking under-
ground. He acts according to the nature of the situations
he is in, wi thout distorting them. The r e are no guiding
images in his behavior, no hidden forces; he is simply
free. And so, we feel relaxed and peaceful in his com-
pany.
Of course, in practice we often don't know just what
our inner forces are.
51
T HE QUALI TY
We live, for months, for years, acting in a certain way,
not knowi ng whether we are free or not, doubting, not
even sure when we are successfully resolved, and when
we aren' t.
Yet still there are those special secret moments in
our lives, when we smile unexpectedlywhen all our
forces are resolved.
A woman can often see these moments in us, better than
a man, better than we ourselves even. Wh e n we know
those moments, when we smile, when we let go, when
we are not on guard at al l t hese are the moments when
our most important forces show themselves; what ever
you are doing at such a moment, hold on to it, repeat i t
for that certain smile is the best knowl edge that we ever
have of what our hidden forces are, and where they lie,
and how they can be loosed.
We cannot be aware of these most precious moments
when they are actually happening.
In fact, the conscious ef f ort to attain this quality, or to be
free, or to be anythi ng, the glance whi ch this creates, will
al ways spoil it.
It is, instead, when we forget ourselves compl etel y:
playing the fool perhaps among a group of friends, or
swi mmi ng out to sea, or wal ki ng simply, or trying to
finish something late at night over a table with a group
of friends, cigarette stuck to l ower lip, eyes tired, earnest
concentration.
52
BEI NG ALI VE
Al l these moments in my own l i f e I only know them
now, in retrospect.
Yet each of us knows from experience the feeling
which this quality creates in us.
It is the time when we are most right, most just, most
sad, and most hilarious.
And for this reason, each one of us can also recognize
this quality when it occurs in buildings.
We can identify the towns and buildings, streets and
gardens, flower beds, chairs, tables, tablecloths, wine bot-
tles, garden seats, and kitchen sinks whi ch have this
qual i tysi mpl y by asking whether they are like us when
we are free.
We need only ask ourselves whi ch pl aceswhi ch towns,
which buildings, which rooms, have made us feel like
t hi swhi ch of them have that breath of sudden passion
in them, whi ch whispers to us, and lets us recall those
moments when we were ourselves.
And the connection between the t wobet ween this
quality in our own lives, and the same quality in our
surroundi ngsi s not just an anal ogy, or similarity. Th e
fact is that each one creates the other.
Places which have this quality, invite this quality to
come to life in us. And when we have this quality in
us, we tend to make it come to life in towns and
buildings which we help to build. It is a self-
53
T H E Q U A L I T Y
supporting, self-maintaining, generating quality. It is
the quality of life. And we must seek it, for our own
sakes, in our surroundings, simply in order that we
can ourselves become alive.
That is the central scientific fact in all that follows.
5 4
CHAPTER 4
PATTERNS OF EVENTS
In order to define this quality in build-
ings and in towns
}
we must begin by
understanding that every place is given
its character by certain patterns of events
that keep on happening there.
MV**
1
We know what the quality without a name is like in
our own lives.
As we shall see in the next few chapters, this qual-
ity can only come to life in us when it exists within the
world that we are part of. We can come alive only to
the extent the buildings and towns we live in are
alive. The quality without a name is circular: it exists
in us, when it exists in our buildings-, and it only
exists in our buildings, when we have it in ourselves.
To understand this clearly, we must first recognize
that what a town or building is, is governed, above all,
by what is happening there.
I mean this in the most general sense.
Activities; events; forces; situations; l i ghtni ng strikes;
fish die; wat er flows; lovers quarrel ; a cake burns; cats
chase each other; a hummi ngbi rd sits outside my wi n-
dow; friends come by; my car breaks down; lovers' re-
uni on; children born; grandparents go broke. . . .
My life is made of episodes like this.
Th e life of every person, animal, plant, creature, is
made of similar episodes.
T h e character of a place, then, is given to it by the
episodes whi ch happen there.
Those of us who are concerned with buildings tend
to forget too easily that all the life and soul of a
place, all of our experiences there, depend not
simply on the physical environment, but on the pat-
terns of events which we experience there.
62
P A T T E RNS OF EVENT S
Wh a t is L i ma w h a t is most memor abl e t he r e e a t i ng
ant i cuchos in the st reet ; smal l pieces of beef heart , on
sticks, cooked over open coals, wi t h hot sauce on t he m;
the dark, badl y lit ni ght streets of Li ma , smal l carts wi t h
the flickering fire of the hot coals, the faces of the
sellers, s hadowy f i gures gat hered r ound, to eat the beef
hearts.
O r in Ge ne v a c he s t nut s , hot, in smal l paper bags, eaten
in the aut umn mist, wa r mi ng the f i ngers.
An d , wha t is it about the Cal i f or ni a c oa s t t he shock of
the waves, the hiss of the surf , st andi ng on a rock whi l e
the whi t e wat er hisses in, runs out , a dash across the we t
sand to the rock, bef ore the sea comes in agai n.
An d it is j ust the same i ndoors. T h i n k of a big r oom,
vast, huge wi ndows , bi g empt y f i repl ace, compl et el y
empt y, no f urni t ure at al l , except an easel and a c h a i r
Picasso' s studi o. Is this not made enti rel y of the situations,
the f orces let loose by the conf i gur at i on of events?
An d wha t of a part y around a ki tchen tabl e, people
dri nki ng t oget her, cooki ng t oget her, dri nki ng wi ne, eat-
i ng grapes, t oget her prepari ng a st ew of beef and wi ne
and garl i c and t omat oes whi ch takes f our hours to c o o k
and whi l e it cooks, we dri nk, and t hen, at last we eat it.
Wh a t of the kind of mome nt we r emember most of al l :
the flickering candl es on the Chr i s t mas tree, the smal l bell
ri ngi ng, the chi l dren wai t i ng, hour by hour, outside the
door, peepi ng t hrough the crack, and f i nal l y rushi ng in,
whe n they hear the t i nkl i ng of that smal l bell, and see the
tree there, lit, wi t h f i f t y whi t e and red candl es, burni ng,
and the smel l of si nged pine needl es wher e a t wi g caught
f i re whi l e the candl es wer e bei ng lit.
63
T HE QUALI TY
Wha t of the process of scrubbing the floor, worki ng the
stiff bristles of the brush, and the pail of wat er, over the
soft boards, with fibers breaking loose, and the smell of
the soap that stays in the wood.
Or saying goodbye, at a train, leaning through the
wi ndow of the train, wavi ng, kissing, as the train pulls
out, runni ng al ong the pl atform. . . .
Or , taking the Sunday wal k, a famil y, abreast, in twos
and threes, wal ki ng al ong the road, pushing the smallest
child perhaps, the others l aggi ng behind to look at frogs,
and an old shoe.
These patterns of events which create the character
of a place are not necessarily human events.
T h e sunshine shining on the wi ndowsi l l , the wi nd bl ow-
ing in the grass are events t oot hey affect us just as
much as social events.
Any combination of events, whi ch has a bearing on
our l i vesan actual physical effect on usaf f ect s our
lives.
I f , for example, there is a stream bed gouged in the rock
outside my house, which fills each time it rains, this is a
situation whi ch has a powerf ul effect on the character of
the envi ronment, yet is not a human situation at all.
Compare the power and importance of these events
with the other purely geometrical aspects of the en-
vironment, which- architects concern themselves with.
64
PATTERNS OF EVENTS
Compare, for instance, t wo ways of including water in a
building.
Suppose, on the one hand, that there is a concrete reflect-
ing pool outside your r oomwi t h no purpose except to
reflect the sky.
And suppose, on the other hand, that there is a stream
outside your room, with a small rowi ng boat on it, where
you can go, to row, lie on the water, struggle against the
stream, tip over . . . .
Which of these two makes the most difference to the
building? The rowing boat, of course, because it
alters the entire experience of the building.
It is the action of these moments, the people involved in
them, and the peculiar situations, whi ch make the im-
pression on our lives.
Th e life of a house, or of a t own, is not gi ven to it,
directly, by the shape of its buildings, or by the ornament
and pl ani t is given to them by the quality of the events
and situations we encounter there. Al ways it is our situa-
tions whi ch allow us to be what we are.
It is the people around us, and the most common ways we
have of meeting them, of being with them, it is, in short,
the ways of being whi ch exist in our worl d, that make it
possible for us to be alive.
We know, then, that what matters in a building or a
town is not its outward shape, its physical geometry
alone, but the events that happen there.
65
T HE QUALI TY
It is all the events whi ch happen t her et he human
events given by the situations whi ch are repeated, the
mechanical events, the rush of trains, the fall of water,
the slow cracki ng of structures, the growi ng of the grass,
the mel ti ng of the snow, the rusting of iron, the flower-
ing of roses, the heat of a summer' s day, the cooking,
l oving, playing, dying, and not only of ourselves, but of
the animals, and plants, and even of the inorganic pro-
cesses whi ch make the whol e.
Of course, some events happen once in a l i feti me;
others happen more of t en; and some happen very often
indeed. But although it is true that a unique event can
sometimes change our lives completely, or leave its mark
on us, it is not too much to say that, by and large, the
overall character of our lives is given by those events
which keep on recurri ng over and over again.
And, by the same token, it is roughly true that any
system, any aspect of the life of a part of the worl d, is
essentially governed by those situations, human or non-
humanwhi c h keep on repeating there.
A building or a town is given its character, essen-
tially, by those events which keep on happening
there most often.
A field of grass is given its character, essentially, by
those events which happen over and over agai nmi l l i ons
upon millions of times. T h e germination of the grass
seed, the bl owi ng wi nd, the flowering of the grass, the
66
PATTERNS OF EVENTS
movement of the worms, the hatching of the in-
sects . . . .
A car is given its character by the events whi ch keep on
happening t heret he rolling of the wheels, the move-
ment of the pistons in the cylinders, the limited to and fro
of the steering wheel and axle, as the car changes
direction.
A family is given its character by the particular events
which keep on happening t heret he small affections,
kisses, breakfast, the particular kinds of arguments which
keep on happening, the way these arguments resolve
themselves, the idiosyncrasies of people, both together and
alone, whi ch make us love them . . . .
And just the same is true in any -person's individual
life.
If I consider my life honestly, I see that it is governed by
a certain very small number of patterns of events whi ch I
take part in over and over again.
Bei ng in bed, havi ng a shower, having breakfast in the
kitchen, sitting in my study wri ti ng, wal ki ng in the
garden, cooking and eating our common lunch at my
office with my friends, goi ng to the movies, taking my
family to eat at a restaurant, havi ng a drink at a friend' s
house, driving on the f reeway, goi ng to bed again. Ther e
are a f ew more.
The r e are surprisingly f ew of these patterns of events in
any one person's way of life, perhaps no more than a
67
T HE QUALI TY
dozen. Look at your own life and you will find the same.
It is shocking at first, to see that there are so f ew patterns
of events open to me.
Not that I want more of them. But when I see how very
f ew of them there are, I begin to understand what huge
effect these f ew patterns have on my life, on my capacity
to live. If these f ew patterns are good for me, I can live
wel l . If they are bad for me, I can' t.
Of course, the standard patterns of events vary very
much from person to person, and from culture to
culture.
For a teenage boy, at high school in Los Angel es, his
situations include hangi ng out in the corridor wi th other
boys; wat chi ng television; sitting in a car with his girl -
friend at a drive-in restaurant eating coke and ham-
burgers. For an old woman, in a European mountain
village, her situations include scrubbing her front door-
step, l ighting a candle in the local church, stopping at the
market to buy fresh vegetables, wal ki ng five miles across
the mountains to visit her grandson.
But each town, each neighborhood, each building, has
a particular set of these patterns of events according
to its prevailing culture.
A person can modi f y his immediate situations. He can
move, change his life, and so on. I n exceptional cases he
can even change them almost whol l y. But it is not possible
68
PATTERNS OF EVENTS
to go beyond the bounds of the collection of events and
pattern of events whi ch our culture makes available to us.
We have a glimpse, theny of the fact that our world
has a structurey in the simple fact that certain patterns
of eventsboth human and nonhumankeep repeat-
ing, and account, essentiallyy for much the greater
part of the events which happen there.
Our individual lives are made from them . . . so are
our lives together . . . they are the rules, through
whi ch our culture maintains itself, keeps itself alive, and
it is by building our lives, out of these patterns of events,
that we are people of our culture . . . .
The r e is no aspect of our lives whi ch is not governed
by these patterns of events. And if the quality without a
name can come into our lives at all, it is clear that it
depends entirely on the specific nature of these patterns
of events from whi ch our worl d is made.
And indeedy the world does have a structurey just be-
cause these patterns of events which repeat them-
selves are always anchored in the space.
I cannot imagine any pattern of events without imagining
a place where it is happening. I cannot think of sleeping,
without i magi ni ng myself sleeping somewhere. Of
course, I can imagine myself sleeping in many different
kinds of pl acesbut these places all have at least certain
physical geometrical characteristics in common. And I can-
69
T HE QUALI TY
not think about the place without also knowi ng, or imagin-
ing, what happens there. I cannot think of a bedroom,
without i magi ni ng the bed, l ovemaki ng, sleeping, dressing
perhaps, waki ng up . . . breakfast in bed . . . .
Consider, for example, the pattern of events which
we might call "watching the world go by."
We sit, perhaps slightly raised, on the front porch, or on
some steps in a park, or on a cafe terrace, with a more or
less protected, sheltered, partly private place behind us,
looking out into a more public place, slightly raised above
it, watchi ng the worl d go by.
I cannot separate it from the porch where it occurs.
Th e action and the space are indivisible. T h e action is
supported by this kind of space. T h e space supports this
kind of action. Th e t wo f orm a unit, a pattern of events
in space.
Th e same in a barbershop. Inside, barbers, customers
sitting in a row al ong one side, chairs for haircuts in
another row, widely spaced, facing the mirrors, the barber
idly talking while he cuts your hair, bottles of pomade
around, hair dryer l yi ng on the table, a basin in front for
rinsing, a strop hangi ng on the wal l , for stropping the
razors . . . . Agai n, the activity and its physical space
are one. The r e is no separating them.
70
PATTERNS OF EVENTS
Indeed, a culture always defines its -pattern of events
by referring to the names of the physical elements of
space which are "standard" in that culture.
If you look back at the patterns of events that I have
mentioned, each one is almost completely defined by the
spatial character of the place where it occurs.
Th e barbershop, the porch, the shower, the study with
its wri ti ng desk, the garden with its path, the bed, the
communal lunch table, the cinema, the f reeway, the high
school corridor, the television set, the drive-in restaurant,
the front doorstep, the candlestand at the back of the
church, the market with its vegetable stalls, the mountain
path. Each of these elements defines a pattern of events.
And the mere list of elements which are typical in a
given town tells us the way of life of people there.
Wh e n you think of Los Angel es, you think of freeways,
drive-ins, suburbs, airports, gas stations, shopping centers,
swi mmi ng pools, hamburger joints, parking lots, beaches,
billboards, supermarkets, free-standing one-family houses,
front yards, traffic lights . . . .
Wh e n you think of a medieval European t own, you
think of the church, the marketplace, the town square, the
wall around the t own, the town gates, narrow wi ndi ng
streets and lanes, rows of attached houses, each one con-
taining an extended family, rooftops, alleys, blacksmiths,
alehouses . . . .
In each case the simple list of elements is intensely evoc-
71
T HE QUALI TY
ative. T h e elements are not just dead pieces of architecture
and bui l di ngeach one has an entire life associated with
it. T h e names of the elements make us imagine and re-
member what people are doing in those elements, and
what life is like in an envi ronment which has those ele-
ments.
This does not mean that space creates events, or that
it causes them.
For example, in a modern t own, the concrete spatial
pattern of a sidewalk does not "cause" the kinds of hu-
man behavior which happens there.
Wh a t happens is much more complex. T h e people on
the sidewalk, being cul ture-bound, know that the space
which they are part of is a sidewalk, and, as part of their
culture, they have the pattern of a sidewalk in their
minds. It is this pattern in their minds which causes them
to behave the way that people do behave on sidewalks,
not the purely spatial aspect of the concrete and the walls
and curbs.
And this means, of course, that in t wo cultures, people
may see sidewalks di fferentl y, that is, they may have
di fferent patterns in their mi ndsand, that they will, as
a result, act di fferentl y on the sidewalks. For example, in
Ne w Yor k, a sidewalk is mainly a place for wal ki ng,
jostling, movi ng fast. And by comparison, in Jamai ca, or
India, a sidewalk is a place to sit, to talk, perhaps to play
music, even to sleep.
It is not correct to interpret this by saying that the t wo
sidewalks are the same.
72
PATTERNS OF EVENTS
It simply means that a pattern of events cannot he
separated from the space where it occurs.
Each sidewalk is a unitary system, which includes both
the field of geometrical relationships which define its con-
crete geometry, and the field of human actions and events,
which are associated with it.
So when we see that a sidewalk in Bombay is used by
people sleeping, or for parking cars . . . and that in
Ne w Yor k it is used only for wa l ki ng we cannot inter-
pret this correctly as a single sidewalk pattern, with t wo
di fferent uses. Th e Bombay sidewalk (space -f- events) is
one pattern; the Ne w Yor k sidewalk (space -J- events) is
another pattern. Th e y are t wo entirely di fferent patterns.
This close connection between patterns of events and
space is commonplace in nature.
T h e word " st r eam" describes a pattern of physical space
and a pattern of events, at the same time.
We do not separate the stream bed from the stream.
Ther e is no distinction in our minds between the bed of
the stream, its banks, its wi ndi ng configuration in the
land, and the rushing of the water, the growt h of plants,
the swi mmi ng of the fish.
And, in the same wayy the patterns of events which
govern life in buildings and in towns cannot be
separated from the space where they occur.
Each one is a living thing, a pattern of events in space,
73
T HE QUALI TY
just like a stream, a waterf al l , a fire, a s t or ma thing
which happens, over and again, and is exactly one of the
elements f rom which the worl d is made.
And it is therefore clear that we can only understand
these patterns of events by seeing them as l iving elements
of space themselves.
It is t;he space itself whi ch lives and breathes; it is the
space which is the wal ki ng, jostling sidewalk in Ne w
Yo r k ; it is the space we call the porch, whi ch is the pat-
tern of events we also call watchi ng the worl d go by.
The lije which happens in a building or a town is not
merely anchored in the space but made up from the
space itself.
For since space is made up of these living elements, these
labeled patterns of events in space, we see that what
seems at first sight like the dead geometry we call a
building or a town is indeed a quick thing, a living sys-
tem, a collection of interacting, and adj acent, patterns of
events in space, each one repeating certain events over
and over again, yet al ways anchored by its place in space.
And, if we hope to understand the life which happens in
a building or a t own, we must therefore try to understand
the structure of the space itself.
We shall now try to find some way of understand-
ing space which yields its patterns of events in a com-
pletely natural way, so that we can succeed in seeing
patterns of events, and space, as one.
74
CHAPTER 5
PATTERNS OF SPACE
These patterns of events are always inter-
locked with certain geometric patterns in
the space. Indeed, as we shall see, each
building and each town is ultimately
made out of these patterns in the space,
and out of nothing else: they are the
atoms and the molecules from which a
building or a town is made.
%i p y J l i j
We are now ready to come to grips with the most
basic -problem of a building or a town: What is it
made of? What is its structure? What is its physical
essence? What are the building blocks of which its
space is made?
We know, from chapter 4, that any town and any
building gets its character from those events and pat-
terns of events which keep on happening there the
most-, and that the patterns of events are linked,
somehow, to space.
So far, though, we do not know fust what aspect of
the space it is that correlates with the events. We do
not have a picture of a building or a town which
shows us how its obvious outward structurethe way
it looks, its physical geometryis interlocked with
these events.
Suppose I want to understand the "st ruct ure" of some-
thing. Just what exactly does this mean?
It means, of course, that I want to make a simple pic-
ture of it, whi ch lets me grasp it as a whol e.
And it means, too, that as far as possible, I want to paint
this simple picture out of as few elements as possible. T h e
f ewer elements there are, the richer the relationships be-
tween them, and the more of the picture lies in the
"st ruct ure" of these relationships.
And final l y, of course, I want to paint a picture whi ch
allows me to understand the patterns of events whi ch
keep on happening in the thing whose structure I seek. In
other words, I hope to find a picture, or a structure,
81
T HE QUALI TY
which wi l l , in some rather obvious and simple sense, ac-
count for the outward properties, for the pattern of
events of the thing whi ch I am studying.
Wh a t then, is the f undamental "st ruct ure" of a build-
ing or a t own ?
In the crudest sensey we know from the last chapter
roughly what the structure of a town or building is.
I t is made up of certain concrete elements, with every
element associated with a certain pattern of events.
On the geometric level> we see certain physical ele-
ments repeating endlesslyy combined in an almost end-
less variety of combinations.
A t own is made of houses, gardens, streets, sidewalks,
shopping centers, shops, workplaces, factories, perhaps a
river, sportgrounds, parking . . .
A^building is made up of walls, wi ndows, doors, rooms,
ceilings, nooks, stairs, staircase treads, doorhandles, ter-
races, counter tops, flowerpots . . . repeated over and
again.
A gothic cathedral is made of a nave, aisles, west door,
transept, choir, apse, ambul atory, columns, wi ndows, but-
tresses, vaults, ribs, wi ndow tracery.
A modern metropolitan region in the Uni t ed States is
made of industrial areas, freeways, central business dis-
tricts, supermarkets, parks, single-family houses, gardens,
high-rise housing, streets, arteries, traffic lights, sidewalks.
82
PATTERNS OF SPACE
And each of these elements has a specific pattern of
events associated with it.
Families living in the houses, cars and buses driving in the
streets, flowers gr owi ng in the f l ower pots, people wal ki ng
through the doors, opening and closing them, traffic lights
changi ng, people gatheri ng for mass on Sundays in the
nave of the cathedral, forces acting on the vaults, when
the wi nd sways the building, light comi ng through the
wi ndows, people sitting at the wi ndows in their living
rooms and looking at the vi ew . . . .
But this picture of space does not explain howor
whythese elements associate themselves with defi-
nite and quite specific patterns of events.
Wh a t is the relation between a church, say, taken as an
el ement and the pattern of events whi ch happens in the
church? It is all very wel l to say that they are connected.
But unless we can see some kind of common sense in the
connection, it explains nothing.
It is certainly not enough merely to say glibly that
every pattern of events resides in space. Tha t is obvious,
and not very interesting. Wh a t we want to know is just
how the structure of the space supports the patterns of
events it does, in such a way that if we change the struc-
ture of the space, we shall be able to predict what kinds of
changes in the patterns of events this change will
generate.
I n short, we want a theory whi ch presents the interaction
83
T HE QUALI TY
of the space and the events, in a clear and unambiguous
way.
Furthery it is very puzzling to realize that the
a
ele-
ments" which seem like elementary building blocksy
keep varyingy and are different every time that they
occur.
For among the endless repetition of elements we also see
an almost endless variation. Each church has a slightly dif-
ferent nave, the aisles are different, the west door is di ffer-
ent . . . and in the nave, the various bays are usually dif-
ferent, the individual columns are di f f erent; each vault
has slightly di fferent ribs; each wi ndow has a slightly dif-
ferent tracery and di fferent glass.
And just so in an urban region. Each industrial area is
di f f erent; each f reeway is di f f erent; each park is di ffer-
ent; each supermarket is di f f er ent even the smaller
individual elements like traffic lights and stop signs, al-
though very similar, are never quite the s ameand there
is al ways a variety of types.
If the elements are different every time that they
occury evidently theny it cannot be the elements them-
selves which are repeating in a building or a town: these
so-called elements cannot be the ultimate "atomic" con-
stituents of space.
Since every church is di fferent, the so-called element we
call " chur ch" is not constant at all. Gi vi ng it a name
84
PATTERNS OF SPACE
only deepens the puzzl e. If every church is di fferent,
what is it that remains the same, from church to church,
that we call " chur ch" ?
Whe n we say that matter is made of electrons, pro-
tons, and so forth, tin's is a satisfying way of understanding
things, because these electrons seem, indeed, to be the
same each time that they occur, and it therefore makes
sense to show how matter can be built up from combina-
tions of these "el ement s, " because the elements are truly
elementary.
But if the so-called elements of which a building or a
town is madet he houses, streets, wi ndows, doorsare
merely names, and the underl yi ng things which they
refer to keep on changi ng, then we have no solidity at all
in our picture, and we need to find some other elements
which truly are invariant throughout the variation, in a
way that we can understand a building or a town as a
structure made up by combination of these elements.
Let us therefore look more carefully at the struc-
ture of the space from which a building or a town is
madey to find out what it really is that is repeating
there.
We may notice first that over and above the elements,
there are relationships between the elements which keep
repeating too, just as the elements themselves repeat. . . .
Beyond its elements each building is defined by cer-
tain patterns of relationships among the elements.
85
T HE QUALI TY
In a gothic cathedral, the nave is flanked by aisles whi ch
run parallel to it. T h e transept is at right angles to the
nave and aisles; the ambul atory is wrapped around the
outside of the apse; the columns are vertical, on the line
separating nave from aisle, spaced at equal intervals. Each
vault connects jour columns, and has a characteristic
shape, cross-like in plan, concave in space. Th e buttresses
are run down the outside of the aisles, on the same lines
as the columns, supporting the load from the vaults. Th e
nave is al ways a long thin rectangleits ratio may vary
between 1:3 and 1:6, but is never 1:2 or 1:20. T h e
aisles are al ways narrower than the nave.
And each urban region, too, is defined by certain fat-
terns of relationships among its elements.
Consider a typical mi d-twenti eth-century Ameri can
metropolitan region. Somewhere towards the center of
the region, there is a central business district, whi ch con-
tains a very high density office bl ock; near these there are
high density apartments. T h e overall density of the
region slopes off with distance from the center, according
to an exfonential law; -periodically there are again peaks
of higher density, but smaller than the central ones; and
subsidiary to these smaller peaks, there are still smaller
peaks. Each of these peaks of density contains stores and
offices surrounded by higher density housing. Towards
the outer fringe of the metropolis there are large areas of
freestanding one-fami l y houses; the farther out from the
center they arc, the larger their gardens. T h e region is
86
PATTERNS OF SPACE
served by a network of freeways. These f reeways are
closer together at the center. Independent of the free-
ways, there is a roughly regular two-dimensional network
of streets. Every five or ten streets, there is a l arger one,
which functions as an artery. A f ew of the arteries are
even bigger than the others: these tend to be arranged
radially, branching out from the center in a star-shaded
fashion. Whe r e an artery meets a f reeway, there is a
characteristic cloverleaf arrangement of connecting lanes.
Where t wo arteries intersect, there is a traffic l i ght;
where a local street meets an artery, there is a stop sign.
T h e maj or commercial areas, whi ch coincide with the
high density peaks in the density distribution, all fall on
the maj or arteries. Industrial areas all fall within half a
mile of a f r eeway; and the older ones are also close to
at least one maj or artery.
Evidently, then, a large fart of the "structure" of a
building or a town consists of patterns of relationships.
For both the city of Los Angel es and the medieval
church get their respective characters as much from these
repeating patterns of relationships, as they do f rom the ele-
ments themselves
At first sight, it seems as though these patterns of
relationships are separate from the elements.
Thi nk of the aisle of the cathedral. I t is parallel to the
nave, and next to it, it shares columns with the nave, it
87
T HE QUALI TY
runs east-west, like the church itself, it contains columns,
on its inner wal l , and wi ndows on its outer wal l . At first
sight, it seems that these relationships are " ext r a, " over
and above the fact of its being an aisle.
When we look closer, we realize thai these relation-
ships are not extra, but necessary to the elements, in-
deed a part of them.
We realize, for instance, that if an aisle were not parallel
to the nave, were not next to it, were not narrower than
the nave, did not share col umns with the nave, did not
run east to west, . . . that it woul d not be an "ai sl e"
at all. It woul d be merel y a rectangle of space, in gothic
construction, floating free . . . and what makes it an
aisle, specifically, is just the pattern of relationships which
it has to the nave, and other elements around it.
When we look closer still, we realize that even this
view is still not very accurate. For it is not merely
true that the relationships are attached to the ele-
ments: the fact is that the elements t hems el ves are
patterns of relationships.
For, once we recognize that much of what we think of as
an " el ement " in fact lies in the pattern of relationships be-
tween this thing and the things in the worl d around it, we
then come to the second even greater realization, that
the so-called element is itself nothing but a myth, ( and
that indeed, the element itself is not just embedded in a
88
PATTERNS OF SPACE
pattern of relationships, but is itself entirely a pattern of
relationships, and nothing else.
In short, the aisle, whi ch needs the pattern of relation-
ships to the nave and the east wi ndow to define it, is itself
also a pattern of relationships between its l ength, its
width, the columns which lie on the boundary with the
nave, the wi ndows which lie on the outer boundary . . . .
And finally, the things which seem like elements dis-
solve, and leave a fabric of relationships behind,
which is the stuff that actually repeats itself, and
gives the structure to a building or a town.
In short, we may forget about the idea that the building
is made up of elements entirely, and recognize instead,
the deeper fact that all these so-called elements are only
labels for the patterns of relationships whi ch really do
repeat.
T h e f reeway, as a whol e, does not repeat. But the fact
that there are cloverleafs which connect the f reeway to
roads at certain i nt erval st hat does repeat. The r e is a
certain relationship between the f reeway and its crossing
arteries and cloverleafs, which does repeat.
But once again, the cloverleaf itself does not repeat.
Each cloverleaf is different. Wh a t does repeat is that
each lane forms a continuously curvi ng off ramp to the
ri ght t here is a relationship between its radius, its tan-
gency, the fact that it is banked, whi ch does repeat.
Ye t once again the " l ane" which figures in this pattern
of relationships does not repeat. Wh a t we call a lane is itself
89
T HE QUALI TY
a relationship among still smaller so-called el ement st he
edges of the road, the surface, the lines whi ch form the
edge . . . and these again, al though they function tem-
porarily as elements, in order to make these relations
clear, themselves evaporate when we look closely at them.
Each one of these patterns is a morphological lawy
which establishes a set of relationships in space.
Thi s morphological l aw can al ways be expressed in the
same general f or m:
X > r ( A, B, . . . ) , whi ch means:
Wi t hi n a context of type X , the parts A, B, . . . are
related by the relationship r.
Thus , for exampl e:
Wi t hi n a gothic cathedral > the nave is flanked on
both sides by parallel aisles.
or:
Whe r e a f reeway meets an artery > the access ramps
of the interchange take the rough form of a clover-
leaf.
And each law or pattern is itself a pattern of relation-
ships among still other lawsy which are themselves
just patterns of relationships again.
For though each pattern is itself apparently composed of
smaller things whi ch look like parts, of course, when we
look closely at them, we see that these apparent "part s"
are patterns too.
90
PATTERNS OF SPACE
Consider, for exampl e, the pattern we call a door. Thi s
pattern is a relationship among the f rame, the hinges,
and the door itself: and these parts in turn are made of
smaller parts: the frame is made of uprights, a crosspiece,
and cover mouldings over joints; the door is made of up-
rights, crosspieces and panels; the hinge is made of leaves
and a pin. Ye t any one of these things we call its "part s"
are themselves in fact also patterns, each one of which
may take an almost infinite variety of shapes, and color
and exact si zewi t hout once losing the essential field of
relationships whi ch make it what it is.
Th e patterns are not just patterns of relationships, but
patterns of relationships among other smaller patterns,
which themselves have still other patterns hooking them
t oget herand we see finally, that the worl d is entirely
made of all these interhooking, interlocking nonmaterial
patterns.
Further, each pattern in the space has a pattern of
events associated with it.
For instance, the pattern of the f reeway contains a cer-
tain fabric of events, defined by rules: drivers drive at
certain speeds; there are rules governi ng the way that
people may change lanes; the cars all face the same way;
there are certain kinds of overtaki ng; people drive a little
slower on the entrances and exits . . . .
And the pattern of a kitchen, in any given culture, also
contains a very definite pattern of events: the way that
people use the kitchen, the way that food is prepared, the
91
T HE QUALI TY
fact that people eat there, or don' t eat there, the fact that
they wash the dishes standing at the sink . . . and on
and on . . .
Of course, the pattern of space, does not "cause" the
pattern of events.
Nei ther does the pattern of events "cause" the pattern in
the space. Th e total pattern, space and events together, is
an element of people's culture. It is invented by culture,
transmitted by culture, and merely anchored in space.
But there is a fundamental inner connection between
each pattern of events, and the pattern of space in
which it happens.
For the pattern in the space is, precisely, the precondition,
the requirement, whi ch al l ows the pattern of events to
happen. In this sense, it plays a f undamental role in
maki ng sure that just this pattern of events keeps on re-
peating over and over again, throughout the space, and
that it is, therefore, one of the things which gives a cer-
tain building, or a certain town, its character.
Go back, for example, to the porch of chapter 4, and
the pattern of events we may call "sitting on the
porch, watching the world go by."
Wh a t aspect of the space is it whi ch is connected to this
pattern of events? Certai nl y it is not the whol e porch,
92
PATTERNS OF SPACE
in its entirety: it is instead, just certain specific relation-
ships.
For instance, in order for the pattern of events " wat ch-
ing the worl d go by" to happen, it is essential that the
porch should be a little raised above the level of the
street; it is essential that the porch be deep enough, to let
a group of people sit there comf ort abl y; and it is essential,
of course, that the front of the porch be open, pierced
with openings, and that the roof is therefore supported on
columns.
It is this bundle of relationships which is essential} be-
cause these are the ones which are directly congruent
with the pattern of events.
By contrast, the l ength of the porch, its height, its color,
the materials of whi ch it is made, the height of the side
walls, the way the porch connects up with the inside of
the house, are less essential so they can vary, without
altering the fundamental and essential nature of the
porch.
And in this same sense, each pattern of relationships
in space is congruent with some specific pattern of
events.
Th e pattern of relationships we call a " f r e e wa y " is just
that pattern of relationships required by the process of
driving fast with limited access to and from side roads: in
short the pattern of events.
93
T HE QUALI TY
Th e pattern of relationships we call a Chinese "ki t chen"
is just that pattern of relationships required for cooking
Chinese f ood: again the underl yi ng pattern of events.
And insofar as there are di fferent "ki nds" of kitchens,
there are di fferent patterns of relationships, responsible for
slightly di fferent patterns of events, in di fferent cultures,
which have di fferent patterns of cooking.
In every case the pattern of relationships in space is that
invariant which must repeat itself with some pattern of
events, because it is exactly these relationships which are
required to sustain that pattern of events.
We realize then that it is just the patterns of events
in space which are repeating in the building or the
town: and nothing else.
Not hi ng of any importance happens in a building or a
t own except what is defined within the patterns whi ch
repeat themselves.
For what the patterns do is at the same time seize the
outward physical geometry, and also seize what happens
there.
The y account entirely for its geometrical structure: they
are the visible, coherent stuff that is repeating, and co-
herent there: they are the background of the variation,
which makes each concrete element a little di fferent.
And, at the same time, they are also responsible for
those events which keep repeating there, and therefore do
the most to give the building or a t own its character . . .
94
PATTERNS OF SPACE
Each building gets its character from just the patterns
which keep on repeating there.
Thi s is not only true of general patterns; it is true of the
entire building: all its details; the shape of rooms, the
character of ornament, the kind of wi ndowpanes it has,
the boards of whi ch the floor is made, the handles on the
doors, the light, the height, the way the ceilings vary, the
relationship of wi ndows to the ceiling, the connection of
the building to the garden and the street, and to the
spaces and the paths and to the detailed seats, and walls
which are around it. . . .
Each neighborhood is dejined) tooy in everything that
matters, by the patterns which keep on repeating
there.
Agai n, it is just those details which give the neighborhood
a "charact er" whi ch are defined by patterns: the kind of
streets whi ch it has, the kind of lots the houses are; the
typical size of houses, the way that the houses are con-
nected or distinct. . . .
Isn' t it true that the features whi ch you remember in a
place are not so much peculiarities, but rather the typical,
the recurrent, the characteristic features: the canals of
Veni ce, the flat roofs of a Moroccan town, the even spac-
i ng of the fruit trees in an orchard, the slope of a beach
towards the sea, the umbrellas of an Italian beach, the
wide sidewalks, sidewalk cafes, cylindrical poster board-
95
T HE QUALI TY
use of two-inch bricks in Georgi an houses, the ratio of
bathroom area to house area, compared with that typical
of an Ameri can house, the use of flagstones on the side-
walks. Th e n down to the tiniest details of al l t he special
shape of Engl i sh faucets, the kinds of handles on an
English metal wi ndow, the shape of the insulators on a
telegraph pole.
Agai n, in each case the patterns define all the typical
events whi ch happen there. So " London, " as a way of
life, lies there completely in these patterns whi ch the Lon-
doners create, and fill with the events that are exactly
congruent with them.
Andy what is most remarkable of ally the number of
the patterns out of which a building or a town is made
is rather small.
One mi ght imagine that a building has a thousand
different patterns in i t; or that a t own has tens of
thousands. . . .
But the fact is that a building is defined, in its essentials,
by a f ew dozen patterns. And, a vast t own like London,
or Paris, is defined, in its essence, by a f ew hundred pat-
terns at the most.
I n short, the patterns have enormous power and depth;
they have the power to create an almost endless variety,
they are so deep, so general , that they can combine in
millions upon millions of di fferent ways, to such an ex-
tent, that when we wal k through Paris we are mainly
overwhel med by the vari ety; and the fact that there are
these deep invariants, lying behind the vast variety, and
98
PATTERNS OF SPACE
generati ng it, is really an amazi ng shock. . . .
In this sense, the patterns are perhaps still deeper and
more powerf ul than the discussion has made clear so far.
Fr om a handful of patterns, a vast, almost incalculable
variety can be made: and a building, with all of its com-
plexity and variety, is generated, actually, by a small
number of them.
They are the atoms of our man-^made universe.
In chemistry we learn that the worl d, in all of its com-
plexity, is made up f rom combinations of some 92 elements,
or atoms. Thi s is an extraordinary fact, amazi ng \to a per-
son who learns chemistry for the first time. It is true that
our conception of these atoms has changed repeat edl y
far f rom being the little billiard balls we once thought,
we know that they are shifting patterns of particles and
waves and that even the most "el ement ar y" part i cl e
the el ect roni s itself a ripple in the stuff of the universe,
not a " t hi ng. " However , all these changi ng views do not
alter the fact that at the level of scale ' where atoms occur,
they do occur, as identifiable recurrent entities. And even
if vast changes occur in physics, and we one day recognize
that these so-called atoms are also merely ripples in a
deeper field, the fact that there are entities of some kind
whi ch correspond to the things we once called atoms will
remain.
Just so, we realize now, that at the larger scale of
towns and buildings, the worl d is also made of certain
fundamental " a t oms " t ha t each place is made from a
99
T HE QUALI TY
But in a courtyard where the -pattern of the opening
and veranda and crossing paths is missing, there are
forces which conflict in such a way that no one can
resolve them for himself.
Consider, for exampl e, dead courtyard surrounded by
walls on all sides, with no porch or hal f way space be-
tween the indoors and the outdoors, and with no more
than one path leading out to it.
In this place, the forces are in conflict. People want to
go out, but their timidity, which makes them seek a
place hal f way to the outdoors, prevents them. The y
want to stay out, but the claustrophobic quality, and the
enclosure, sends them back inside again. T h e y hope to be
there, but the lack of paths across the courtyard make it a
dead and rarely visited place, whi ch does not beckon
them, and which instead tends to be filled with dead
leaves, and forgotten plants. Thi s does not help them come
to l i fei nstead it only causes tension, and frustrates them,
and perpetuates their conflicts.
1 1 0
PATTERNS WHI C H ARE ALI VE
And the same can happen even In a window: A win-
dow with a "window place" helps a person come to
life.
Everyone knows how beautiful a room is when it has a
bay wi ndow in it, or a wi ndow seat, or a special ledge
next to the wi ndow, or a small alcove which is entirely
glassed. Th e feeling that rooms with these kinds of places
in them are especially beautiful is not merely whi msy. It
has a fundamental organic reason behind it.
Wh e n you are in a living room for any length of time,
t wo of the many forces acting on you are the f ol l owi ng:
1. Yo u have a tendency to go towards the light.
People are phototropic, biologically, so that it is
often comfortable to place yourself where the light
is.
2. If you are in the room for any length of time, you
probably want to sit down, and make yourself
comfortabl e.
In a room whi ch has at least one wi ndow that is a
" pl a c e " a wi ndow seat, a bay wi ndow, a wi ndow with
a wide low windowsil l that invites you to pull your
favorite chair over to it because you can see out so easily,
a special ledge next to the wi ndow, or a small alcove
which is entirely gl assedi n this room you can give in to
both forces: you can resolve the conflict for yourself.
In short, you can be comfortabl e.
But a room which has no window place, in which the
windows are just "holes'' sets up a hopeless inner
conflict in me which I can't resolve.
i n
T HE QUALI TY
And now we see just- how the circle of the argument
completes itself.
In our own lives, we have the quality wi thout a name
when we are most intense, most happy, most whol ehearted.
Thi s comes about when we allow the forces we experi-
ence to run freely in us, to fly past each other, when we
are able to allow our forces to escape the locked-in conflict
whi ch oppresses us.
But this freedom, this limpidity, occurs in us most easily
when we are in a worl d whose patterns also let their forces
loose . . . because, just as we are free when our own
forces run most freely within us, so the places we are in
are also free when their own forces (whi ch include the
forces that arise in us) themselves run free, and are them-
selves resolved . . .
The quality without a name in us, our liveliness, our
thirst for life, depends directly on the patterns in the
world, and the extent to which they have this quality
themselves.
Patterns which live, release this quality in us.
But they release this quality in us, essentially be-
cause they have it in themselves.
122
CHAPTER 7
THE MULTIPLICITY OF
LIVING PATTERNS
The more living patterns there are in a
thinga room, a building, or a
townthe more it comes to life as an
entirety, the more it glows, the more it
has this self-maintaining fire, which is
the quality without a name.
T HE QUALI TY
leaking flow of water from above, within the beami s
suddenly unstable and inadequate also, because the con-
text and the forces whi ch it tries to put in balance, have
changed.
In the endy the whole system must collapse.
Th e slight stress caused by the overf l ow of forces from
these first unstable patterns spreads first to nearby pat-
t ernsand then spreads still further, since these nearby
patterns become unstable and destructive too.
Th e delicate configuration whi ch is self-creating, and
in balance with its forces, is for some reason i nt errupt ed
prevented from occurri ng, placed in a position in which
its configuration can no l onger recreate itself.
Wha t then happens to the forces in this system?
So l ong as the self-creating balanced configuration
existed, the forces were in balance.
But once the configuration is put out of balance, these
forces remain in the system, unresolved, wi l d, out of
balance, until in the end, the whol e system must collapse.
By contrasty assume now that each one of the fifty
patterns out of which the building is made is alive
and self-resolving.
In this case just the opposite occurs. Each pattern en-
compasses, and contains, the forces whi ch it has to deal
wi t h; and there are no other forces in the system. Under
130
MUL T I P L I C I T Y OF LI VI NG PATTERNS
these circumstances, each event whi ch happens is re-
solved. T h e forces come into play, and resolve them-
selves, within the patterns as they are.
Each pattern helps to sustain other patterns.
T h e quality wi thout a name occurs, not when an isolated
pattern occurs, but when an entire system of patterns,
interdependent, at many levels, is all stable and alive.
We may see the sand ripples anywhere where we choose
to put loose sand under the wi nd.
But when the wi nd blows across the sea, over the inland
marshes, and the sand ripples support the dunes between
the two, and the sandpiper wal ks out, the sand fleas hop,
the shifting of the dunes is held in check by grasses which
maintain themselves and the sandpi perthen we have a
portion of the worl d, alive at many levels at once, begin-
ning to have the quality without a name.
The individual configuration of any one pattern re-
quires other patterns to keep itself alive.
For instance, a WINDOW PLACE is stable, and alive, only
if many other patterns which go with it, and are needed
to support it, are alive themselves: for instance, LOW
WINDOWSILL, to solve the problem of the view and the
relation to the ground; CASEMENT WINDOW to solve the
problem of the way the air comes in, to allow people to
lean out and breathe the outside air; SMALL PANES to
131
T HE QUALI TY
let the wi ndow generate a strong connection between the
inside and the outside.
If these smaller patterns, which resolve smaller systems
of forces in the wi ndow place, are missing from the wi ndow
place itself, then the pattern doesn' t work. Imagi ne for
instance a so-called wi ndow place, with high windowsills,
fixed wi ndows, and huge sheets of plate glass. The r e are
so many subsidiary forces, still in conflict, that the
wi ndow place still cannot work, because it still fails to
resolve the special system of forces it is supposed to solve.
T o be in balance, each pattern must be supported by a
situation in whi ch both the l arger patterns it belongs to,
and the smaller patterns it is made of, are themselves
alive.
In an entrance which is whole> many patterns must
cooperate.
T r y to imagine an entrance whi ch is whol e. I have in
mind an entry way, perhaps to a l arger building: and for
it to be whol e, it must contain at least these elements: the
arch, or beam, whi ch brings the loads down f rom above;
a certain heaviness perhaps, in the members whi ch bring
these forces down, and mark the edges of the entrance
way; a certain depth, or penetration, which takes the en-
trance a distance in, deep enough, so that the light is
changi ng on the way through the entry wa y ; some
ornament, around the archway, or the opening, whi ch
marks the entrance as distinct, and gives it lightness; and,
in some f orm, things that I woul d somehow see as
132
MUL T I P L I C I T Y OF LI VI NG PATTERNS
" f e e t " t hi ngs sticking out, at the bases of the si des
they might be seats, the feet of columns, something any-
way whi ch connects the sides to the ground, and makes
them one. Now this entrance might be whol e. I doubt
though, if it could be whol e with any less than that.
So, somehow, this system of patterns, which I have
loosely sketched, forms the basis for what is needed in the
entrance of a l arger building: and these patterns are a
system; they are interdependent. It is true that each one
can be explained, in its own terms, as an isolated thing,
which is needed to resolve certain forces. But , also, these
f ew patterns form a whol e, they work as a system. . . .
The same in a neighborhood.
Agai n there are certain rough patterns, more rooted in
human events, in this case, which must be there together,
in order for us to experience the neighborhood as
whol e. . . .
A boundary, certainly, more or less clearly marked; and
gateways, not emphatic, but gently present, where the
paths that pass in and out cross the boundaries: inside, a
piece of common land, with children playing, animals
grazi ng maybe; seats where old people sit and watch
what is happening; a focus to the whol e; the families
themselves, grouped in some kind of clusters, so that
there are a visible number of them, not too many, for
the neighborhood as a whol e; water somewhere; work-
shops and work, perhaps towards the boundary; houses
of course, too, but clustered; trees, s omewher eand sun-
133
T HE QUALI TY
light somewhere too, intense in at least one place. Now,
the neighborhood begins to form a whol e.
Now we begin to see what happens when the patterns
in the world collaborate.
Each living pattern resolves some system of forces, or
al l ows them to resolve themselves. Each pattern creates
an organization whi ch maintains that portion of the
worl d in balance.
And a building in whi ch all the patterns are alive has no
disturbing forces in it. T h e people are rel axed; the plants
are comfortabl e; the animals pursue their natural paths;
the forces of erosion are in balance with the natural pro-
cess of repair whi ch the confi gurati on of the building en-
courages; the forces of gravi ty are in balance wi th the
configuration of the beams and vaults, and columns, and
the bl owi ng of the wi nd; the rai nwater flows natural l y,
in such a way that it helps just those plants to gr ow
whi ch, for other reasons, are themselves in balance with
the cracks in the pavi ng stones, the beauty of the en-
trance, the smell of roses in the eveni ng outdoor
room . . . .
The more life-giving patterns there are in a building
the more beautiful it seems.
It shows, in a thousand small ways, that it is made, with
care and wi th attention to the small things we mi ght
need.
134
MU L T I P L I C I T Y OF L I V I NG P A T T E RNS
A seat, an armrest , a door handl e whi c h is comf or t abl e to
hol d, a t errace shaded f r om the heat , a flower g r o wi ng
j ust al ong the ent rance whe r e I can bend down and
smel l it as I pass i nto the gar den, l i ght f al l i ng on the top
of the stair, whe r e it is dark, so that I can wa l k t owar d
it, col or on the door, or nament ar ound the door, so that
I k no w, wi t h a smal l l eap of the heart , that I am back
agai n, a cel l ar hal f d o wn i nto the gr ound, whe r e mi l k
and wi ne can be kept cool .
Just so in a town.
T h e t o wn whi c h is al i ve, and beaut i f ul , f or me, shows, in
a t housand ways , ho w al l its institutions wor k t oget her to
make people comf or t abl e, and deep seated in respect f or
themsel ves.
Pl aces out doors whe r e people eat, and danc e; old people
si tti ng in the street, wat c hi ng the wor l d go by; pl aces
whe r e t eenage boys and gi rl s ha ng out, wi t hi n the nei gh-
borhood, f ree enough of thei r parent s t hat t hey feel
t hemsel ves al i ve, and stay t her e; car places whe r e cars
are kept, shi el ded, if there are ma ny of t hem, so that they
don' t oppress us by thei r presence; wo r k goi ng on a mo ng
the fami l i es, chi l dren pl ayi ng whe r e wo r k is goi ng on,
and l ear ni ng f r om it.
And finally the quality without a name a-pfears> not
when an isolated pattern lives
y
but when an entire
system of fatterns
y
interdependent at many levels
y
is
all stable and alive.
1 3 5
T H E Q U A L I T Y
A bui l di ng or a t own becomes al i ve whe n ever y pat t ern
in it is al i ve: whe n it al l ows each person in it, and eacli
pl ant and ani mal , and ever y st ream, and bri dge, and wal l
and roof , and every human gr oup and ever y road, to be-
come al i ve in its o wn t erms.
A n d as that happens, the whol e t own reaches the state
that i ndi vi dual people somet i mes reacJi at their best and
happiest moment s , whe n they are most f ree.
Remember the warm peach tree, flattened against
the wall, and facing south.
At this stage, the whole town will have this quality,
simmering and baking in the sun of its own processes.
1 3 6
CHAPTER 8
THE QUALITY ITSELF
And when a building has this fire
y
then it
becomes a fart of nature. Like ocean
waves
y
or blades of grass
y
its farts are
governed by the endless flay of repeti-
tion and variety
y
created in the presence
of the fact that all things pass. This is
the quality itself.
I l i p i t i i f
tMmmmmm
' -I-"..
Finally
y
in this last chapter of fart i
y
we shall see
what happens geometrically
y
when a building or a
town is made entirely of patterns which are living.
For when a town or building lives
y
we can always
recognize its lifenot only in the obvious happiness
which happens there
y
not only in its freedom and re-
laxednessbut in its purely physical appearance too.
It always has a certain geometric character.
Wh a t happens in a wo r l d a building or a t o wni n
whi ch the patterns have the quality wi t hout a name, and
are alive?
T h e most i mportant thi ng whi ch happens is that every
part of it, at every l evel , becomes unique. T h e patterns
whi ch control a portion of the wor l d, are themselves
fairly simple. But when they interact, they create slightly
di f f erent overal l conf i gurati ons at every place. Thi s
happens because no t wo places on earth are perfectl y
alike in their conditions. And each small di f f erence, itself
contributes to the di f f erence in conditions whi ch the
other patterns face.
This is the character of nature.
" T h e character of nat ur e" is no mere poetic metaphor. I t
is a specific morphol ogi cal character, a geometri c char-
acter, whi ch happens to be common to all those things in
the worl d whi ch are not man- made.
T o make this character of nature cl ear, let me contrast it
wi t h the character of the buildings bei ng built today. One
of the most pervasive features of these buildings is the fact
H3
T H E Q U A L I T Y
that t hey are " mo d ul a r . " T h e y are f ul l of i denti cal con-
crete bl ocks, i denti cal rooms, i denti cal houses, i denti cal
apart ment s in i denti cal apar t ment bui l di ngs. T h e idea
that a bui l di ng c a n a n d o u g h t t o be made of modul ar
units is one of the most pervasi ve assumpti ons of t went i et h-
cent ur y archi t ect ure.
Na t ur e is never modul ar . Nat ur e is f ul l of al most
similar uni ts ( wav e s , rai ndrops, blades of g r a s s ) b u t
t hough the units of one ki nd are all alike in their broad
st ruct ure, no t wo are ever alike in detai l .
1. T h e same broad f eat ures keep r ecur r i ng over and
over agai n.
2. I n their detai l ed appearance these broad f eat ures are
never t wi ce the same.
O n the one hand all oak trees have the same overal l
shape, the same t hi ckened twi sted t runk, the same
cri nkl ed bark, the same shaped l eaves, the same propor-
tion of l i mbs to branches to t wi gs. O n the ot her hand, no
t wo trees are quite the same. T h e exact combi nat i on of
hei ght and wi dt h and cur vat ur e never repeats i tsel f; we
cannot even find t wo l eaves whi c h are the same.
The ocean waves all have this character.
T h e pat t erns out of whi ch the wa v e is made are al ways
the s ame: the curl of the wa v e ; the drops of spr ay; the
spaci ng of the wa v e s ; the f act that r oughl y ever y sevent h
wav e is l ar ger than the others . . . T h e r e are not ma ny
of these patterns.
Y e t at the same t i me, the act ual concret e wa v e s t hem-
selves are al ways di f f er ent . T h i s happens because the
1 44
T H E Q U A L I T Y I T S E L F
patterns i nt eract di f f er ent l y at ever y spot. T h e y i nteract
di f f er ent l y wi t h one anot her. A n d they i nt eract di f f er -
ent l y wi t h the details of their surroundi ngs. So every
act ual wa v e is di f f er ent , at the same time that al l its
pat t erns are the same preci sel y as the pat t erns in the
ot her waves .
So do the drops within the waves.
T h e distinction bet ween the " g l o ba l " pat t erns and the
concret e details is not a mat t er of si ze. Wh a t is true f or
the waves is also true f or the i ndi vi dual dropl ets. Ea c h
drop of a gi ven size has mor e or less the same s ha pe y e t ,
agai n, under a f i ner mi croscope, each one is sl i ghtl y
di f f er ent f r om the next . A t each scale there are gl obal
i nvari ants, and detail ed vari ati ons. I n such a syst em, there
is endl ess var i et y; and yet at the same ti me there is
endl ess sameness. No wonde r we can wat c h the wave s
f or hours; no wo nde r that a bl ade of grass is still fas-
ci nat i ng, even af t er we have seen a mil l ion of t hem. I n
al l this sameness, we never feel oppressed by sameness. I n
all this vari et y, we never f eel lost, as we do in the presence
of vari ety we cannot underst and.
Even the atoms have this character.
I t ma y surprise you to real i ze t hat the same rul e even
hol ds f or at oms. No t wo at oms are the same. Ea c h at om
is sl i ghtl y di f f er ent , accor di ng to its i mmedi at e envi r on-
ment .
I t is part i cul arl y cruci al to discuss this f act about atoms,
145
T H E Q U A L I T Y
because so many people take " mo d u l a r " constructi on for
gr ant ed. If you chal l enge the bui l der of a modul ar en-
vi r onment , and say that such an envi r onment cannot be
al ive, he wi l l very l i kel y say that nat ure itself is built f r om
modul ar c o mpo ne nt s na me l y a t o ms a n d that wha t is
good for nat ure is good enough f or hi m. I n this sense,
at oms have become the archet ypal i mages of modul ar
const ruct i on.
But at oms are all uni que, j ust like rai ndrops and bl ades
of grass. Because we use the symbol C for every at om of
carbon, and because we know that every at om of carbon
has the same number of protons and el ectrons in it, we
assume that all at oms of carbon are i denti cal . W e thi nk of
a crystal as an array of i denti cal parts. Y e t the f act is that
the orbits of the el ectrons are i nf l uenced by the orbits of
el ectrons in nearby at oms, and are t heref ore di f f erent in
each at om, accor di ng to its position in the crystal . If we
coul d exami ne every at om in ver y gr eat detail i ndeed, we
woul d f i nd that no t wo at oms are exact l y al i ke: each is
subtly di f f er ent , accor di ng to its position in the l ar ger
whol e.
There is always repetition of the patterns.
T h e pat t erns repeat t hemsel ves because, under a gi ven
set of ci rcumst ances, there are al ways certai n fields of
relationships whi ch are most nearl y wel l adapted to the
f orces whi ch exist.
T h e shape of the wave is gener at ed by the dynami cs of
the wat er , and it repeats itself whe r e ve r these dynami cs
occur. T h e shape of the drops is gener at ed by the bal ance
1 46
T H E Q U A L I T Y I T S EL F
bet ween gravi t y and surf ace tension in the f al l i ng drop,
and it repeats itself under all ci rcumst ances whe r e these
are the domi nat i ng f orces. An d the shape of the at oms is
creat ed by the i nner f orces a mong particl es, whi ch once
agai n repeats itself, approxi mat el y, e ve r ywhe r e these
particles and f orces coi nci de.
But there is always variation and uniqueness in the
way the patterns manifest themselves.
Ea c h pat t ern is a generi c sol uti on to some system of
f orces in the wor l d. But the f orces are never quite the
same. Since the exact conf i gur at i on of the surroundi ngs at
any one place and time is al ways uni que, the c onf i gur a-
tion of the f orces whi c h the system is subj ect to is also
uni q ue no other system of f orces is ever subj ect to
exact l y the same conf i gur at i on of f orces. If the system is
responsible to the f orces it is subj ect to, it f ol l ows that the
system too, must be uni que; it cannot be exact l y like any
ot her, even t hough it is r oughl y si mi l ar. Th i s is not an
acci dent al consequence of the uni queness of each syst em:
it is an essential aspect of the life and whol eness of each
part .
In short
y
there is a character in natural things which
is created by the fact that they are reconciled
y
exactly
y
to their inner forces.
F o r f r om the pl ay of repetition and vari ety at every
l evel , it f ol l ows that the overal l ge ome t r y is al ways loose
and fl ui d. T h e r e is an i ndef i nabl e roughness, a looseness,
147
T H E Q U A L I T Y
a rel axedness, whi c h nat ure al ways has: and this rel axed
geomet r y comes di rectl y f r om the bal ance of the repetition
and vari et y.
I n a f orest whi ch is al i ve, it woul d be impossible f or
all the trees to be i dent i cal ; and it woul d be impossible
f or one tree itself to be al i ve, if its l eaves wer e all the
same. No system whose component parts are so unrespon-
sive to the f orces they are subj ect to, coul d mai nt ai n itself
successf ul l y; it coul d not be al i ve or whol e. It is a cruci al
f act about the whol eness of the tree that every leaf be
sl i ghtl y di f f er ent f r om the next . A n d of course, since
the same a r gume nt applies at every l evel , it means that
the component parts of nat ure are uni que at every
l evel .
This character will happen anywhere, where a part
of the world is so well reconciled to its own inner
forces that it is true to its own nature.
Al l those thi ngs whi ch we loosely call naturethe grass,
the trees, the wi nt er wi nd, deep blue wat e r , yel l ow
crocuses, f oxes, and the r a i ni n short the thi ngs whi c h
man has not ma d e a r e j ust those thi ngs whi ch are true
to their own nat ure. T h e y are j ust those thi ngs whi ch
are perf ect l y reconci l ed wi t h their o wn i nner f orces. A n d
the thi ngs whi ch are not " na t u r e " are j ust those thi ngs
whi ch are at odds wi t h their own i nner f orces.
A n d any system whi c h is whol e must have this charact er
of nat ure. T h e mor phol ogy of nat ure, the softness of its
lines, the al most i nfi ni te vari ety and the l ack of g a ps a l l
1 48
T H E Q U A L I T Y I T S EL F
this f ol l ows di rectl y f r om the f act that nat ure is whol e.
Mount ai ns , ri vers, forests, ani mal s, rocks, f l ower s all have
this charact er. But t hey do not have it si mpl y by acci dent .
T h e y have it because they are whol e , and because all
their parts are whol e. A n y system whi c h is whol e must
have this charact er.
It follows that a building which is whole must always
have the character of nature
}
too.
Th i s does not mean that a bui l di ng or a t own whi ch is
alive wi l l l ook like a tree, or like a forest. But , it wi l l
have the same bal ance of repetition and vari ety that na-
ture does.
On the one hand> patterns will repeat themselves>
just as they do in nature.
I f the pat t erns out of whi c h a t hi ng is made are al i ve,
t hen we shall see t hem over and over agai n, j ust because
they make sense. If the wa y a wi ndo w l ooks ont o a tree
makes sense, then we shall see it over and agai n; if the
rel ationship bet ween the doors make sense, we shall see it
f or al most every door ; if the wa y that the tiles are hung
makes sense, we shal l see al most all the tiles hung in
this wa y ; if the a r r a nge me nt of the ki tchen in the house
makes sense, it wi l l be repeated in the nei ghbor hood.
I n short, we shall f i nd the same el ement s, repeat i ng over
and over a g a i n a n d we shall see the r hyt hms of their
repeti ti on. T h e boards in the si di ng of the house, the
1 49
T H E Q U A L I T Y
bal usters in the rai l i ng of a bal ustrade, the wi ndows in
the bui l di ngs, panes wi t hi n the wi ndows , the same ap-
proxi mat e roof shape repeated over and agai n, the similar
col umns, similar rooms, similar cei l i ngs, or nament s re-
peat ed, trees and the boles of trees repeated in their
pat t ern, seats repeat ed, whi t ewas h repeat ed, col ors re-
peated, avenues, gar dens, f ount ai ns, roadside places, t rel -
lises, arcades, pavi ng stones, blue tiles . . . all repeat ed,
whi c hever of t hem are appropri ate in any gi ven pl ace.
On the other hand, of course, we shall find the physi-
cal farts in which the patterns manifest themselves
unique and slightly different each time that they
occur.
Because the patterns i nteract, and because the condi ti ons
are sl i ghtl y di f f er ent ar ound each i ndi vi dual occur r ence,
the col umns in an arcade wi l l al l be sl i ghtl y di f f erent ,
the boards in the si di ng of the house wi l l be sl i ghtl y
di f f erent , the wi ndo ws wi l l var y sl i ghtl y, the house wi l l
var y, trees' positions wi l l var y, seats wi l l be di f f er ent
even at the same time that they recur . . .
The repetition of patterns is quite a different thing
from the repetition of parts.
Wh e n t wo physi cal wi ndows are i denti cal the rel ati on-
ships whi ch they have to their surroundi ngs are di f f erent ,
because their surroundi ngs are di f f er ent .
But whe n the relationships to their s ur r oundi ngs t hei r
1 5 0
T H E Q U A L I T Y I T S E L F
pat t e r ns ar e the same, the wi ndows t hemsel ves wi l l all
be di f f er ent , because the sameness of the pat t erns, i nt er-
act i ng wi t h the di f f er ence of the cont ext s, makes the
wi ndows di f f er ent .
Indeed, the different, far is will be unique because the
patterns are the same.
Consi der, f or exampl e, the pattern SUNNY PLACE, whi ch
creates a spot in the sun, al ong the south side of a bui l d-
i ng, j ust wher e the out door space gets used, and wher e
the bui l di ng opens out to it. Th i s pattern may create a
series of similar spots, al ong the sout hern edge of a l ong
r ow of hous e s but then, j ust wher e the houses turn a
corner, it generat es a special pl ace, whi ch sticks out half
into the street, l ow wal l s to protect its sides, perhaps a
canvas c a n o p y a place whi ch everyone in the nei ghbor-
hood r emember s and l ooks f or.
Th i s unique pl ace is not creat ed by some arbi t rary
searchi ng f or uni queness. I t is created by the repetition of
the pattern whi ch calls f or a spot in the sun, and by the
i nteracti on of this pat t ern wi t h the wor l d.
An d we shall find the same at every scal e. Wh e r e
there are many houses, the houses wi l l be similar in f or m,
but each wi l l be uni que, accor di ng to the nat ure of the
people wh o live in it, and because each has a sl i ghtl y di f-
f erent combi nat i on of relationships to the l and, the sun,
the streets, the c ommuni t y.
T h e wi ndo ws of a gi ven house wi l l all be broadl y
similar, accor di ng to their patterns, but agai n no t wo will
151
T H E Q U A L I T Y
be the same in de t a i l e a c h wi l l be di f f erent accor di ng to
its exact position, the di recti on of the l i ght, the size of
r oom, the pl ants outside the r oom.
And from the repetition of the patterns, and unique-
ness of the parts, it follows, as it does in nature, that
buildings which are alive are fluid and relaxed in
their geometry.
Ag a i n, this doesn' t mean that bui l di ngs ought to l ook like
ani mal s, or pl ants. T h e verti cal , the hori zont al , and the
ri ght angl e are too cent ral to the nat ure of human space
to make that possible. But in a pl ace whi ch is al i ve, these
ri ght angl es are rarel y exact ; the spaci ng of parts is hardl y
ever perf ect l y even. On e c ol umn is a little thi cker t han
anot her, one angl e is a little l ar ger t han a ri ght angl e, one
door way is j ust a little smal l er than the next , each roof
line departs j ust an i nch or t wo f r om the hori zont al .
A bui l di ng in whi c h angl es are all perf ect l y ri ght angl es,
in whi c h all wi ndo ws are exact l y the same si ze, and in
whi ch all col umns are perf ect l y vert i cal , and all floors
perf ect l y hori zont al , can onl y reach its false perf ect i on by
i gnor i ng its surroundi ngs ut t erl y. T h e apparent i mper f ec-
tions of a place whi c h is alive are not i mperf ect i ons at al l .
T h e y f ol l ow f r om the process whi ch al l ows each part to
be fitted caref ul l y to its position.
This is the character of nature. But its fluidity, its
roughness, its irregularity, will not be true, unless it
is made in the knowledge that it is going to die.
1 5 2
T H E Q U A L I T Y I T S E L F
No mat t er how muc h the person who makes a bui l di ng is
able to underst and the r hyt hm of regul ari t y and i rregu-
l ari ty, it wi l l mean not hi ng so l ong as he creates it wi t h
the idea that it must be preserved because it is so preci ous.
If you wa nt to preserve a bui l di ng, you wi l l try to
make it in mat eri al s whi ch last and last f orever. Y o u wi l l
try to make sure that this creati on can be preserved i ntact,
in j ust its present state, f orever. Ca nv a s must be rul ed out
because it has to be r epl aced; tiles must be so hard that
they wi l l not crack, and set in concret e, so that they can-
not move, and so that weeds wi l l not g r o w up to split the
pavi ng; chairs must be made perf ect , of mat eri al s whi ch
never wear or f ade; trees must be nice to look at, but may
not bear f rui t, because the dropped frui t mi ght of f end
someone.
But to reach the qual i ty wi t hout a name, a bui l di ng
must be made, at least in part, of those mat eri al s whi ch
age and cr umbl e. Sof t tile and bri ck, sof t pl aster, f adi ng
coats of paint, canvas whi ch has been bl eached a little and
t orn by the wi nd, . . . f rui t, droppi ng on the paths, and
bei ng crushed by people wal ki ng over it, grass g r o wi ng
in the cracks bet ween the stones, an old chai r, pat ched,
and pai nted, to increase its c omf or t . . .
None of this can happen in a wor l d whi ch lasts f orever.
The character of nature can't arise without the pres-
ence and the consciousness of death.
So l ong as human i mages distort the charact er of nat ure,
it is because there is no whol ehear t ed accept ance of the
153
T H E Q UA L I T Y
nat ure of thi ngs. So l ong as there is not whol ehear t ed ac-
ceptance of the nat ure of thi ngs, people wi l l distort nat ure,
by exagger at i ng di f f erences, or by exagger at i ng si mi l ari -
ties. T h e y do this, ul t i mat el y, in order to stave of f the
t hought and f act of deat h.
So finally the fact is, that to come to this, to make a
thing which has the character of nature, and to be
true to all the forces in it, to remove yourself, to let
it be, without interference from your image-making
selfall this requires that we become aware that all
of it is transitory; that all of it is going to pass.
Of course nature itself is also always transitory.
The trees, the river, the humming insectsthey are
all short-lived; they will all -pass. Yet we never feel
sad in the presence of these things. No matter how
transitory they are, they make us feel happy, joyful.
But when we make our own attempt to create nature
in the world around us, and succeed, we cannot
escape the fact that we are going to die. This quality,
when it is reached, in human things, is always sad; it
makes us sad; and we can even say that any place
where a man tries to make the quality, and be like
nature, cannot be true, unless we can feel the slight
presence of this haunting sadness there, because we
know at the same time we enjoy it, that it is going to
pass.
1
54
T H E G A T E
To reach the quality without a name
we must then build a l iving pattern
language as a gate.
CHAPTER 9
THE FLOWER AND THE SEED
This quality in buildings and in towns
cannot be made, but only generated,
indirectly, by the ordinary actions of the
people, just as a flower cannot be made,
but only generated from the seed.
We are now In a fosition to recognize, at least in
hazy outline, the character of towns and buildings
with the quality without a name in them.
Next we shall see that there is a specific concrete
process by which this quality comes into being.
Indeed the main thing we shall deal with in these next
nine chapters, is the fact that the quality wi thout a name
cannot be made, but only generated by a process.
It can flow f rom your actions; it can flow wi th the
greatest ease; but it cannot be made. I t cannot be con-
trived, t hought out, designed. It happens when it flows
out f rom the process of creation of its own accord.
But we must give lip al together the idea that it is some-
thing we can capture, consciously, by wor ki ng over
drawi ngs at the dr awi ng board.
Consider the process by which the Samoans make a
canoe, from a tree.
T h e y cut the tree do wn; scrape the branches f rom the
t r unk; take off the bark; hol l ow the inside out ; carve
the outside shape of the hul l ; f orm the prow and the
stern; carve decorations on the prow . . . .
Each canoe made by this process is di f f erent ; each one is
beautiful in its own way, because the process is so or-
dinary, so simple, so direct. The r e is no time lost wonder-
i ng what kind of canoe ought we to build, what shape to
make the hull, should we put seats in i t al l those deci-
sions are made before you s t ar t s o that all the energy
1 59
T H E GA T E
and f eel i ng whi ch its makers have goes into the specific
charact er of this part i cul ar canoe. . . .
The quality of life is just like that: it cannot be made,
but only generated.
Wh e n a t hi ng is made, it has the wi l l of the maker in it.
But whe n it is gener at ed, it is gener at ed, f reel y, by the
operati on of egol ess rul es, act i ng on the reality of the situ-
ati on, and gi vi ng bi rth, of their o wn accord. . . .
T h e brush stroke becomes beaut i f ul , wh e n it is visible
onl y as the end product of a pr o c e s s whe n the f orce of
the process takes over the cramped wi l l of the maker .
T h e maker lets go of his wi l l , and lets the process take
over.
An d j ust so, any t hi ng whi ch l ives can onl y be achi eved
as the end product of a process, whose f orce takes over
and repl aces the wi l l f ul act of creat i on.
In our time we have come to think of works of art as
"creations " conceived in the minds of their creators.
An d we have come to thi nk of bui l di ngs, even t owns, also
as " c r e a t i o ns " a g a i n t hought out , concei ved enti re,
desi gned.
T o gi ve birth to such a whol e seems like a monume nt al
task: it requi res that the creat or t hi nk, f r om not hi ng, and
gi ve birth to somet hi ng whol e : it is a vast task, f or -
bi ddi ng, hug e ; it c ommands respect ; we underst and how
1 60
T H E F L OWE R AND T H E SEED
hard it is; we shri nk f r om it, perhaps, unless we are very
certai n of our po we r ; we are af rai d of it.
Al l this has def i ned the task of creati on, or desi gn, as a
huge task, in whi ch somet hi ng gi gant i c is br ought to
birth, suddenl y, in a si ngl e act, whose i nner wor ki ngs
cannot be expl ai ned, whose substance relies ul t i mat el y on
the ego of the creat or.
The quality without a name cannot be made like this.
I magi ne, by cont rast , a system of simple rules, not compl i -
cat ed, pati entl y applied, until they gr adual l y f or m a t hi ng.
T h e t hi ng may be f or med gr adual l y and built all at once,
or built gr adual l y over t i me b u t it is f or med, essentially,
by a process no mor e compl i cat ed t han the process by
whi ch the Samoans shape their canoe.
He r e there is no mast ery of unnameabl e creati ve pro-
cesses: onl y the pati ence of a c r af t s man, chi ppi ng a wa y
s l owl y; the mast ery of wha t is made does not lie in the
depths of some i mpenet rabl e e g o; it lies, i nstead, in the
simple mast ery of the steps in the process, and in the
def i ni ti on of these steps.
The same thing, exactly, is true of a living organism.
A n organi sm cannot be made. I t cannot be concei ved, by
a wi l l f ul act of creat i on, and then buil t, accor di ng to the
bl uepri nt of the creat or. I t is far too compl ex, far too
subtl e, to be born f r om a bolt of l i ght ni ng in the creat or' s
1 61
T H E GAT E
mi nd. I t has a thousand billion cells, each one adapted
perf ect l y to its c ondi t i ons and this can onl y happen be-
cause the organi sm is not " ma d e " but generat ed by a
process whi c h al l ows the gr adual adaptati on of these cells
to happen hour by hour. . . .
I t is the process whi c h creates the o r g a ni s ma nd it
must be so. No t hi ng whi ch l ives can possibly be made in
any ot her wa y .
If you zvant to make a living flower
y
you don't build
it physically
y
with tweezers
y
cell by cell. You grozv it
from the seed.
Suppose you are t r yi ng to create a flowera ne w kind of
flower. Ho w wi l l you do it? O f course you wi l l not try
to build it cell by cel l , wi t h t weez er s . Y o u k now that
any at t empt to build such a compl ex and del icate t hi ng
di rectl y woul d l ead to not hi ng. T h e onl y flowers whi ch
men have built di rectl y, piece by piece, are plastic flow-
ers. If you wa nt to make a l i vi ng flower, there is onl y
one wa y to do i t y o u wi l l have to build a seed f or the
f l ower and then let ity this seed, generat e the flower.
This hinges on a simple scientific proposition: the
great complexity of an organic system
y
which is es-
sential to its life
y
cannot be created from above
directly; it can only be generated indirectly.
T h e sheer amount of di f f erent i at i on makes this certai n.
For i nstance, in a flower there are mor e t han a billion
1 62
T H E F L OWE R AND T H E SEED
c el l s eac h one di f f er ent . Obvi ous l y, no process of con-
structi on can ever create this ki nd of compl exi t y di rect l y.
On l y those i ndi rect g r owt h processes, in whi ch order
mul tipl ies itself, onl y these kinds of processes can generat e
this bi ol ogi cal compl exi t y.
This cannot happen unless each part is at least partly
autonomous, so that it can adapt to the local condi-
tions in the whole.
T h e qual i ty wi t hout a name, like all f orms of organi c
whol eness, depends essentially on the degree of adapt a-
tion of the parts wi t hi n the whol e.
I n a system whi c h approaches the charact er of nat ure,
the parts must be adapted wi t h an al most i nfi ni te degree
of subt l et y: and this requi res that the process of adapt a-
tion be goi ng on t hr ough the system, const ant l y.
I t requires that each part at ever y l evel , no mat t er how
smal l , has the powe r to adapt itself to its own processes.
T h i s cannot happen unless each part is aut onomous.
A building which is natural requires the same.
I n the bui l di ng, ever y wi ndowsi l l and every col umn must
be shaped by an aut onomous process whi c h al l ows it to
adapt correct l y to the whol e .
Ea c h bench, each wi ndowsi l l , each tile, needs to be
made by a person, or a process, in tune wi t h the subtle
mi nut e f orces there, maki ng it a little di f f er ent at each
point al ong its l engt h and di f f er ent f r om all the others.
1 63
T H E GAT E
And the same in the town.
I n the t own, each bui l di ng and each gar den must also
by shaped by an aut onomous process, whi ch al l ows it to
adapt to its uni que parti cul ars.
Th i s vast vari ety can onl y be created by the peopl e.
Ev e r y house al ong a road must be shaped by a di f f er ent
person f ami l i ar wi t h the di f f er ent f orces pecul i ar to that
pl ace. An d wi t hi n the house, the wi ndo ws must be shaped
by people wh o are l ooki ng out , and seei ng wha t the
boundari es of the wi ndo w need to be.
T h i s does not mean that every person has to desi gn the
place he l ives in. I t si mpl y means that the l ove, and care,
and pati ence needed to bri ng ever y part into adj us t ment
wi t h the f orces act i ng on it, can onl y exist whe n each
detail ed part is cared f or, and shaped, by someone wh o
has the time and pati ence and knowl e dge to underst and
the f orces act i ng on it. I t is not essential that each person
desi gn or shape the pl ace whe r e he is goi ng to live or
wor k. Obvi ous l y people move, are happy in old houses, and
so on.
I t is essential onl y that the people of a soci ety, t oget her,
all the mi l l i ons of t hem, not j ust professi onal archi tects,
desi gn all the mi l l i ons of places. T h e r e is no other wa y
that human vari et y, and the real i ty of specific human
lives, can find their wa y into the st ruct ure of the places.
But of course, autonomous creation of the farts, if
taken by itself, will produce chaos.
1 64
T H E F L OWE R AND T H E SEED
T h e parts wi l l not f or m any l ar ger whol e, unless the
i ndi vi dual adapt at i on of the parts is under some sort of
deeper regul at i on, whi ch guar ant ees t hat the l ocal process
of adaptati on wi l l not onl y make the l ocal part trul y
adapt ed to its o wn processes, but that it wi l l also be shaped
to f or m a l ar ger whol e.
What makes a flower whole> at the same time that all
its cells are more or less autonomous
y
is the genetic
code
y
which guides the process of the individual parts
y
and makes a whole of them.
T h e di f f er ent cells are able to act in har mony because
each one of t hem cont ai ns the same genet i c code.
Ea c h part ( c el l ) is f ree to adapt l ocal l y to its o wn pro-
cesses, and is hel ped in this process by the genet i c code
whi ch gui des its g r owt h.
Y e t at the same ti me, this same code contai ns f eat ures
whi ch guar ant ee that the sl ow adaptati on of the i ndi vi dual
parts is not mer el y anarchi c, and i ndi vi dual , but that
each part si mul t aneousl y helps to create those l ar ger parts,
systems, and pat t erns whi c h are needed f or the whol e.
And
y
fust as the flower needs a genetic code to keep
the wholeness of its parts
y
so do the building and the
town.
T h e i ndi vi dual bui l di ng needs a code, whi ch guar ant ees
t hat all the col umns and the wi ndows , as they get i n-
165
T H E GAT E
di vi dual l y shaped, wi l l f or m a whol e. I t must provi de
the i ndi vi dual bui l der wi t h a sequence of i nstructi ons so
cl ear, and so fluid, that he can f reel y make each porti on
of the bui l di ng perf ect l y, accor di ng to its pl ace.
A n d the t own needs a code, whi ch makes the ma ny ac-
tions of the great vari et y of people whol e. I t must provi de
the people of the t own wi t h i nstructi ons so cl ear that all
of t hem can take part in the shapi ng of the t o wn: j ust
like tlie genet i c process whi ch creates the flower, this
process must al l ow each person to shape his o wn corner of
the wor l d, so that each bui l di ng, each r oom, each door-
step, is uni que accor di ng to its pl ace wi t hi n the w h o l e
but wi t h the built-in guar ant ee that the t own whi ch
emer ges f r om these i ndependent acts, wi l l also be alive
and whol e.
So I began to wonder if there was a code
}
like the
genetic code
}
for human acts of building?
Is there a fluid code, which generates the quality
without a name in buildings> and makes things live?
Is there some process which takes place inside a per-
son's mind
y
when he allows himself to generate a
building or a place which is alive? And is there in-
deed a process which is so simple too
y
that all the
people of society can use it
y
and so generate not only
individual buildings
}
but whole neighborhoods and
towns?
It turns out that there is. It takes the form of
language.
1 66
CHAPTER IO
OUR PATTERN LANGUAGES
The people can shape buildings for
themselves, and have done it for cen-
turies, by using languages which I call
pattern languages. A pattern language
gives each person who uses it, the power
to create an infinite variety of new and
unique buildings, just as his ordinary
language gives him the power to create
an infinite variety of sentences.
We know, from chapter g, in very vague and general
terms, that life cannot be made, but only generated
by a process.
In the case of buildings and of towns, this process
must be one which lets the people of a town shape
rooms, and houses, streets, and churches, for them-
selves.
No w we shall begin to see what kind of processes can
make this possible.
In traditional cultures these processes were common-
place.
Each person knew how to make a house, a wi ndow, or a
bench, just ri ght.
Each building was a member of a family, and yet
unique.
Ev e n though there are a hundred farmhouses, in a valley
of the alps, all similar, yet still each one is beauti ful , and
special to the place where it occurs, and fil l ed wi t h the
same elements, but in unique combinations, so that it is
alive and wonder f ul .
Each room is a little different according to the view.
Each garden is di f f erent accordi ng to its relation to the
sun; each path di f f erentl y placed accordi ng to the best
route f rom the street; each stair has a slightly di f f erent
174
OUR P A T T E R N L A NGUA GE S
slope, wi t h di f f er ent steps, to fi t j ust ni cel y in bet ween
the rooms wi t hout a wast e of space. . . .
Each tile is set a little differently in the ground
y
ac-
cording to the settling of the earth.
Ea c h wi ndowpa ne is sl i ghtl y di f f er ent accor di ng to the
shri nkage of the wo o d ; each wi ndo w di f f er ent accor di ng
to the vi ew it l ooks upon; each shelf di f f er ent accor di ng
to wha t it wi l l carry, and how it is pl aced; each orna-
ment a di f f er ent col or, accor di ng to the or nament s and
col ors round about i t; each col umn wi t h a di f f er ent
capital , accor di ng to the mo me nt in the life of the car ver
wh o made i t; each step wo r n di f f er ent l y, accor di ng to the
wa y that f eet move over i t ; each door a sl i ghtl y di f f erent
hei ght and shape accor di ng to its position in the f r a me ;
each pl ant a di f f er ent one accor di ng to the angl e of the
sun, and the position of the wi nd; each f l ower box wi t h
di f f er ent flowers in it accor di ng to the likes and dislikes of
the people i nsi de; each stove made di f f er ent l y, accor di ng
to the number of the people in the r oom, and the size of
the r oom; each board cut to fi t its posi ti on; each nail
dri ven accor di ng to the gi ve and shri nkage of the wood.
How was this possible?
Ho w was it possible that any simple f ar mer coul d make a
house, a t housand ti mes mor e beaut i f ul t han all the st rug-
gl i ng archi tects of the last f i f t y years coul d do?
Or s t i l l s i mp l e r ho w, f or i nstance, coul d he make
175
T H E GAT E
a barn? Wh a t is it that an i ndi vi dual f ar mer di d, whe n he
deci ded to build a barn, that made his barn a member of
this f ami l y of barns, si mi l ar to hundr eds of ot her barns, yet
nevert hel ess uni que?
At first sight, we might imagine that each farmer made
his barn beautiful, simply by paying attention to its
function.
Ev e r y barn must have a doubl e door, so that the f ar mer
can dri ve his ha y - wa g on ri ght i nto the barn f or unl oad-
i ng; every barn must provi de enough hay storage to f eed
the c ows t hr oughout the wi nt e r ; it must al l ow the c ows to
stand in a wa y that makes it easy to feed t hem, and easy
to move the hay f r om the pl ace wher e it is stored to the
place wher e the c ows eat ; it must provi de an easy wa y of
was hi ng do wn the c ow dung and urine whi ch ac c umu-
l at e; it must provi de a way of support i ng the roof and
wal l s agai nst wi nd l oads. . . .
Ac c or di ng to this t heory, the f ar mer is abl e to make
liis barn beaut i f ul , because he is so deepl y in touch wi t h
it f unct i on.
But this does not explain the similarity of different
barns.
I f ever y ne w barn we r e created f r om scrat ch, purel y
f r om the f unct i onal nat ure of the probl em, we shoul d
expect to see a muc h gr eat er vari ety of f or ms than
actual l y exists. Wh y are there no ci rcul ar barns? Wh y
1 7 6
OUR P A T T E R N L ANGUAGE S
do some barns not have a doubl e nave, to provi de even
mor e st orage, or a doubl e pi tched roof ? I t may be true
that these kinds of barns woul d not wor k as wel l as the
ones whi c h are bui l t ; but how coul d the bui l ders know
that, wi t hout t r yi ng it?
T h e f act is that they don' t try it. T h e y are si mpl y copy-
i ng the ot her barns whi ch they al ready know.
An d , i ndeed, ever yone wh o has ever built anyt hi ng
knows that he goes about it in this wa y . Wh e n you put
floor joists at 1 6 " cent ers, you don' t wor k out the st ruc-
tural cal cul ati ons every ti me you do i t; once you are
persuaded that this is a good wa y to build floors, you go
on doi ng it that wa y , unti l you have some reason to re-
thi nk it.
We might imagine then, that the farmer got his
-power to build a barn by copying the other barns
around him.
I magi ne f or a mome nt that the f ar mer act ual l y had a
detailed pi cture of anot her barn, or several other barns
in his mi nd, compl et e down to the last details, and that
whe n he starts to make his o wn barn, he si mpl y modi f i es
this ideal barn in his mi nd.
T h i s woul d cert ai nl y expl ai n wh y one barn looks like
other barns in the val l ey, even whe r e purel y f unct i onal
consi derati ons don' t require it.
But this does not explain the great variety of barns.
1 7 7
T H E GA T E
An d it does not expl ai n the enor mous vari ati ons whi ch the
f ar mer is able to make, in his o wn barn, wi t hout goi ng
wr o ng .
Fo r exampl e, a mo ng the old barns in Cal i f or ni a, I know
t wo whi c h are radi cal l y di f f er ent f r om the " s t a nda r d"
type. On e of t hem has the same cross section as u s u a l
but it is very very l ong, about 240 f e e t a n d its mai n
doors, instead of bei ng at the ends, run t hr ough it, at
ri ght angl es to the mai n axis. T h e ot her one is nestled
i nto the slope of a hill, and it has three stories. T h e t wo
l owe r stories are j ust like the nor mal floors of a barn,
but one above the ot her, and approached f r om opposite
di recti ons.
Y o u can say that these barns are copies too. But , obvi -
ousl y in these cases, the total ar r ange me nt of the " t ypi c al "
barns has not been copied at all. T h e patterns whi ch are
typical of ot her barns are still present in these t wo barns;
but the wa y in whi ch the patterns are combi ned is utterl y
di f f erent .
The pro-per answer to the question
}
"How is a farmer
able to make a new barn?
>}
lies in the fact that every
barn is made of patterns.
I t is not the idea of copyi ng whi ch is at f aul t ; onl y the
concept i on of " wh a t is copi ed. " Obvi ous l y the f ar mer
does have some sort of i mage of a barn in his mi nd, whe n
he starts to make a ne w barn. But this i mage of the barn,
whi ch he has in his mi nd, is not an i mage like a dr a wi ng
or a bl uepri nt or a phot ograph. I t is a system of patterns
whi ch f unct i ons like a l anguage.
178
OUR P A T T E R N L ANGUAGE S
An d the f ar mer is able to make a ne w barn, unl i ke the
ones whi ch he has seen bef ore, by t aki ng all the patterns
whi ch he knows , f or barns, and combi ni ng t hem in a ne w
wa y .
These patterns are expressed as rules of thumb> which
any farmer can combine and re-combine to make an
infinite variety of unique barns.
Her e are some of the patterns f or tradi ti onal Cal i f or ni a
barns.
Ma k e a barn in the shape of a rect angl e, 3 0 - 5 5 feet
wi de, 4 0 - 2 5 0 f eet l ong, the l engt h at least 3 * feet,
wher e * is the number of c ows the barn has to hol d.
Or i e nt the barn so t hat its ends connect easily wi t h the
paths whe r e c ows come in f r om the fiel ds, and wi t h the
l ocal road.
Di vi de the inside of the barn i nto three paral l el aisles:
t wo c ow mi l ki ng aisles do wn the out er sides, and a cent ral
hay- st orage aisle.
Ma k e the cent ral aisle 1638 feet wi de, and the outer
aisles 1 01 6 f eet wi de. I n certai n cases, one of the side
aisles can be short er t han the cent ral aisle, thus t aki ng a
not ch out of the rect angl e.
Be t we e n the outer edge of the cent ral aisle and the t wo
out er aisles, pl ace t wo r ows of col umns. T h e col umns are
equal l y spaced, and the distance bet ween the last col umn
and the end wal l is equal to the distances bet ween
col umns. Choose a c ol umn spaci ng bet ween 7 and 17
f eet .
I f the c ol umn spaci ng is 7 - 1 0 f eet , make the col umns
1 7 9
T H E GAT E
4 x 4
>
s. I f the c ol umn spaci ng is 1 0 - 1 4 f eet , make the
col umns 6 x 6' s. I f the col umn spaci ng is 1 41 7 f eet ,
make the col umns 8 x 8' s. T h e col umns are tied to-
get her , al ong the l engt h of the barn, by the mai n purl i ns
sitting on top of the col umns.
Ma k e the roof of the barn a symmet r i cal pi tched
roof , and make the pitch over the outer aisles flatter, or
equal to, the pitch over the cent ral ai s l es o that the
pitch wi l l usual l y break over the mai n col umns al ong the
purl ins. Bot h pitches are bet ween 20 and 40 degrees to
the hor i zont al .
I f the l engt h of the barn is less than 1 5 0 f eet , place the
mai n doors at the ends, r oughl y on the cent er l ine of the
cent ral aisle. If the barn is mor e t han 1 50 f eet l ong,
place the mai n doors in the side wal l s, r oughl y ha l f wa y
al ong, and let the side aisles be i nt errupt ed by the
doors.
I f the t wo r ows of col umns whi ch def i ne the cent ral aisle
are mor e t han 18 f eet apart , tie t hem t oget her by hori -
zont al tie beams, all at the same hei ght , and wi t hi n 3 f eet
of the tops of the col umns.
Ma k e the side wal l s 710 f eet hi gh, and the peak of the
roof 1 5 - 2 5 feet hi gh.
Fr a me the side wal l s by a system of vert i cal studs, con-
nect ed by hori zont al sill ( bot t om) and plate ( t o p) , and,
if you wi sh, by a mi ddl e hor i zont al me mb e r a l l these
member s 2 x 4*s.
Pl ace the studs in the side wal l s to line up wi t h the
col umns of the cent ral aisle, and pl ace the mai n raf t ers in
the same pl anes as the studs and col umns, si tti ng on the
plates and purl i ns whi ch run over these member s.
1 80
OUR P A T T E R N L ANGUAGES
Pl ace raf t ers f r om opposite sides of the roof , meet i ng
the mai n ri dge beam.
Br ac e ever y corner in the f r a mi ng of the side wal l s
wi t h a di agonal 2 x 4 , about 3 f eet l ong.
Co nne c t the tie beams r unni ng across the cent ral aisle to
the mai n col umns, by di agonal braces.
Co nne c t the mai n purl i ns to the mai n col umns, wi t h
di agonal braces 34 f eet l ong. If the col umn spaci ng is
mor e t han about 21 f eet also use doubl e braces, the outer
ones about 6 f eet l ong.
To understand
y
in detail
y
how these patterns work we
must extend our definition of "a pattern "
I n chapters 4 and 5 we l earned to see a pattern as
somet hi ng " i n the w o r l d " a uni tary pat t ern of acti vi ty
and space, whi ch repeats itself over and over agai n, in any
gi ven pl ace, al ways appeari ng each ti me in a sl i ghtl y di f-
f erent mani f est at i on.
Wh e n we ask, no w, j ust wher e these pat t erns come
f r om, and also wher e the vari at i on comes f r om, whi ch al-
l ows each pat t ern to take on a sl i ghtl y di f f er ent f or m each
ti me that it occurs, we have been l ed to the idea that
these pat t erns " i n the wo r l d " are creat ed by us, because
we have ot her, similar pat t erns in our mi nds f r om whi ch
we i magi ne, concei ve, creat e, bui l d, and live these actual
patterns in the wor l d.
The s e pat t erns in our mi nds are, mor e or less, ment al
i mages of the pat t erns in the wor l d: they are abstract rep-
resentati ons of the very morphol ogi cal rules whi c h def i ne
the patterns in the wor l d.
1 81
T H E GAT E
Ho we v e r , in one respect they are very di f f er ent . T h e
patterns in the wor l d mer el y exist. But the same patterns
in our mi nds are dynami c. T h e y have f orce. T h e y are
generat i ve. T h e y tell us wha t to do; they tell us how
we shal l , or may, gener at e t he m; and they tell us too,
that under certai n ci rcumst ances, we must create t hem.
Each pattern is a rule which describes what you have
to do to generate the entity which it defines.
Consi der , f or exampl e, the pat t ern of hillside t erraci ng,
used in hilly countri es, to make usable f ar ml and on hilly
slopes. A s a " f a c t , " this pat t ern mer el y has certai n
characteri sti cs. For exampl e: the terraces f ol l ow the
cont our l i nes; the terraces are spaced vert i cal l y at
r oughl y equal i nt erval s; the terrace is f or med by a wal l ,
al ong its outer edge, whi c h keeps the earth f r om sl i di ng;
each of these outer wal l s rises sl i ghtl y above the l evel of
the terrace whi ch it retains, so that it also keeps wat er
there, evens out the rai nf al l , and prevent s erosi on. Al l
this def i nes the pat t ern. The s e are the relationships whi ch
def i ne the pattern " i n the wo r l d . "
No w consi der the same pattern " i n the f ar mer ' s mi nd. "
I t contai ns the same i nf or mat i on: more detai l ed prob-
abl y, less superfi ci al . But it contai ns, in addi ti on, t wo
other aspects. Fi rst , it i ncl udes the knowl e dge whi c h is
requi red to build a system of terraces like this. T h e f act
that the wal l s are built bef ore the terraces are fi l l ed in and
l evel ed; the f act that there are smal l drai n hol es in the
outer wal l s ; in short, the t erraci ng is described no w as a
1 82
OUR P A T T E R N L ANGUAGES
rule. I t is a rule whi ch tells the f ar me r wha t to do on an
exi sti ng hillside to t r ans f or m it into the state whi ch has
this pat t ern in i t i n short, to gener at e the pat t ern itself,
in the wor l d.
An d there is an i mperat i ve aspect to the pat t ern. T h e
pat t ern sol ves a probl em. I t is not mer el y " a " pat t ern,
whi ch one mi ght or mi ght not use on a hillside. I t is a
desirable pat t er n; and f or a person wh o want s to f ar m a
hillside, and pr event it f r om erosi on, he must create this
pat t ern, in order to mai nt ai n a stable and heal t hy wor l d.
I n this sense, the pat t ern not onl y tells hi m how to create
the pat t ern of t erraci ng, if he want s to; it also tells hi m
that it is essential for hi m to do so, in certai n parti cul ar
cont ext s, and that he must create this pat t ern there.
It is in this sense that the system of patterns forms a
language.
Wh e n the barn bui l der applies the patterns f or a barn to
one anot her in the proper order, he is able to create a
barn. T h i s barn wi l l al ways have the parti cul ar rel ati on-
ships requi red by the pat t erns; howe ve r , all ot her sizes,
angl es, and relationships depend on the needs of the situa-
tion, and the whi m of the bui l der. T h e f ami l y of barns
produced by this system all share the morphol ogi cal f ea-
tures specified by the rules (these are the morphol ogi cal
l aws we have obs er ved) , but beyond that there is l i teral l y
endl ess vari et y.
From a mathematical point of view, the simplest kind
of language is a system which contains two sets:
1 83
T H E GA T E
1. A set of el ement s, or symbol s.
2. A set of rul es f or combi ni ng these symbol s.
T h e l ogi cal l anguages are an exampl e. I n a l ogi cal
l anguage, the symbol s are compl et el y abstract, the rul es
are the rul es of l ogi cal synt ax, and the sentences are cal l ed
we l l - f or me d f ormul as. Fo r i nstance, such a l anguage
mi ght be def i ned by the set of symbol s + , = and by
the rul e " T h e same symbol must never appear t wi ce
in a r o w. " I n this l anguage, *- [ - *- [ - *- [ - *- [ - . * and
= *x woul d be sent ences ( or we l l - f or me d
f or mul as ) , but ^ r z i - r r ^ - f
-
* *
-
!
-
woul d not be, because
* appears t wi ce in a r ow.
A natural language like English is a more complex
system.
Ag a i n, there is a set of el ement s, in this case the set of
wor ds. A n d agai n there are rul es whi ch describe the
possible ar r ange me nt of the wor ds . But , there is, in addi -
tion, a st ruct ure on the wo r d s t h e compl ex ne t wor k of
semant i c connect i ons, whi ch def i nes each wo r d in t erms
of ot her wor ds , and shows how wor ds are connect ed to
other wor ds .
T a k e f or i nstance, a ver y simpl e sentence like " T h e
tree is st andi ng on the hi l l . " T h e wor ds here are el e-
ment s: " T h e , " " t r e e , " " hi l l " . . . and so on. T h e
el ement s are combi ned accor di ng to certai n rul es, whi c h
create a sent ence. T h e simplest of these rul es are the rul es
of g r a mma r , whi ch make it cl ear that the wor d " t o be "
must be t r ansf or med i nto " i s " in this cont ext ; that the
1 84
OUR P A T T E R N L ANGUAGE S
wor d " t h e " comes bef ore the nouns to whi ch it ref ers,
and so on.
Fur t her , the meani ng of the sentence comes f r om the
ne t wor k of connect i ons a mong the wor ds whi ch tells us,
f or exampl e, that a " t r e e " g r o ws in the " g r o u n d " and
that a " hi l l " is a kind of " g r o u n d , " and that a tree can
t heref ore stand on a hill.
A pattern language is a still more complex system of
this kind.
T h e el ement s are patterns. T h e r e is a st ruct ure on the
patterns, whi c h describes how each pattern is itself a
pattern of ot her smal l er patterns. An d there are also
rules, embedded in the patterns, whi c h describe the wa y
that they can be creat ed, and the wa y that they must be
ar r anged wi t h respect to other pat t erns.
Ho we v e r , in this case, the patterns are both el ement s
and rul es, so rul es and el ement s are i ndi sti ngui shabl e. T h e
patterns are el ement s. A n d each pat t ern is also a rul e,
whi ch describes the possible ar r angement s of the el ement s
t he ms e l v e s agai n ot her pat t erns.
An ordinary language like English is a system which
allows us to create an infinite variety of one-
dimensional combinations of words, called sentences.
Fi rst of al l , it tells us whi c h ar r angement s of wor ds are
l egi t i mat e sentences, in a gi ven si tuati on, and whi c h are
not. An d , f ur t her mor e, whi ch ar r angement s of wor ds
1 85
T H E GAT E
make sense in any gi ven situation, and whi c h ones don' t .
I t nar r ows down the total possible ar r angement s of wor ds
whi c h woul d make sense in any gi ven situation.
Second, it act ual l y gi ves us a system whi ch al l ows us
to produce these sent ences whi c h make sense. So, it not
onl y def i nes the sentences whi c h make sense in a gi ven
si tuati on; it also gi ves us the apparatus we need to create
these sentences. It is, in other wor ds , a generat i ve system,
whi ch al l ows us to generat e sentences that are appropri ate
to any gi ven si tuati on.
A pattern language is a system which allows its users
to create an infinite variety of those three dimensional
combinations of patterns which we call buildings
y
gardens
y
towns.
Fi rst, it def i nes the l i mi ted number of ar r angement s of
spaces that make sense in any gi ven cul t ure. T h i s is a f ar
smal l er col l ecti on t han the total number of ar r angement s
of j umbl ed nonsense, the piles of bricks and space and
air and wi ndows , ki tchens on top of f r e e wa y i nt er-
changes, trees g r o wi ng upside do wn inside a r ai l way
s t at i ont hat coul d be put t oget her, but woul d make no
sense at all.
An d second, a pat t ern l anguage act ual l y gi ves us the
power to generat e these coherent ar r angement s of space.
T h u s , as in the case of nat ural l anguages, the pat t ern l an-
guage is generative. It not onl y tells us the rul es of ar -
r angement , but shows us how to const ruct ar r angement s
a s many as we wa n t wh i c h satisfy the rul es.
1 86
OUR P A T T E R N L ANGUAGE S
In summary: both ordinary languages and pattern
languages are finite combinatory systems which allow
us to create an infinite variety of unique combina-
tions, appropriate to different circumstances, at will.
Natural Language
Wo r d s
Rul es of g r a mma r and
me ani ng whi ch gi ve con-
necti ons
Sent ences
Pattern Language
Pat t er ns
Pat t er ns whi c h speci fy
connect i ons bet ween
pat t erns
Bui l di ngs and pl aces
Here is the outline of a pattern language for a farm-
house in the Bernese Oberland.
NORTH SOUTH AXIS
WEST FACING ENTRANCE DOWN THE SLOPE
TWO FLOORS
HAY LOFT AT THE BACK
BEDROOMS IN FRONT
GARDEN TO THE SOUTH
PITCHED ROOF
HALF- HIPPED END
BALCONY TOWARD THE GARDEN
CARVED ORNAMENTS
Ea c h of these pat t erns is a field of relationships whi c h can
take an i nfi ni te vari ety of specific f orms. An d , in addi ti on,
each one is expressed in the f or m of a rul e, whi c h tells the
f ar mer wh o is maki ng his house j ust wha t to do.
1 87
T H E G A T E
Y o u can see that the vari et y of possible houses whi c h
such a si mpl e syst em of pat t erns can creat e is al most i n-
f i ni t e. F o r i nst ance, here are some houses whi c h it g e n-
erates.
Here is the outline of another simple pattern lan-
guage for stone houses in the South of Italy.
SQUARE MAIN ROOM, ABOUT 3 METERS
TWO STEP MAIN ENTRANCE
SMALL ROOMS OFF THE MAIN ROOM
ARCH BETWEEN ROOMS
MAIN CONICAL VAULT
SMALL VAULTS WI THI N THE CONE
WHITEWASHED TOP TO THE CONE
FRONT SEAT, WHITEWASHED
188
OUR P A T T E R N L ANGUAGE S
Th i s l anguage generat es the ver y simple houses in this
dr a wi ng :
An d the more compl i cat ed, and less similar houses, in this
second dr a wi ng :
UL
189
T HE GATE
In this casey the pattern language not only helps the
people shape their housesy but also helps them shape
their streets and town collectively.
For instance, there are further patterns in the l anguage
whi ch i ncl ude:
NARROW STREETS
STREET BRANCHING
190
OUR PATTERN LANGUAGES
FRONT DOOR TERRACES
CONNECTED BUILDINGS
PUBLIC WELLS AT INTERSECTIONS
STEPS IN THE STREET
These larger patterns create the structure of the town. If
every person who makes an individual house, at the same
time fol l ows these larger patterns, step by step, and does
whatever he can with the layout and placing of his house
to help create these l arger patterns too, then the town
slowly gets its structure from the incremental aggregati on
of their individual acts.
Each person uses the l anguage a little differently. Each
person uses the l anguage to make a building which re-
flects his dreams, to meet the special needs of his own
family, the way they live, the animals they keep, the site,
and its relation to the street . . . But overall, throughout
the differences, there is a constancy, a harmony, created
by the repetition of the underl ying patterns.
At this stage, we have defined the concept of a pat-
tern language clearly. We know that it is a finite
system of rules which a person can use to generate an
infinite variety of different buildingsall members
of a familyand that the use of language will allow
the people of a village or a town to generate exactly
that balance of uniformity and variety which brings a
place to life.
In this sense, then, we have found an example of the
kind of code which does, at certain times play just the
1,91
T HE GATE
role in buildings and in towns that the genetic code
flays in a living organism.
What we do not know yet, is that these kinds of
languages are ultimately responsible for every single
act of building in the world.
1 92
CHAPTER I I
OUR PATTERN LANGUAGES:
CONTINUED
These pattern languages are not confined
to villages and farm society. All acts of
building are governed by a pattern
language of some sort, and the patterns
in the world are there, entirely because
they are created by the languages which
people use.
We have seeny so fary that pattern languages were the
secret of the farmer's power to build in simple vil-
lages.
But languages are more widespread^ and more pro-
found than that. The fact is that every work of build-
ingy large or smally humble or magnificenty modern
or ancient, is made in this same way.
For the use of pattern languages is not merely
something that happens in traditional societies. It is
a fundamental fact about our human naturey as funda-
mental as the fact of speech.
For example, our own towns and buildings, just like any
others, are all made of patterns, too.
Look around our worl d. Our worl d is made of free-
ways, gas stations, houses, sidewalks, kitchens, buildings,
bare concrete walls, flat roofs, front doors, television,
parking garages, skyscrapers, elevators, high schools, hos-
pitals, parks, parki ng places, gutters, trees in concrete
boxes, tubs of artificial fl owers, neon signs, telephone
wires, picture wi ndows, front gardens, back gardens, gilt
plastic-framed pictures, motels, supermarkets, hamburger
joints, sandwich machines.
The patterns of our timey like all other patterns in
the built environmenty come from the pattern lan-
guages which people use.
For instance, the f reeways are built f rom handbooks,
which contain, more or less exactly in the f orm of pat-
terns, rules which prescribe the optimum spacing of exits,
198
OUR PATTERN LANGUAGES I CONTI NUED
at di fferent densities, the best configurations for the exits
under di fferent conditions, the proper curvature and in-
clination of the petals of a cloverleaf . . . .
And the gas stations built by any one company are often
built f rom a little book, whi ch describes the essential
features of, for instance, a "Shel l " gas st at i onand de-
scribes how these essential features may be combined dif-
ferently, in di fferent situations, to provide a gas station
whi ch is still one of the family of Shell gas stations, but
adapted to a local site.
Indeedy as we shall see now, these patterns always
come from languages. They come into the man-made
worldy because we always fut them thereand we
fut them there by using languages.
Each wi ndow, room, house, street and neighborhood, gets
those patterns which identify it, whi ch give it its structure,
from a l anguage: and each entity within the worl d is
governed, and guided in its development, by an internal
pattern l anguage whi ch functions for it, just as the
genetic code works for an organism.
Of coursey these patterns do not come only from the
work of architects or flanners.
Archi tects are responsible for no more than perhaps 5
percent of all the buildings in the worl d.
Most buildings, streets, shops, offices, rooms, kitchens,
cafes, factories, gas stations, freeways, bridges . . .
199
T HE GATE
which give the worl d its form, come from an entirely
di fferent source.
They come from the work of thousands of different
people.
The y come from the decisions of administrators, hard-
ware storeowners, housewives, the officials in the build-
ing department, local bankers, carpenters, public works
departments, gardeners, painters, city councils, fami-
lies . . . .
Each of them builds by following some rules of
thumb.
Exampl e: T h e British government makes the decision to
build Stevenage Ne w T o wn , a t own of 50, 000 people,
30 miles outside of London. T h e pattern governi ng this
decision was created by Ebenezer Howard in 1890, and
was known to the British Government for 50 years
before they used it to.build Stevenage.
Exampl e : A group of hi ghway engineers, from the Cal i -
fornia State Hi ghway Depart ment , locate, and design, a
f reeway interchange, on Interstate 80, east of San Fr an-
cisco. Th e y are f ol l owi ng patterns which are explicitly
laid down in the f orm of rules, in the A A S H O manual :
these rules define the optimum spacing of f reeway inter-
changes, most efficient ramp configurations, mi ni mum
radii and super-elevations for di fferent design speeds, etc.
Exampl e: A Ne w Yor k architect defines the outer shape
200
OUR PATTERN LANGUAGES I CONTI NUED
of an office building on Park Avenue. He is confined by
l aw, to make the building envelope conform to the day-
light requirements of the building code, and knows before
he starts that he will have to create a more or less
pyramidal envelope.
Exampl e: A housewife asks her husband to build shelves
across the kitchen wi ndows, the way she saw in last
month' s House and Garden. Agai n, the pattern, which
says that shelves across a kitchen wi ndow are a good
idea in general , is in her mind before her decision to try
it in her own kitchen.
Everybody follows rules of thumb.
Exampl e: A man who is fi xi ng the bathroom goes to the
local hardware store, and buys an expandi ng shower
curtain rail, which can be force-fitted between the bath-
room walls above the bathtub. T h e fact that this fixture
is available on the market, and is the easiest to fi x, is the
controlling force behind the pattern in his mind which
tells him how to place the shower curtain rod.
Exampl e: A small t own decides to close off the central
street in t own, to f orm a pedestrian precinct. It is prob-
ably acting under guidance from architects: and the
architects base their advice on a pattern that has been
emergi ng in architectural thought for more than twenty
years.
Exampl e: Th e landscape architect who is called in to do
the detailing of the pedestrian precinct, uses brick walks,
planters, and benchesal l part of the current vernacul ar
201
T HE GATE
for pedestrian precincts, and all in his mind l ong before
he started this particular j ob.
Exampl e: A bank decides to lend money to one devel-
oper, and not to another. Th e bank bases its decisions on
rules of thumb about the density of land coverage whi ch
will bring a reasonable financial return. Thei r patterns tell
them not to lend money to people who want to put small
buildings on large pieces of l and, in central cities.
Exampl e: Th e Parks Depart ment is thinning the trees in
the park. If they are pine trees, they are l eft spaced at
about 15 foot centers; any extra trees are taken out, so
that the trees won' t stop each other from gr owi ng. Thi s
spacing for pine trees is a wi del y known pattern taught
in forestry school and used all over the worl d.
And all these rules of thumbor fat ternsare fart
of larger systems which are languages.
For, of course, these rules of thumb, which I have given
as examples, do not exist, independently, isolated, free-
floating.
Each one is part of a system of other rules of thumb,
organi zed, so that the rules of thumb, or patterns, can be
used, not only to make isolated decisions, but to create
complete t hi ngscompl et e parks, buildings, park benches,
f reeway interchanges . . . and so on.
Every ferson has a fattern language in his mind.
Yo ur pattern l anguage is the sum total of your knowl -
202
OUR PATTERN LANGUAGES I CONTI NUED
edge of how to build. T h e pattern l anguage in your mind
is slightly di fferent from the l anguage in the next person's
mi nd; no two are exactly alike; yet many patterns, and
fragments of pattern languages, are also shared.
Whe n a person is faced with an act of design, what he
does is governed entirely by the pattern l anguage which
he has in his mind at that moment. Of course, the pattern
l anguages in each mind are evol vi ng all the time, as each
person's experience grows. But at the particular moment
he has to make a design, he relies entirely on the pattern
l anguage he happens to have accumulated up until that
moment. His act of design, whether humble, or giganti-
cally complex, is governed entirely by the patterns he has
in his mind at that moment, and his ability to combine
these patterns to form a new design.
This is as true of any great creative artist, as of the
humblest builder.
Palladio used a pattern l anguage to make his designs.
And Frank Ll oyd Wr i g ht too used a pattern l anguage to
make his designs. Palladio happened to record his patterns
in books, with the idea that other people could use them
too. Wr i ght tried to keep his patterns secret, like a master
chef who keeps his recipes secret. But this difference is
inessential. Wh a t matters is that both of them, and all the
other great architects who have ever lived, have had their
own pattern l anguages, the condensation of their own
experience, in the form of private rules of thumb, which
they could use whenever they began to make a building.
203
T HE GATE
And you yourself make your designs by using a fat-
tern language.
Imagi ne for a moment that I ask you to lay out a simple
cottage for yourself.
No w, let me ask you this: Ar e the rooms in your
cottage circular? Al most certainly not. Most likely you
have a rule in your mind whi ch tells you that the
rooms in your building should be more or less rough
rectangles.
I do not say, for the moment, that this rule is good or
bad. I only ask you to recognize that you do have a rule
of some kind, which tells you roughl y what kind of shape
to make the rooms. . . .
And you have many, many rules like this.
Indeed it is the system of these rules that is your
-present language.
And your creative power is entirely given by the power
of these patterns. Yo ur power to create a building is
limited entirely by the rules you happen to have in your
l anguage now.
At the moment when a person is faced with an act of
designy he does not have time to think about it from
scratch.
He is faced with the need to act, he has to act fast; and
the only way of acting fast is to rely on the various rules
204
OUR PATTERN LANGUAGES I CONTI NUED
of thumb whi ch he has accumulated in his mind. In short,
each one of us, no matter how humbl e, or how elevated,
has a vast fabric of rules of thumb, in our minds, whi ch
tell us what to do when it comes time to act. At the time
of any act of design, all we can hope to do is to use the
rules of thumb we have collected, in the best way we
know how.
Even when a person seems to "go back to the basic
problem" he is still always combining patterns that
are already in his mind.
Al t hough he may manage to transform these patterns,
slightly, according to a new analysis of the problem, it is
still the pattern l anguage in his mind whi ch forms the
groundwork of what he does.
You may think: We l l , I do not have any kind of pat-
tern l anguage in my mind at present.
Ther e are people who may deny the existence of patterns
in their own minds. T o such a person I ask a simple
question: If you know anythi ng about how to make
buildings, what is it that you know?
Your answer may be that you rely on the depths of your
emotion and intuition to respond, in a unique fashion, to
each new problem that presents itself to you. But even
this emotion and intuition is guided by some principles
however deep. Even if you have never tried to make these
principles explicit for yourself, and even if you cannot do
so, still, somewhere in your mind there are these princi-
ples, couched in who knows what f or ma nd it is these
205
T HE GATE
principles whi ch come into action, through intuition and
emotion, when you make a design.
It is only because a person has a pattern language in
his mindy that he can be creative when he builds.
Yo u may be unwi l l i ng to admit that your creative power
comes from a l anguage in your mind, because you are
afraid that the rules of a l anguage in your mi nd may
prevent you f rom being free and creative. T h e very op-
posite is true. A pattern l anguage is the very source of
creative power in the individuals who use it, and wi th-
out a l anguage they could create nothing. It is the
l anguage they could create nothing. It is the l anguage
whi ch makes them creative.
Remember Engl i sh. It woul d be ridiculous to say that the
rules of Engl i sh in your head restrict your freedom.
Wh e n you say something, you say it in Engl i sh; and
even though you may sometimes be frustrated by what
cannot be said, still, when you speak you have no wish to
be free of the rules. I n fact, a vast part of what you know
is captured in the fabric of these rul esevery concept
which you understood because you can express it in terms
of other concepts is part of the Engl i sh in your mind.
The rules of English make you creative because they
save you from having to bother with meaningless com-
binations of words.
Most possible combinations of words are mere jumbles
206
OUR P A T T E R N L A NGUA GE S I C ONT I NUE D
( " c a t wo r k house tea i s, " and so o n) . T h e r e are f ar
more of these nonsensi cal combi nat i ons than of the com-
binations whi c h make sense.
Suppose you had to search in your mi nd, a mo ng all the
possible combi nat i ons of wor ds every time you want e d to
say s o me t hi ng y o u woul d never even get to the t hi ngs
you wa nt to say: and you cert ai nl y woul d be unabl e to
say anyt hi ng that expressed deep f eel i ng or meani ng.
T h e rul es of Engl i s h steer you a wa y f r om the vast
number of nonsensi cal sentences, and t owar ds the smal l er
t h o u g h still v a s t n u mb e r of sent ences whi ch make
sense; so that you can pour all your ef f or t i nto the f i ner
shades of meani ng. I f it we r e not f or the rul es of Engl i s h,
you woul d spend all your time s t r uggl i ng to say anyt hi ng
at all.
A pattern language does the same.
A pat t ern l anguage is real l y not hi ng mor e t han a precise
wa y of descri bi ng someone' s experi ence of bui l di ng. I f a
ma n has a gr eat deal of experi ence of bui l di ng houses, his
l anguage f or houses is ri ch and c ompl e x; if he is a gr een-
horn, his l anguage is nai ve and si mpl e. A poet of houses,
a mast er bui l der, coul dn' t possibly wor k wi t hout his
l a n g u a g e i t woul d be as if he we r e a gr eenhor n.
Ag a i n, if you t hi nk of al l the possible combi nat i ons of
col umns, and studs, and wal l s, and wi ndows , most of
t hem are meani ngl ess j umbl es. T h e number of meani ng-
less combi nat i ons is vastl y l ar ger t han the number of
combi nat i ons whi c h make sense as bui l di ngs. A ma n wi t h-
207
T HE GATE
out a l anguage woul d have to comb his mind to find even
one meani ngf ul design among all these meaningless com-
binations, and he woul d never even get to the subtleties
which make a building work.
So the use of language is not merely something that
haffens in traditional societies. It is a fundamental
fact about our human nature, as fundamental as the
fact of sfeech.
Ever y creative act relies on l anguage. It is not only
those creative acts which are part of a traditional society
which rely on l anguage: all creative acts rely on pattern
l anguages: the f umbl i ng inexperienced constructions of a
novice are made within the l anguage which he has. T h e
works of idiosyncratic genius are also created within some
part of l anguage too. And the most ordinary roads and
bridges are all built within a l anguage too.
And now at last it becomes clear just where the fat-
terns in the world come from.
In chapter 5, we saw that every part of the worl d is
given its character, essentially, by a small number of
patterns whi ch repeat themselves over and over again.
Patterns whi ch repeat to create the floorboards in the
floor; patterns whi ch repeat to create the rooftops of a
t own; patterns whi ch create the overall arrangement of
the t own whi ch gives one place the character of Paris,
and another place the character of London . . .
Whe r e does all this repetition come from? Whe r e does
208
OUR PATTERN LANGUAGES I CONTI NUED
the order come f rom? Whe r e does the coherence come
f rom? Whe r e , above all, do the patterns come f rom, and
why are just a f ew of them repeated over and over
again ?
We now know the answer to this question.
The patternsy which repeat themselvesy come simply
from the fact that all the people have a common lan-
guage y and that each one of them uses this common
language when he makes a thing.
Each person has his own version of this common
l anguage, no doubt; but, broadly speaking, each person
knows the same patterns, and the same patterns therefore
keep repeating and repeating and repeating, al ways with
infinite variety, simply because these are the patterns in
the l anguage whi ch people use.
Every single part of the environment is governed by
some portion of a pattern language.
The r e are l anguages for the l ayout of fields, for the
arrangement of the streets, for public squares, for build-
i ng public buildings, churches, temples, l anguages for
l ayi ng out the way that buildings group, for mendi ng
walls, f or maki ng stairs, for the arrangement of the shops
and cafes al ong the street, and for the way the inside of
the shops are goi ng to be made and used. . . .
And the enormous repetition of patternsy which makes
up the worldy comes about because the languages
209
T HE GATE
which feofle use to make the world are widely
shared.
T h e millionfold repetition of patterns comes into the
worl d because a million people share the l anguages which
have these patterns in them.
Al l the places that man has ever built, traditional or
newl y invented, built a thousand years ago or built today,
designed by architects or l aymen, under the influence of
l aws or not, by many people or by one person, all of
them get their shape directly f rom the l anguages their
builders use.
At all timesy in every human culture, the entities of
which the world is made are always governed by the
fattern languages which feofle use.
Every windowy every doory each roomy each housey
each gardeny every streety each neighborhoody and
every town: it always gets its shafe directly from
these languages.
They are the origin of all the structure in the man-
made world.
21 0
CHAPTER I 2
THE CREATIVE POWER
OF LANGUAGE
Andy beyond that> it is not just the shape
of towns and
f
buildings which comes to
them from pattern languages
}
it is
their quality as well. Even the life and
beauty of the most awe-inspiring great
religious buildings came from the lan-
guages their builders used.
From chafler 11, we see that pattern languages are
responsible for all the ordinary structure of the world.
But 'pattern languages are still more basic, even than
that. It is not just the form of buildings, but their life
as well, their beauty as created things, which comes
from -pattern languages. The patterns are responsible
not only for the specific shape a building has, but also
for the extent to which the building comes to life.
T h e life and beauty of the great cathedrals comes from
their pattern l anguages. So does the beauty of the smallest
place which comes to life. And the degree to which a
building comes to life, and moves us, al ways hinges on the
power of the pattern l anguage which its builders used.
Let us start by seeing how the great cathedrals,
Chartres and Notre Dame, were made within a pat-
tern language too.
In one sense, this is obvious. Of course, the rules which
formed the great cathedrals were, to some extent, com-
mon rules of thumb, whi ch defined the general form of
' o
" a " cathedral. Nave, aisles, transepts, east end, west end,
tower . . . .
And it was not only the obvious l arge scale organization
which was composed of common patterns. At a smaller
scale, there were patterns too. T h e clustering of columns,
the form of the arch, the great rose wi ndow in the west,
the chapels round the east end, the spacing of the columns,
the buttresses and flying buttresses.
21 5
T HE GATE
Indeed, even the most beautiful details were patterns too.
Th e col umn capitals, the wi ndow tracery, the way the
stones within the vaults were cut, the hammer beam
roof, the gargoyl es on the flying buttresses, the carvings
round the doorways, stained glass in the wi ndows, polished
stones whi ch made the floor, the carved and inlaid
tombs . . . .
Of course, these buildings were not built by lay
feofle.
The r e were hundreds of people, each maki ng his part
within the whol e, worki ng, often for generations. At any
given moment there was usually one master builder, who
directed the overall layout . . . but each person in the
whol e had, in his mi nd, the same overall l anguage. Each
person executed eacJi detail in the same general way, but
with minor differences. T h e master builder did not need
to force the design of the details down the builders'
throats, because the builders themselves knew enough of
the shared pattern l anguage to make the details correctl y,
with their own individual flair.
But still the fower and beauty of the great cathedrals
came mainly from the language which the master
builder and his builders shared.
Th e l anguage was so coherent that anyone who under-
stood this l anguage wel l and devoted his whol e life to
the building of a single building, worki ng at it sl owl y,
21 6
T HE CREATI VE POWER OF LANGUAGE
piecemeal, shaping all the parts within the common l an-
guage, woul d be able to make a great work of art.
Th e building grew sl owl y, magni fi centl y, from the
impact of the common pattern l anguage from whi ch it
was made, guiding its individual parts, and the acts which
created them, just as the genes inside the f l ower' s seed
guide and then generate the f l ower. . . .
All the great buildings in history have been built
like this, by languages.
Chartres, T h e Al hambra, the mosque at Kai rouan,
Japanese houses, Brunelleschi' s dome . . .
We imagine, because of the distorted view of archi-
tecture we have learnt, that some great architect created
these buildings, wi th a f ew marks of the pencil, worked
out laboriously at the dr awi ng board.
T h e fact is that Chartres, no less than the simple f arm-
house, was built by a group of men, acting within a
common pattern l anguage, deeply steeped in it of course.
It was not made by "desi gn" at the drawi ng board.
The same process which the simple farmer used to
make his house, the same process exactlyy was the
process which allowed people to generate these
greater buildings.
T h e builders who built the great cathedrals, the great
mosques, the palaces, and the Al hambra, used the same
l anguage as ordinary people.
21 7
T HE GATE
T h e people had a passing knowl edge of their l anguage,
they built no more than a house or two, and helped to
build a public bui l di ngt hey were occupied essentially
with something else.
But the builders were those men who spent their whol e
lives with that same l anguage, deepening it, under-
standing more about its patterns, practicing, building
over and again, until they knew exactly how to realize
these patterns best.
You may have a fundamental doubt about the possi-
bility of capturing the deepest architectural knowl-
edge in any "language."
It is, after all, common to say that a great creator has
talent whi ch lesser persons do not have, and to assume
that the power to create a wonderf ul building whi ch is
full of life, depends simply on this talent.
However , many people will agree that a great archi-
tect's creative power, his capacity to make something beau-
tiful, lies in his capacity to observe correctly, and deeply.
A painter' s talent lies in his capacity to s eehe sees more
acutely, more precisely, what it is that really matters in a
thing, and where its qualities come from. And an archi-
tect's power also comes from his capacity to observe the
relationships whi ch really mat t er t he ones which are
deep, prof ound, the ones which do the work.
I n this sense, then, a pattern l anguage whi ch is deep
is a collection of patterns whi ch correspond to profound
observations about what makes a building beautiful.
21 8
T HE CREATI VE POWER OF LANGUAGE
We have a habit of thinking that the dee-pest insights,
the most mystical, and spiritual insights, are some-
how less ordinary than most thingsthat they are
extraordinary.
Thi s is only the shallow refuge of the person who does
not yet know what he is doing.
In fact, the opposite is true: the most mystical, most
religiousy most wonderfulthese are not less ordinary
than most thingsthey are more ordinary than most
things.
It is because they are so ordinary, indeed, that they strike
to the core.
And this is connected to the fact that these things can,
indeed, be expressed clearly, discovered, talked about.
These deep things which really matter, they are not
f ragi l et hey are so solid that they can be talked about,
expressed quite clearly. Wh a t makes them hard to find is
not that they are unusual, strange, hard to expressbut
on the contrary that they are so ordinary, so utterly
basic in the ordinary bread and butter senset hat we
never think of l ooki ng for them. Le t me give t wo ex-
amples: one f rom the beauty of old prayer rugs, the
other from the art of building.
The old Turkish prayer rugs, made two hundred
years ago, have the most wonderful colors.
21 9
T HE GATE
Al l of the good ones fol l ow this rul e: wherever there are
two areas of color, side by side, there is a hairline of a
di fferent third color, between them. Thi s rule is so
simple to state. And yet the rugs which fol l ow this rule
have a brilliancc, a dance of color. And the ones which
do not follow it are somehow flat.
Of course this is not the only rule which makes a rug
gr eat but this one rule, simple, banal, almost as it seems,
will triple the brilliance and the beauty of a rug. A per-
son who knows this rule may be able to make a beautiful
rug. A person who does not will almost certainly not be
able to.
And all the other features of the great old rugs also
depend on other rather simple rules. But now most of
these rules have been f or got t enand now they can no
longer make the great rugs, with their glorious colors.
T h e depth, and spirituality, of the rug is not made
less by the fact that this rule can be expressed, nor that
it is so simple. Wh a t matters, simply, is that this rule is
extremel y deep, extremel y powerf ul .
And the light in many glorious rooms is also gov-
erned by a simple rule.
Consider the simple rule that every room must have day-
light on at least t wo sides (unless the room is less than
8 feet deep). Thi s has the same character, exactly, as the
rule about the colors. Rooms whi ch fol l ow this rule are
pleasant to be in; rooms whi ch do not fol l ow it, with a
f ew exceptions, are unpleasant to be in.
220
T HE CREATI VE POWER OF LANGUAGE
Or consider one of the most beautiful small buildings
in the world: the shrine at Ise, in Japan.
Wh a t is it that makes it beautiful? It is the steepness of
the roof, the way the roof beams cut the sky, the wal k
around the building, the height of the railing, the per-
fectly smoothed and rounded wooden columns, the brass
covers, whi ch protect the open grain at the end of every
beam, the brass bolts let into the smoothly polished planks,
the spacing of the col umns in the walls, the fact that there
are col umns at the corners, marki ng space, the gravel
path which leads around the building, the position of the
steps, which form the entrance, and provide a place to
stop . . . .
Again it is the particular patterns there, and the
repetition of the patterns, which creates the magic of
the building.
For each of., these facts about the building is not just a
chance event. It is a rule which is repeated over and
agai n; it is fol l owed exactl y, and the building varies only
in the way permitted by these rules; the rules adapt to
di fferent places in the building and create a slightly
di fferent version there; but it is above all their repetition,
over and again, and the fact that there is almost nothing
else whi ch makes the building come to life, and sit, in-
spiring, and inviting for us. . . .
You may wonderif the rules are so simple to
expresswhat is there that makes a builder greatP
221
T HE GATE
And indeed, there is an answer. Ev e n though the rules
are simple, by the time you have twenty, perhaps fi fty
rules like this in your mind, it takes almost i nhuman
singleness of purpose to insist on t he mnot to let go of
them.
It is so easy to s ay oh wel l , it is too hard to have light
on two sides of this room, and that r ooma t the same
time as all other things we are tryi ng to do. It will be
alright if we allow this room to have light on just one
side. T h e fact is that it will not be alright. But to insist,
to keep all the rules whi ch matter, freely in your mind,
and not to let go of t hemt hat does perhaps require
unusual character of purpose.
But of course, the fact that these rules are simple does
not mean that they are easy to observe, or easy to
invent.
Just as a great artist is one who observes very carefully
the things whi ch make the di f f erenceso it does, indeed,
take great powers of observat i ongreat depth, great
concentration, to formul ate these simple rules.
A man who knows how to build has observed hundreds
of rooms, and has finally understood the "secret " of
maki ng a room with beautiful proportions say. . . . Thi s
knowl edge exists, in his mind, in the f orm of a rudi-
mentary pattern, whi ch tells him, under such and such
circumstances, create the f ol l owi ng field of relation-
ships . . . for such and such reasons. It may have taken
years of observation for him finally to understand this
rule.
222
T HE CREATI VE POWER OF LANGUAGE
It may be hard to believe that one might make a
work of art by simply combining patterns.
It sounds almost as though there was a box of " magi c "
parts, so powerf ul , that anyone can make a beautiful
thing, simply by combining them.
Thi s is absurd, because, of course, it is not possible to
make something beautiful, merel y by combining fixed
components.
But once again, the difficulty of believing it may have
to do with the fact that we tend to think of patterns
as "thingsand keep forgetting that they are com-
plex\ and potent fields.
Each pattern is a fieldnot fixed, but a bundle of rela-
tionships, capable of being di fferent each time that it
occurs, yet deep enough to bestow life wherever it occurs.
A collection of these deep patterns, each one a fluid
field, capable of being combined, and overlapping in en-
tirely unpredictable ways, and capable of generati ng an
entirely unpredictable system of new and unforeseen
relationships.
Wh e n we remember this, it may be easier to recognize
how powerf ul they ar eand that we do indeed, have our
creative power as a result of the system of patterns whi ch
we have.
The source of life which you create lies in the power
of the language which you have.
223
T HE GATE
If your l anguage is empty, your buildings cannot be full.
If your l anguage is poor, you cannot make good buildings
until you enrich your l anguage. If your l anguage is rigid,
your buildings must be rigid. If your l anguage is florid,
your buildings will be florid. Your l anguage generates the
buildings which you make, and the buildings live or not,
according to the life your l anguage has.
Pattern l anguages are the source of beauty and of
ugliness. The y are the source of all creative power:
nothing is made wi thout a pattern l anguage in the
maker' s mi nd; and what that thing becomes, its depth, or
its banality, comes also f rom the pattern l anguage in the
builder' s mind.
And now we realize the truly immense -power which
pattern languages have.
For it is not only true that every building gets its
structure from the languages which people use.
It is also true that the spirit which the buildings
have, their power, their life, comes from the pattern
languages their builders use as well. The beauty of
the great cathedrals, the fire in the windows, the
touching grace of ornaments, the carving of the
columns and the column capitals, the great silence
of the empty space which forms the heart of the
cathedral . . . all these come from the pattern lan-
guages their builders use as well.
224
CHAPTER I 3
THE BREAKDOWN OF LANGUAGE
But in our time the languages have
broken down. Since they are no longer
shared
y
the processes which keep them
deep have broken down: and it is there-
fore virtually impossible for anybody
y
in our time
y
to make a building live.
i i i i i i i i i i i i i i i i i i i i gi l l Ml l l l l l l l l
iiiIiKiiliiiilliiii|H|iii|iiliiiil|
iiiiiiiLiiiiiiiiiiiiaiii L
iiiiaiiKilIiilIiBlliiii|i2|||i|i||l#
iiiiiliiHiiilliiiiiiiBiliHiiiiiiiii
iiiiiiiliiiiiiiiiBiiBiiligiiiliiiili
iiiillliiiiiilli!iii|i||||!l!!!l!liii
jiiiiiiiiijiiaiiii9liiiiaiaai>i
iiiiiiiiiiBiililliiiaiiiiiiiiiii
i l l i i HB B i B B I i B i i i B i l i i i U B Mi MI i l B
H I I B B H I i i i i i l l i l l i l l i l l i l l i l l l B l H i
We know now that language has the -power to bring
things to life. The most beautiful houses and vil-
lagesthe most touching paths and valleysthe most
awe inspiring mosques and churchesattained the life
they have in them because the languages their builders
used were powerful and deep.
But, so far, we have not dealt at all with the condi-
tions under which a language is itself alive, or the
conditions under which a language dies.
For all the ugliest and most deadening places in
the world are made from patterns as well.
Consi der, for example, the l anguage whi ch generated
my office at school.
It is an ugl y place, terrible, dark and dead. It is one
of many similar offices, in the same building: and these
offices are generated by the f ol l owi ng l anguage :
LONG AND NARROW
DAYLIGHT AT ONE END ONLY
WINDOW THE FULL WIDTH OF THE WALL
CONCRETE WAFFLE CEILING, 5' GRID
FLUORESCENT LIGHTS AT I O' CENTERS
FLAT CONCRETE WALL
UNPAINTED CONCRETE CEILING SURFACE
STEEL WINDOW
PLYWOOD WALL SURFACE
Thi s terrible l anguage has generated hundreds of
offices. But the person who has this l anguage in his
mind can never make an office live until he abandons this
l anguage altogether. The r e is not a single pattern in this
228
T HE BREAKDOWN OF LANGUAGE
list, except perhaps the fourth, whi ch is not derelict and
at odds with the forces actually at work in such a context.
It is therejore obvious that the mere use of pattern
languages alone does not ensure that people can
make places live.
Some towns and buildings live, and others don' t. If all
of them are made by pattern l anguages, there must be
some distinction in the content of these l anguages, and in
the way that they are used.
Andy indeed} there is a fundamental difference be-
tween those societies in which people are able to
make their environment alivey and those in which the
towns and buildings become dead.
Pattern l anguages are used in both. But the pattern
l anguages in the t wo kinds of societies are di fferent. I n
one case, the pattern l anguages themselves are somehow
alive and help people give life to their surroundings. I n
the other case, the l anguages themselves are dead: and
with these l anguages it is only possible for people to make
towns and buildings whi ch are dead.
In a town with a living languagey the pattern lan-
guage is so widely shared that everyone can use it.
In agricultural societies everyone knows how to build;
everyone builds for himself, and helps his neighbor build.
229
T HE GATE
And in later traditional societies there are bricklayers,
carpenters, pl umbersbut everyone still knows how to
design. For exampl e, in Japan, even fifty years ago, every
child learned how to lay out a house, just as children
learn football or tennis today. People laid out their houses
for themselves, and then asked the local carpenter to build
it for them.
Wh e n the l anguage is shared, the individual patterns
in the l anguage are profound. T h e patterns are al ways
simple. Not hi ng whi ch is not simple and direct can survive
the slow transmission from person to person. The r e is
nothing in these l anguages so compl ex that someone can-
not understand it.
Cornerstones for a stone building; a shelf by the
wi ndow; a seat by the front door; dormer wi ndows;
care for a tree; light and shade where we sit; runni ng
water in the nei ghborhood; a brick edge to the
water . . .
Just because every detail has to make sense to every
man and woman, the patterns are heartfelt, and pro-
found.
The language covers the whole of life.
Ever y facet of human experience is covered, in one way
or another, by the patterns in the l anguage.
T h e seven ages of man are all covered, and the variety
of all possible acts is covered. T h e entire culture, and
the envi ronment whi ch supports it, forms a single un-
broken fabric.
230
T HE BREAKDOWN OF LANGUAGE
The connection between the users and the act of
building is direct.
Ei ther the people build for themselves, with their own
hands, or else they talk directly to the craftsmen who
build for them, with almost the same degree of control
over the small details which are built.
T h e whol e emerges by itself and is continually re-
paired. Each person in a t own knows that his own
small acts help to create and to maintain the whol e. Each
person feels tied into society, and proud because of it.
The adaptation between people and buildings is pro-
found.
Each detail has meani ng. Each detail is understood. Each
detail is based on some person' s experience, and gets
shaped right, because it is sl owl y thought out, and deeply
felt.
Because the adaptation is detailed and profound, each
place takes on a unique character. Sl owl y, the variety of
places and buildings begins to reflect the variety of
human situations in the town. Thi s is what makes the
t own alive. T h e patterns stay alive, because the people
who are using them are also testing them.
But, by contrast, in the early phases of industrial so-
ciety which we have experienced recently, the pattern
languages die.
Instead of being widely shared, the pattern languages
231
T HE GATE
which determine how a t own gets made become special-
ized and private. Roads are built by hi ghway engineers;
buildings by architects; parks by planners; hospitals by
hospital consultants; schools by educational specialists;
gardens by gardeners; tract housing by developers.
T h e people of the t own themselves know hardly any
of the l anguages whi ch these specialists use. And if they
want to find out what these l anguages contain, they
can' t, because it is considered professional expertise. T h e
professionals guard their language jealously to make
themselves indispensable.
Even within any one profession, professional jealousy
keeps people from sharing their pattern languages. Archi -
tects, like chefs, jealously guard their recipes, so that they
can maintain a unique style to sell.
Th e l anguages start out by being specialized, and
hidden from the people; and then within the specialities,
the l anguages become more private still, and hidden from
one another, and f ragmented.
Most people believe themselves incompetent to design
anything and believe that it can only be done properly
by architects and planners.
Thi s has gone so far that most people shrink, in fear,
from the task of designing their surroundings. The y are
afraid that they will make foolish mistakes, afraid that
people will laugh at them, afraid that they will do some-
thing "i n bad taste. " And the fear is justified. Once
people wi t hdraw f rom the normal everyday experience of
232
T HE BREAKDOWN OF LANGUAGE
building, and lose their pattern l anguages, they are liter-
ally no l onger able to make good decisions about their
surroundings, because they no l onger know what really
matters, and what doesn' t.
Peofle lose touch with their most elementary intui-
tions.
If they read somewhere that large plate glass picture
wi ndows are a good idea, they accept this as wisdom from
a source wiser than t hemsel veseven though they feel
more comfortable sitting in a room with small wi ndow-
panes, and say how much they like it. But the fashionable
taste of architects is so persuasive that people will be-
lieve, against the evidence of their own inner feelings,
that the plate glass wi ndow is better. The y have lost con-
fidence in their own j udgment . Th e y have handed over
the right to design, and lost their own pattern l anguages
so utterly that they will do anythi ng which architects tell
them.
Yet , architects themselves, have lost their intuitions too.
Since they no l onger have a widely shared l anguage which
roots them in the ordinary feelings people have, they are
also prisoners of the absurd and special l anguages which
they have made in private.
Even the buildings built by architects start to be full
of obvious "mistakes."
2
33
T HE GATE
T h e recently built Col l ege of Envi ronment al Desi gn at
the University of Cal i forni a, Berkel ey, was designed by
three wel l - known architects. In a certain part of this
building, at the end of each floor, there are t wo seminar
rooms. Thes e seminar rooms are l ong and nar r ow; one
of the short walls is filled by a wi ndow; the blackboard
is mounted al ong one of the l ong wal l s; each room is
filled by a l ong narrow table. These rooms are f unc-
tionally defective in a number of obvious ways. First of
all, a l ong narrow table, and the l ong narrow group of
people whi ch form around it, are not suitable for intense
discussion; this is a seminar r oomi t should be more
nearly square. Second, the position of the blackboard with
respect to the wi ndow means that half of the people in
the room see the wi ndow reflected on the blackboard, and
can' t read what is written t heret he blackboard should
be opposite the wi ndow. Thi r d, because the wi ndow is so
large, and so l ow, the people who sit near it appear
silhouetted to those who are sitting further away. It is ex-
tremely difficult to talk properly with someone seen in
si l houettetoo many of the subtle expressions of the face
get lost. Seminar communication suffers. T h e wi ndow
should be above the height of a sitting person's head.
Specific patterns, like, for instance, the LIGHT ON TWO
SIDES pattern, vanish from people's knowledge about
building.
At one time it would have been unthinkable to build any
234
T HE BREAKDOWN OF LANGUAGE
room, except a stable or a workshed, without wi ndows on
t wo sides. In our own time, all knowl edge of this pattern
is forgotten. Most rooms, in most buildings, have light
from one side only. And even a " gr eat " architect like
Le Corbusier, builds whol e apartments, l ong and narrow,
with wi ndows only at the narrow e ndas he did in the
Marseilles apartments bl ockwi t h terrible glare and dis-
comf ort as results.
The r e is not a single building built in recent times, nor
a single part of a city laid out by planners, in whi ch such
trivial mi stakescaused by the loss of pat t ernscannot
be described a hundredfol d. Thi s is as true of the greatest
works of so-called modern masters, as of the most mun-
dane works built by tract developers.
And those few patterns which do remain within our
languages becomes degenerate and stupid.
Thi s fol l ows naturally from the fact that the l anguages
are so highly specialized. T h e users, whose direct ex-
perience once formed the languages, no l onger have
enough contact to influence them. Thi s is almost bound
to happen, as soon as the task of building passes out of the
hands of the people who are most directly concerned, and
into the hands of people who are not doing it for them-
selves, but instead for others.
So l ong as I build for myself, the patterns I use will be
simple, and human, and full of feeling, because I under-
stand my situation. But as soon as a f ew people begin to
build for "t he many, " their patterns about what is needed
become abstract; no matter how wel l meani ng they are,
235
T HE GATE
their ideas gradual l y get out of touch with reality, because
they are not faced daily with the living examples of what
the patterns say.
If I build a fireplace for mysel f, it is natural for me to
make a place to put the wood, a corner to sit in, a mantel
wide enough to put things on, an opening which lets the
fire draw.
But, if I design fireplaces for other peopl enot for
mysel f t hen I never have to build a fire in the fireplaces
I design. Gradual l y my ideas become more and more in-
fluenced by style, and shape, and crazy not i onsmy feel-
ing for the simple business of maki ng fire leaves the
fireplace altogether.
So, it is inevitable that as the work of building passes
into the hands of specialists, the patterns whi ch they use
become more and more banal, more wi l l ful , and less an-
chored in reality.
Of course, even now a town still gets its shape from
pattern languages of a sort.
T h e architects and planners and bankers have pattern l an-
guages whi ch tell them to build gigantic steel and concrete
buildings. Th e users have a f ew shattered patterns left in
their vocabul ary: a sheet of plastic to make a kitchen
counter; huge plate glass wi ndows for the living room;
wal l -t o-wal l carpet in the bat hr oomand they enthusi-
astically piece these patches together, whenever they have
a free weekend.
236
T HE BREAKDOWN OF LANGUAGE
But these remnants of our former languages are dead
and empty.
Th e y are based mainly on the by-products of industry.
People use plate glass wi ndows, formica counters, wal l -to-
wal l carpet, because industry makes them available, not
because these patterns contain anythi ng essential about life,
or how to live it.
Th e time when a pattern l anguage was a song, in which
people could sing the whol e of life, is gone. T h e pattern
l anguages in society are dead. Th e y are ashes and f rag-
ments in the people's hands.
As the pattern languages die, everyone can see the
chaos which emerges in our towns and buildings.
But the people do not know that it is the pattern lan-
guages whi ch cause it. Th e y know that buildings are
less human than they used to be. Th e y are wil l ing to
pay great prices for old buildings whi ch were made at a
time when people still knew how to make them human.
Th e y complain bitterly about the lack of life, the danger,
the merciless inhuman quality of their envi ronment. But
they do not know what to do about it.
In panic, people try to replace the lost order of the or-
ganic process, by artificial forms of order based on
control.
237
T HE GATE
Since the natural processes of building towns no l onger
work, in panic, people look for ways of "cont rol l i ng" the
design of towns and buildings. Those architects and
planners who have become concerned by the insignifi-
cance of their influence on the envi ronment make three
kinds of efforts to gain "total desi gn" control of the
envi ronment :
1. Th e y try to control l arger pieces of the environ-
ment (this is called urban desi gn).
2. Th e y try to control more pieces of the envi ronment
(this is called mass production or system-bui l di ng).
3. Th e y try to control the envi ronment more f i rml y,
by passing l aws (this is called planning cont rol ).
But this makes things still worse.
These totalitarian efforts, although they do control
more of the envi ronment, have exactly the wr ong effect.
Th e y cannot create a whol e envi ronment, because they
are not sufficiently responsive to the real needs, forces,
demands, problems, of the people invol ved. Instead of
maki ng the envi ronment more whol e, they make it less
whol e.
At this stage, the pattern l anguages become still more
f ragmented, and more dead. Th e y are controlled by even
f ewer people; they have even less of the living connection
with the people whi ch they need.
The variety, once created by organic and natural pro-
cesses, disappears altogether.
238
T HE BREAKDOWN OF LANGUAGE
Experts try to make towns and buildings whi ch are
adapted to people's needs, but they are al ways trivial. The y
can only deal with general forces, whi ch are common to
all men, and never with the particular forces that make
one particular man unique and human.
Adaptation of buildings to people becomes impossible.
Even when experts make buildings whi ch are "adapt-
abl e, " to solve this problem, the result is still trivial, be-
cause the unique particulars are still subservient to the
common generalities. Huge machine-like buildings whi ch
al l ow people to move the walls around, so that they can
express themselves, still make them subject to the "syst em. "
Andy finallyy people lose their ability to make life al-
together.
Onc e the common l anguage has been broken down, the
individual l anguages whi ch, in a living culture, are al ways
private versions of the common l anguage, are also broken
down.
And not only that. It even becomes impossible for peo-
ple to create, or re-create, new private languages, because
the absence of a common l anguage means that they lack
the core of fundamental stuff they need to f orm a living
l anguage for themselves.
At this stage, people can no l onger even make a wi ndow
or a door whi ch is beautiful.
239
T HE GATE
It must be obvious from all of this, that a town can-
not become alive without a living language in it.
It is impossible, utterly impossible, to make a building or
a t own whi ch is alive by control from above. And it is
impossible for the people to make the town for themselves
with the ashes of the dead l anguage whi ch they now
have.
The fact is, that the creation of a town, and the
creation of the individual buildings in a town, is
fundamentally a genetic process.
No amount of planning or design can replace this genetic
process.
And no amount of personal genius can replace it either.
Our emphasis on objects, has blinded us to the essential
fact that it is above all the genetic process whi ch creates
our buildings and our towns, that it is above all this ge-
netic process which must be in good order . . . and that
this genetic process can only be in good order, when the
l anguage whi ch controls it, is wi del y used, and widely
shared.
People need a living l anguage, in order to make buildings
for themselves. But the l anguage needs the people too . . .
so that its constant use, and feedback, keeps its patterns in
good order.
And this conclusion, simple though it is, calls for a
shattering revision of our attitude to architecture and
planning.
240
T HE BREAKDOWN OF LANGUAGE
In the past, eacli act of planning or design was thought of
as a self-contained, and original, response to the demands
of a local situation. Archi tects and planners assumed, im-
plicitly, that the structure of the town is generated by
the accumulation of these self-contained acts.
Th e picture which our arguments have led us to is
radically di fferent. Accordi ng to this vi ew, there are
underl ying l anguages whi ch already contain most of the
structure that will appear in the envi ronment. T h e acts
of design which have been thought of as central are acts
which use the structure already present in these under-
lying languages to generate the structure of specific
buildings.
In this vi ew, it is the structure of the underl ying lan-
guage which is doing most of the hard work. If you
want to influence the structure of your t own, you must
help to change the underl ying l anguages. It is useless to
be innovative in an individual building, or an individual
plan, if this innovation does not become part of a living-
pattern l anguage which everyone can use.
And we may conclude, even more strongly, that the
central task of "archi tecture" is the creation of a single,
shared, evol vi ng, pattern l anguage, whi ch everyone con-
tributes to, and everyone can use.
So long as the feofle of society are separated, from
the language which is being used to shape their build-
ings, the buildings cannot be alive.
If we want a l anguage which is deep and powerf ul , we
can only have it under conditions where thousands of peo-
241
T HE GATE
pie are using the same l anguage, exploring it, maki ng it
deeper all the time.
And this can only happen when the l anguages are
shared.
In the next four chapters, we shall see how it is pos-
sible to share our language, and to make it living once
242
CHAPTER 14
PATTERNS WHI CH CAN BE SHARED
To work our way toward a shared and
living language once again, we must first
learn how to discover patterns which are
deep, and capable of generating life.
If we hope to bring our towns and buildings back to
life, we must begin to re-create our languages, in such
a way that all of us can use them: with the patterns
in them so intense, so full of life again, that what we
make within these languages will, almost of its own
accord, begin to sing.
To start with this requires simply that we find a
way of talking about patterns, in a way that can be
shared.
How can this be done? I n a traditional culture, these
patterns exist as independent entities within your mind,
but it is not necessary for you to recognize them as
separate atomic units, nor to know them by name, nor to
be able to speak about them. It is no more necessary than
it is for you to be able to describe the rules of gr ammar
in the l anguage whi ch you speak.
However , in a period when l anguages are no longer
widely shared, when people have been robbed of their in-
tuitions by specialists, when they no l onger even know the
simplest patterns that were once implicit in their habits, it
becomes necessary to make patterns explicit, precisely and
scientifically, so that they can be shared in a new w a y
explicitly, instead of i mpl i ci tl yand discussed in public.
In order to make patterns explicit, so that they can
be shared in this new way, we must first of all review
the very complex structure of a pattern.
Thr oughout this book we have had a gradual awakeni ng,
246
PATTERNS WHI C H CAN BE SHARED
a gr owi ng understanding of what a pattern is. Thi s
awakeni ng began, in chapters 4 and 5, where the concept
was first def i ned; the concept was then extended and re-
fined, in chapter 6 and then again in chapters 10, I I
and 12.
I shall now describe the structure of a single pattern
precisely, in a way that includes all the properties whi ch
living patterns have to have, as they have been discussed
in all these chapters.
Each pattern is a three-part rule, which expresses a
relation between a certain context, a problem, and a
solution.
As an el ement in the worl d, each pattern is a relationship
between a certain context, a certain system of forces
whi ch occurs repeatedly in that context, and a certain
spatial configuration whi ch allows these forces to resolve
themselves.
As an el ement of l anguage, a pattern is an instruction,
whi ch shows how this spatial configuration can be used,
over and over again, to resolve the given system of forces,
wherever the context makes it relevant.
Th e pattern is, in short, at the same time a thing,
whi ch happens in the worl d, and the rule whi ch tells us
how to create that thing, and when we must create it. It
is both a process and a thi ng; both a description of a
thing whi ch is alive, and a description of the process
whi ch will generate that thing.
Patterns can exist at all scales.
247
T HE GATE
Patterns can be stated equally wel l for the human details
of buildings, the overall layout of a building, ecology,
large-scale social aspects of urban planning, regional
economics, structural engineering, details of building
construction.
For example, the distribution of subcultures in a region,
the layout of maj or roads, the organization of work
groups in an industry, the arrangement of trees at the
edge of a forest, the design of a wi ndow, the planting of
flowers in a garden, the layout of a sitting room, mi ght
all be specified by patterns.
And a pattern may deal with almost any kind of
forces. (All the following patterns are defined in
Volume 2 of this series).
ENTRANCE TRANSITION resolves a conflict among inner
psychic forces.
MOSAIC OF SUBCULTURES resolves a conflict among
social and psychological forces.
WEB OF SHOPPING resolves a conflict among economic
forces.
EFFICIENT STRUCTURE resolves a conflict among struc-
tural forces.
GARDEN GROWING WILD resolves the conflict between
forces of nature, the natural gr owi ng process in plants,
and people's natural actions in a garden.
WEB OF TRANSPORTATION resolves forces which lie
partly in the field of human need, and partly in the
politics of public agencies.
STILL WATER resolves conflicts among forces whi ch are
248
PATTERNS WHI C H CAN BE SHARED
partly ecological, and partly in the realm of human fear
and danger.
COLUMNS AT THE CORNERS resolves conflicts among
forces which arise within the process of construction.
WINDOW PLACE resolves forces which are purely
psychological.
To make a -pattern explicit, we merely have to make
the inner structure of the pattern clear.
Le t us start with a very simple commonsense example.
Suppose that we are in a place. We have a general sense
that something is " r i ght " there; something is wor ki ng;
something feels good; and we want to identify this
"somet hi ng" concretely so that we can share it with
someone else, and use it over and over again.
Wh a t do we have to do? As we shall now see, there
are al ways three essential things we must identify.
What, exactly, is this something?
Why, exactly, is this something helping to make the
place alive?
And when, or where, exactly, will this pattern work?
We must first define some physical feature of the
place, which seems worth abstracting.
Ta ke , for the sake of an example, Ost enf el dgaar dena
beautiful old Danish house built in 1685, now in the
Copenhagen Open Ai r Museum. As soon as I went
there, I knew that it had special qualities whi ch woul d
be useful even today, if I could only pin them down. How
249
T H E GAT E
is it possible to pin t hem d o wn i n a wa y t hat is precise
enough to use over and over agai n?
Suppose f or the sake of ar gume nt , I start wi t h f eat ures
like " c oz i nes s " or "spaci ousness. " The s e f eat ures are
no doubt there. Bu t they are not di rectl y usabl e. Ev e n if
I try to pin down the idea of cozi ness f ur t her , by sayi ng
that the f or m of the house s omehow ties the f ami l y to-
get her , and that this is wha t makes it c oz y, this is still not
cl ear enough to copy in anot her house. Unt i l I have
i denti f i ed parti cul ar spatial relationships whi c h hel p to
create this qual i ty at Os t enf el dgaar den, I have still not
manage d to i dent i f y anyt hi ng whi c h I can use di rectl y in
anot her house.
Suppose, then, in an ef f or t to be mor e concret e, I seize
on a part i cul ar spatial rel at i on: the f act t hat there are
al coves r ound the edge of the mai n r oom, t hat the al -
2 5 0
PATTERNS WHI C H CAN BE SHARED
coves have seats in them, that each one is big enough to
hold one or t wo members of the family, and that they
both open into the common living room. It is complex,
but fairly wel l defined. It identifies certain parts (l i vi ng
room, alcoves, seats) and specifies a spatial relation be-
tween these parts.
Thi s pattern is wel l defined. If you were designing a
house, you could incorporate this pattern in the design
directly. Yo u could explain the idea to a third person and
he could decide, by looking at any house plan, whet her or
not it had this feature. So far so good. But even so this
pattern still woul dn' t be sharable. For it to be sharable,
we have to be able to criticize it. And to criticize it, we
have to know its functional purpose.
Next, we must define the problem, or the field of
forces which this pattern brings into balance.
Wh y is it a good idea? Wh a t is the problem whi ch is
solved by putting alcoves around a room? In answer to
this question, I mi ght propose something like: Li vi ng
rooms without alcoves don' t work, for the f ol l owi ng
reasons: the members of a family like to be together; but,
in the evenings and on weekends, when they could be,
each one follows up his personal hobbi essewi ng, home-
work . . . . Because these things are messy, and often
need to be l eft standing, people cannot do them in the
living r oomt he living room is a place which mustn' t get
too messy, since visitors mi ght come at any moment , and
it must be a suitable place to receive them. Instead the
various members of the family go off to their own private
251
T HE GATE
areas to do these t hi ngst he kitchen, the bedroom, the
basement and the family cannot be together.
Ther e are three forces at work here:
1. Each person in the family has his own private
hobbi essewi ng, carpentry, model-building, homework.
These activities being what they are, things often need to
be l eft l yi ng about. People therefore tend to db them
somewhere where things can be left lying safely.
2. Communal places in the house have to be kept tidy,
partly on account of visitors, and partly also so that no
one person or thing encroaches too heavily on the com-
fort and convenience of the whol e family.
3. T h e people in the family woul d like to be together
while they are doing these di fferent things.
In an ordinary house, with an ordinary living room,
these three forces are mutually incompatible. T h e alcove
brings them into resolution.
Finally, we must define the range of contexts where
this system of forces exists and where this -pattern of
physical relationships will indeed actually bring it
into balance.
By now the pattern is clear and sharable. But one ques-
tion is Still open. Whe r e exactly does this pattern make
sense? Does it make sense in an igloo? Hardl y. Does it
make sense in the living room of a cottage where a
single person lives alone? Obvi ousl y not. Wh e n exactly
does it make sense?
252
PATTERNS WHI C H CAN BE SHARED
T o make the pattern really useful, we must define
the exact range of contexts where the stated problem
occurs, and where this particular solution to the problem
is appropriate.
In this case, we should have to define the fact that
the pattern applies to the living rooms of all dwel l i ngs for
large families in the Uni ted States and West ern Europe
(perhaps in other cultures, t oodependi ng on specific
local habits and life styl es). Furt her, if a dwel l i ng has
more than one "l i vi ng r oom" in i t l i ke some Engl i sh
houses which have a front parlor and a back pa r l or
then the alcove idea woul d not apply to both these living
r oomsonl y to the one where the family spends most of
its time.
We see, in summary, that every pattern we define
must be formulated in the form of a rule which
establishes a relationship between a context, a system
of forces which arises in that context, and a configura-
tion which allows these forces to resolve themselves
in that context.
It has the f ol l owi ng generic form :
Cont ext > System of forces > Conf i gurat i on.
And, in the previous case, the f ol l owi ng specific content:
Communal > Confl i cts between > Al cove opening
rooms privacy and off communal
communi ty room.
253
T HE GATE
Every living pattern is a rule of just this kind.
And because it includes the context, each pattern is a self-
contained logical system which makes a double statement
of fact, not merely a declaration of value, and is therefore
capable of being true or not. It says, first, that the given
problem (the stated conflict among forces) exists within
the stated range of contexts. Thi s is an empirical statement,
which can be true or false. And it says, second, that in the
given context, the given solution solves this gi ven problem
a g a i n an empirical statement, whi ch can again be true
or false.
The statement that the pattern is alive, is thus not a
matter of taste, or culture, or point of vi ew. It establishes,
instead, a definite empirical relationship between a limited
context, a set of forces which occurs there, and the pattern
which resolves those forces.
In order to discover patterns which are alive we must
always start with observation.
T h e discovery of a pattern which lives is not di fferent
from the discovery of any profound thing. It is a slow,
deliberate process, tentative, in whi ch we seek to discover
something profound, and where we recognize that we
shall usually be wr ong to start wi th, and that we may
only approach a proper formulation sl owl y.
Eet us take the case of entrances, as an example.
Start by wal ki ng around, looking at house entrances,
254
PATTERNS WHI C H CAN BE SHARED
paying attention to whet her they feel right to you or not,
whether they feel comfortable, alive . . . .
Put the entrances in t wo classes: those in which the
process of entering feels good, and those in which it
doesn' t.
Now try to discover some property which is common
to all the ones which feel good, and missing from
all the ones which don't feel good.
Of course, you won' t be able to do this perfectly. One
entrance may not feel good, but be beautiful in some
entirely di fferent wa y ; however, with all the imperfec-
tions of experiment, come as near as you can to defi ni ng
some property which all the good ones have, and which
all the bad ones l ack: in short, look for the criterial
property whi ch makes the di fference.
This property will be a, highly complex relationship.
It will not be simple, like "al l the good ones are blue, all
the bad ones are not bl ue. " In the case of the entrance,
for instance, it turns out, in my experience, that all the
good ones have an actual place, between the road and the
front door, a place in whi ch there is a change of surface;
change of vi ew; perhaps a change of l evel ; perhaps you
pass under a branch, or hangi ng rose, there is often a
change of direction, and there is above all this actual
place, hal f way between the t wo, so that you pass first
from the street to this place, and then again from this
255
T HE GATE
place to the front door. Very often, in the best cases, this
place has a glimpse of a distant v i e wof something which
you cannot see from the street, nor from the door, but
see only for an instant in between the two. Or some-
times it is more mundane than that. I n a typical London
house, there is just a short pl atform, raised a couple of
steps, marked by its railing, a place to pause. It is rudi-
ment ary; perhaps too sparse; but in this very restricted
and dense situation, at least it does a little of the work.
Now try to identify the problem which exists in
entrances which lack this property.
T o do this, we must try to make explicit just which
forces are at wor k; and we must formulate the pattern
in terms which make it clear just why it helps resolve
some system of forces that cannot be resolved without it.
If we ask ourselves why entrance transitions are im-
portant, we recognize that they create some kind of "i n
bet ween, " a breathing space between the outside and
the i nsi dea place of preparation, where a person can
change his frame of mind, and adapt to the di fferent
conditions: from the loud, noisy, public, vulnerable, ex-
posed feeling of the street, to the private, quiet, intimate,
protected feeling of indoors.
If we try to formul ate the forces governi ng this transi-
tion precisely, we see that they shed a great deal of light
on the invariant whi ch makes transitions function.
For example, there is evidence that this "cl eani ng" of
256
PATTERNS WHI C H CAN BE SHARED
the street mask happens when a person passes through a
zone whi ch is di fferent to the senses, and discontinuous,
from the street.
Knowledge of the problem then helps shed light on
the invariant which solves the problem.
I f , indeed, there are forces like this at work, we can
deduce that the kind of transitions whi ch will work best
are those where many di fferent sensual qualities change
duri ng the passage from outside to inside: a change of
vi ew, a change of surface underfoot, a change of light, a
change of sound, a change of height or level, steps, a
change of s mel l a hangi ng jasmine . . . .
If we then take this, the product of our inferences
whi ch can be built out of the statement of the problem,
and look at some more entrances, with this in mind, our
capacity to distinguish entrances whi ch work from those
whi ch don' t gets sharper.
Th e statement of the problem and the forces helps us
to sharpen the pattern whi ch is responsible for maki ng
the system of forces come to equilibrium.
Th e process of observation does not proceed in linear
fashion f rom the problem to the solution, nor from the
solution to the problem . . . it is a global process, in
whi ch, by any means we can, looking at the matter from
all directions at once, we try to identify a solid and
reliable invariant, whi ch relates context, problem, and
solution, in an unchangi ng way.
257
T HE GATE
Sometimes we find our way to this invariant by
starting with a set of positive examples.
Tha t is what we have been doing in the case of the
alcoves, and in the case of the entrances. In each of these
cases we manage to identify an essential feature of a
number of places which make us feel good.
At other times, we may discover the invariant by
starting from the negative examples, and resolving
them.
For instance: Th e problem of gl oom and shade. I notice,
let's say, that the northside of my house is dank and dark,
and no one ever seems to go there, or use it for anythi ng.
I wonder what I can do about it.
I look around, and find that those outdoor areas to
the immediate north of buildings often seem to be
disarrayed, not worki ng, falling apart.
T h e n I start looking for those places which don' t have
any land around them like that. I realize that it happens
whenever houses, buildings, manage to be to the north of
the land they sit on, with the open land l eft to the south,
so that it gets the sun.
I do an experiment to find out if my intuitions are
born out by careful observation.
We ask people where they sit outside, around their
houses, and where they never sit. I n 19 cases out of 20,
the place they sit is to the south of the house, and next to
it; the places where they never sit are all to the north.
And so we formulate the pattern, SOUTH FACING OUT-
DOORS.
258
PATTERNS WHI C H CAN BE SHARED
T h e positive and negative approaches to finding pat-
terns are al ways compl ement ar ynot exclusive. T h e al-
cove whi ch started as an abstraction from something posi-
tive, mi ght just as wel l have started with an analysis of
what is wr ong in modern living rooms.
T h e south facing open space, whi ch started as a solu-
tion to something negat i vet he dark and unused char-
acter of northern spaceal so is inspired, of course, by
the wonder and warmt h of sunny l awns and terraces
which do face south.
And occasionally, we do not start from concrete ob-
servation at all, but build up the invariant by purely
abstract argument.
For of course, the discovery of patterns is not al ways his-
torical. Some of the examples I have given mi ght make it
seem as though the only way to find patterns is to get
them from observation. Thi s woul d imply that it was i m-
possible to find patterns whi ch do not already exist in the
worl d al ready: and woul d therefore imply a claustrophobic
conservatism; since no patterns whi ch do not already exist
could ever be discovered. T h e real situation is quite dif-
ferent. A pattern is a discovery in the sense that it is a
discovery of a relationship between context, forces, and
relationships in space, whi ch holds absolutely. Thi s dis-
covery can be made on a purely theoretical level.
For exampl e, the pattern PARALLEL STREETS was dis-
covered by purely mathematical reasoning, based on the
forces whi ch connect high speed vehicular movement to
the needs of pedestrians, the problem of accidents, the
259
T HE GATE
huge travel time, the very slow average speeds, etc. At
the time we discovered it, we were unaware that it actu-
ally was an emergi ng pattern in the worl d of the 1 9 6 0s
and only later realized that separated parallel arteries,
without cross streets, was emergi ng as a pattern in several
maj or cities.
In this same sense, it was possible to "di scover"
uranium, by postulating the existence of a chemical ele-
ment wi th certain properties, before it had actually been
observed.
In all these cases, no matter what method, is used} the
-pattern is an attempt to discover some invariant fea-
ture, which distinguishes good places from bad places
with respect to some particular system of forces.
It tries to capture just that essencethat field of rela-
ti onshi pswhi ch is common to all possible solutions to
the stated problem, in the stated context. It is the in-
variant behind the huge variety of forms which solve the
problem. The r e are millions of particular solutions to any
gi ven probl em; but it may be possible to find some one
property whi ch will be common to all these solutions.
Th a t is what a pattern tries to do.
Many people say they don' t like the fact that a pat-
tern gives "one solution" to a problem. Thi s is a serious
misunderstanding. Of course, there are thousands, mil-
lions, in fact, an infinite number of solutions to any given
problem. The r e is, of course, no way of capturing the
details of all these solutions in a single statement. It is
al ways up to the creative imagination of the designer to
260
PATTERNS WHI C H CAN BE SHARED
find a new solution to the problem, which fits his particu-
lar situation.
But when it is properly expressed, a pattern defines an
invariant field which captures all the possible solutions to
the problem given, in the stated range of contexts.
The task of finding, or discovering, such an invariant
field is immensely hard. It is at least as hard as any-
thing in theoretical physics.
My experience has shown that many people find it hard
to make their design ideas precise. Th e y are wil l ing to
express their ideas in loose, general , terms, but are
unwi l l i ng to express them wi th the precision needed to
make them into patterns. Above all, they are unwil l ing to
express them as abstract spatial relations among wel l -
defined spatial parts. I have also found that people aren' t
al ways very good at it; it is hard to do.
It is easy to say that a house entrance should have a
sort of mysterious quality, which both hides the
house from the public domain, and also exposes it
to the public domain.
Ar gument about the statement is imprecise. Archi tects
have been doing a great deal of wool l y thinking of this
sort. It is a kind of ref uge. If I say, instead, that the
front door must be at least 20 feet f rom the street, that it
should be visible, that wi ndows from the house should
look onto the area in front of the house, but that it should
not be possible to see into these wi ndows from the street,
261
T HE GATE
that a change of surface is required during the transition,
that the person arriving should enter a domain of a char-
acter as different as possible from either of the t wo do-
mains inside the house or on the street, and that he should
have a glimpse of some view whi ch lies entirely hidden
from the st reet t hen these statements can be chal l enged,
because they are precise.
But it is very hard to be precise.
Even once you are determined to do it it is terribly hard
to make precise statements whi ch really get to the heart
of the matter. Ever y observation, like the one about the
mystery of house entrances whi ch I have just made, starts
with intuition. T h e task of identifying just precisely those
relations that are at the heart of such an intuition, is no
easier in architecture than it is in physics, or biology, or
mathematics. Maki ng abstractions whi ch are powerf ul
and deep is an art. It requires tremendous ability to go
to the heart of things, and get at the really deep abstrac-
tion. No one can tell you how to do it in science. No one
can tell you how to do it in design.
And it is especially hard to be precise, because there is
never any one formulation of the pattern which is
perfectly exact.
I t is easy to understand this, when we recogni ze how
limited we are in our mathematical ability even to express
simple patterns in precise terms.
262
PATTERNS WHI C H CAN BE SHARED
Ta ke , for example, the idea of a "rough ci rcl e. " If I
ask you to point to things whi ch are rough circles, you can
do it easily. But if I ask you to define precisely, what we
mean by a rough circle, it turns out to be very hard to
do. T h e strict mathematical definition of a circle (points
exactly equidistant f rom a point cent er) , is much too nar-
row. None of the rough circles in nature fol l ow this rule
exactly. On the other hand a looser definition (points be-
tween nine and ten inches from some given point, for in-
stance), is much too loose. It woul d include, for instance,
a weird zi gzaggi ng structure, in whi ch no t wo points of
the "ci rcumf erence" are near each other. Ye t even a
rough circle has some kind of continuity al ong its cir-
cumf erence.
T o capture "rough circles" exactl y, we must find a
formulation whi ch just falls between being too narrow
and too loose. But this turns out to be a deep, and diffi-
cult mathematical problem.
And if a simple circle, whi ch any child can draw with
his finger in the dust, is so hard to pin down, it is easy to
see that a complex invariant like "ent rance transition," is
almost impossibly difficult to pin down precisely.
Instead, to strike the balance between being too nar-
row and too loose, you must express and visualize a
pattern as a kind of fluid image, a morphological feel-
ing, a swirling intuition about form, which captures
the invariant field which is the pattern.
T h e pattern ENTRANCE TRANSITION deals wi th the fact
263
T H E GAT E
that people on the street are in a publ ic f r ame of mi nd,
and need to pass t hr ough a zone wher e they can take this
f r ame of mi nd of f , bef ore they ent er into the kind of
personal i nt i macy or closeness typical in a house.
As we have seen, there is evi dence that tin's " c l e a ni ng "
of the street mask happens wh e n a person passes t hr ough
a zone whi ch is di f f er ent to the senses, and di sconti nuous,
f r om the street. So at one time the pat t ern was f or mul at ed
as: " Ma k e a path bet ween the street and the f r ont door of
the house; pass t hr ough a transi ti on z one wher e it changes
di recti on, changes l evel , changes surf ace, has a change of
vi ew, and a change of l i ght qual i t y. "
Fo r some houses ( e. g. , in a Cal i f or ni a subur b) this
phrasi ng is perf ect l y exact . But , at hi gher densities, whe n
there are no f r ont gardens, and the f ront door on the
street al l ows people to stand in door ways , in t ouch wi t h
the street, the transition cannot be bet ween the f ront door
and the street.
I n a case like this, the transi ti on can be handl ed per-
f ect l y wel l inside. F o r exampl e, in our Per u houses, we
made a court , inside the f ront door, wi t h the mai n l i vi ng
areas sur r oundi ng this court , and openi ng of f it. T o pass
into the house, a person fi rst passes t hr ough the f ront
door, then under the darkness of the ent ry wa y , then into
the l i ght of the patio, then into the cool of the veranda
whi ch connect s the f ami l y r oom and sala. T h i s is the
tradi ti onal Spanish sol ut i on: and, of course, it does con-
tain the ENTRANCE TRANSITION pat t ern.
But , as you can see, it does not f ol l ow the l etter of
264
PATTERNS WHI C H CAN BE SHARED
tiie pattern, as stated above: the transition isn't between
the street and the front door i t is inside the front door.
T o make sense of the design, it was necessary to fol l ow
the spirit of the pattern, not the letter.
Wha t is really happening, is that there is a feeling for
a certain kind of morphol ogy, which is geometrical in
character, but which is a feeling, not a precisely statable
mathematically precise relationship.
A pulsating, fluid, but nonetheless definite entity swims
in your mi nd' s eye. It is a geometrical i mage, it is far
more than the knowl edge of the probl em; it is the knowl -
edge of the problem, coupled with the knowl edge of the
kinds of geometries whi ch will solve the problem, and
coupled with the feeling whi ch is created by that kind of
geometry solving that problem. I t is above all, a feel i ng-
a morphological feeling. Thi s morphological feeling,
whi ch cannot be exactly stated, but can only be crudely
hinted at by any one precise formul ation, is the heart of
every pattern.
Then, once you discover a fluid field of relationships
like this, you must redefine it, as an entity, to make
it operational.
It is only when you do this that it becomes useful opera-
tional i nstructi onbecause you can now tell a person to
build "one of those."
Remember that our patterns are the building blocks in
terms of whi ch we see the worl d. Ki t chen, sidewalk,
265
T H E GAT E
hi gh-ri se offi ce bui l di ngt hes e are the pat t erns of our
t i me a n d our wor l d is made of t hem. Ne w patterns
must be the ne w bui l di ng bl ocks wi t h whi ch we wi sh to
bui l d; and the ment al bui l di ng bl ocks wi t h whi ch we
wa nt to see the wor l d.
Ag a i n, consi der the ENTRANCE TRANSITION. T h e
ori gi nal di scovery s hows that it is necessary f or a person
to experi ence many di f f er ent changes of l i ght , surf ace,
di recti on, vi ew, and sound whe n he enters a house, so
that the house is suffi ci entl y separated f r om the public
qual ity of the street. T h i s can be expressed as a propert y
of the path f r om the street to the door. W e can say
that this path must have certai n properti es. But at this
state the i nsi ght is not yet properl y f or mul at ed as a
pat t ern.
T o make it a pat t ern, I have to ask mysel f : Wh a t ne w
entity do I wa nt to put in the wor l d, to create these
properties?
Wh a t enti ty sums up and captures the f i el d of rel ati on-
ships whi ch these properti es def i ne? T h e ans wer is, an
enti ty whi ch is an act ual space, "the transition " wher e
the l i ght and col or and vi ew and sound and surf ace
change.
O f course, we k no w that this " t h i n g " I no w cal l "the
transition" is real l y not hi ng but a f i ct i on: because wha t
seems to be a t hi ng is enti rel y def i ned by its field of rel a-
tions a ny wa y . But there is some qual ity of the human
mi nd whi ch requi res that this field be treated as a t hi ng,
in order f or it to be underst ood, and made, and used as
part of a l anguage.
266
PATTERNS WHI C H CAN BE SHARED
We must make each pattern a thing so that the human
mind can use it easily and so that it may take its part
among the other patterns of our pattern languages.
For the same reason you must be able to draw it.
If you can' t draw a di agram of it, it isn't a pattern. If
you think you have a pattern, you must be able to draw
a diagram of it. Thi s is a crude, but vital rule. A pattern
defines a field of spatial relations, and it must therefore
al ways be possible to draw a diagram for every pattern.
In the di agram, each part will appear as a labeled or
colored zone, and the layout of the parts expresses the
relation which the pattern specifies. If you can' t draw it,
it isn't a pattern.
And finally, for the same reason too, you must give it
a name.
T h e search for a name is a f undamental part of the
process of inventing or discovering a pattern. So l ong as
a pattern has a weak name, it means that it is not a clear
concept, and you cannot clearly tell me to make " one. "
Suppose that I am in the process of discovering the
entrance transition pattern. Let us say that I am dimly
aware of the need for some kind of change, between
street and house: that sensory changes help to create this
change of attitude. Perhaps, to begin with, I call the
pattern ENTRY PROCESS. I hope that the name entry
process will explain this fact to you. But actually, you do
267
T H E GAT E
not know how to create this change of f eel i ng. Ent r y
process is still too vague.
No w suppose I call the pattern HOUSE STREET RELA-
TIONSHIP. A t this st age, I am awar e that it is the con-
crete geomet r y whi ch wi l l create the transi ti on. I am
awar e that some ki nd of relationship is needed. But I still
don' t know wha t the relationship is.
No w suppose I change the name to FRONT DOOR IN-
DIRECTLY REACHED FROM STREET. T h i s specifies a par-
ti cul ar rel ati onshi p, whi ch I can act ual l y build into a
desi gn. But it is still a relationship. I t is still hard f or me
to be sure that it is present, or mi ssi ng, f r om a desi gn.
So f i nal l y, I call it ENTRANCE TRANSITION, wi t h the
idea that the transition is an actual pl ace, bet ween the
street and the house, wi t h certai n defi ni te characteri sti cs.
N o w I can simpl y ask mysel f , Ha v e I built an ENTRANCE
TRANSITION into my desi gn? I can ans wer at once. I f I
tell you to build an ENTRANCE TRANSITION, you know
exact l y wha t to do, and you can do it.
It is concret e, and easy to do. An d it is mor e accurat e.
Fi nal l y, I underst and f ul l y wha t the probl em is, and
know wha t to do about it.
At this stage, the pattern is clearly sharahle.
Peopl e can discuss it, reuse it themsel ves, i mprove it,
check the observati ons out f or themsel ves, decide f or
t hemsel ves whet her they wa nt to use the pat t ern in a
parti cul ar bui l di ng whi c h they are maki ng. . . .
An d , perhaps even mor e i mport ant , the pat t ern is open
enough to become empi ri cal l y vul nerabl e.
268
PATTERNS WHI C H CAN BE SHARED
We can ask ourselves: Is it true that this system of
forces actually does occur, within the stated context?
Is it true that the actual solution, as formul ated, really
does resolve this field of forces in all cases?
Is it true that the precise formulation of the solution is
actually necessary: that any entrance which lacks this
feature must inevitably have unresolvable conflicts in it,
whi ch will communicate themselves to people who pass
through it?
We can therefore intensify our empirical observations,
and begin a second round of observations, which will
fine-tune the first observations.
Of course} even now the pattern is still tentative.
It is an attempt to define an invariant: but al ways only an
attempt. Just because it is solid, a thing, precisely f ormu-
lated, does not alter the fact that it is still only a guess at
what it is that makes an entrance marvelous.
Thi s guess may be wr ong in the formulation of the
probl em; for instance, the forces may be quite di fferent
from the ones which have been described.
And this guess may be wr ong in the formulation of the
solution; for instance, the actual pattern of relationships
required to help resolve the problem may be incorrectly
stated.
But it is clear enough, now, so that it can be shared.
Anyone who takes the trouble to consider it carefully can
understand it. It has a clearly formulated problem, based
269
T HE GATE
on empirical findings that anyone can check for them-
selves, and check against their own experience. It has a
clearly formulated solution that anyone can understand,
and check against the entrances which seem to work in
their part of the worl d. It has a clearly formulated con-
text, whi ch allows a person to decide whether the pattern
is applicable or not, and lets him see if he agrees with the
range of its application. And final l y, of course, anyone
can use the pattern. It is so concrete, so clearly expressed
as a rule, and as a thing, that anyone can make one, or
conceive one, in the buildings where he lives, or in a
building whi ch is goi ng to be created.
In short, whether this formul ation, as it stands, is cor-
rect or not, the pattern can be shared, precisely because
it is open to debate, and tentative. Indeed, it is the very
fact that it is open to debate, that makes it ready to be
shared.
To show you finally, how natural it is for anyone to
formulate -patterns which can be shared, I shall now
describe a conversation with an Indian friend, in
which I tried to help her define a pattern from her
own experience.
Chri s: First of all, just tell me a place that you like very
much.
Gi t a: Shall I describe it to you?
C. No, just think of it, and visualize it, and remem-
ber what you like about it.
G. It is an Inn.
270
P A T T E R NS W H I C H CAN BE S HARED
C . No w, please try to tell me what it is that makes
this I nn a special and wonde r f ul pl ace.
G. We l l , it is the thi ngs whi ch happen t he r e i t is a
pl ace wher e people wh o are on l ong j our neys
meet , and spend a little time t oget her, and it is
the wonde r f ul at mosphere of all these thi ngs
whi ch happen there. I like it ver y muc h.
C . C a n you try and isolate any f eat ure of its desi gn,
whi ch makes it so wonde r f ul ? I woul d like you
to try and tell me, as cl earl y as you can, wha t I
must do to create anot her place whi c h is as nice as
your I nnpl e a s e gi ve me an i nstructi on whi ch
capt ures one of the good t hi ngs about the desi gn.
G. I t is not the bui l di ng whi ch makes this I nn so
wonde r f ul , but it is the t hi ngs that happen t h e r e
it is the people you meet , the thi ngs you do there,
the stories whi ch people tell bef ore they go to
sleep.
C . Ye s , this is exact l y wha t I mean. O f course, it is
the at mosphere whi ch makes the I nn so wonde r -
f u l n o t the beauty of the bui l di ng, or its g e om-
e t r y ; but I am aski ng you if you can def i ne f or
me, whi ch f eat ures of the bui l di ng it is, that make
this at mosphere possible, the people who pass
t hr ough the I nn to creat e this at mosphere, all
that . . . .
G . I don' t underst and what you are sayi ng. I have
j ust told you it real l y doesn' t depend on the bui l d-
i ngs, but on the peopl e.
C . We l l , l et me put it like this. I magi ne an Ame r i c a n
27 1
T H E GA T E
mot el . Coul d the at mosphere whi ch you are de-
scri bi ng, happen in an Ame r i c a n mot el ?
G . Oh , no w I see wha t you are sayi ng. No , in these
Ame r i c a n mot el s it does not happen; there are so
ma ny pri vate rooms, and the people wh o come to
the mot el , t hey j ust pass t hr ough the mai n l obby,
they tal k at the count er f or a f ew mi nut es, and
then they go to thei r rooms. T h e I nn I am t al ki ng
about is not like t h a t b u t perhaps it is not possible
in Ame r i c a to have an I nn like t hi s i t is a social
pr o b l e mhe r e in Ame r i c a people wa nt to be so
pri vat e, they do not wa nt to meet and t a l k a n d
they do not like to sleep wi t h thei r husbands or
wi ves whe r e ever yone can see t h e ms o perhaps
it is very special, this at mosphere that I am de-
s c r i bi ngi t depends on the people wh o are usi ng
the I nn, and thei r habits, and wa y of l i f e.
C . Ye s , that is f i ne. Ev e r y pat t ern has a cont ext . O f
course, the pat t ern you are t r yi ng to def i ne may
not make sense f or the Uni t e d St at es per haps it
applies onl y to the cont ext of I ndi a. Le t ' s say that
this pat t ern is t rue j ust f or I n d i a n o w try and
tell me wha t it is all about .
G . Al r i ght . I n I ndi a, there are many of these I nns .
T h e r e is a cour t yar d whe r e the peopl e meet , and
a place to one side of the court yard wher e they
eat, and also on one side there is the person wh o
l ooks af t er the I nn, and on the ot her t hree sides
of the cour t yar d t here are the r o o ms i n f r ont of
the rooms is an arcade, maybe one step up f r om
27 2
P A T T E RNS WH I C H CAN BE S HARE D
the cour t yar d, and about ten feet deep, wi t h an-
ot her step l eadi ng into the rooms. Du r i n g the
eveni ng ever yone meet s in the cour t yar d, and they
talk and eat t og e t he r i t is ver y s pec i al and then
at ni ght they al l sleep in the arcade, so they are
al l sl eepi ng t oget her , round the cour t yar d. I think
it is very i mpor t ant , too, that all the r ooms are
si mi l ar, so t hat wh e n they stay here, all the people
feel equal , and free to talk to ever yone else.
C . I t sounds wonde r f ul . N o w , l et' s talk about the
probl em whi c h the pat t ern sol ves. Is it necessary?
D o you think that people can manage j ust as wel l ,
wi t hout the pat t ern as you have described it?
G . I can' t see how else it can be done; if you have
the r ooms separate, and pri vat e, it becomes j ust
like a mot el , and ever yone is al one. An d if you
don' t eat t oget her, wha t chance is there f or tal k-
i ng? I thi nk it must be j ust as I have described it.
Al l the I nns I know in the rel i gi ous t owns of
Indi a are like t h i s I can' t even i magi ne one
whi c h isn' t.
C . Le t ' s def i ne the probl em like this: " Wh e n people
t ravel , they are a little l onel y; and also since they
t ravel in or der to be opened up to the wor l d, they
wa nt a chance to be t oget her wi t h ot her t r avel er s. "
No w , can you tell me wh e n you think this pat-
tern makes sense, and whe n it d o e s n' t wh a t is
the ri ght cont ext for the pat t ern?
G . We l l , it must be f or a pl ace whe r e people are
t r avel i ng ver y f ar, and wher e tliey are in this
2
73
T HE GATE
moodi n India most of these Inns are at religious
places, where people come to make a pi l gr i mage
I think it is a very special kind of crossroads in a
j ourney, that must be like this.
C. Woul d it make sense in Greenl and?
G. I do not understand.
C. Do you think the climate is a part of the context?
G. Oi l yes, it is very important that it is a hot place,
so that you do not sleep out under the arcade for
social reasons onl y, but because of the heat al s o
you find the place where there are breezes, and
you put your bed where it is most comfortabl e.
C. So the pattern makes sense for any Inn where
people are on l ong j ourneys, in a society where
they are open to meet people in a very simple way,
and in a climate where it is hot so that people want
to sleep outside.
Now, once again, we have the beginnings of a
pattern.
We have stated a probl emso far in very intuitive
t er ms as an atmosphere which plays an important part
in the life of an inn where people meet. We have de-
scribed the field of spatial relationshipsthe arrangement
of the courtyard, the arcade, the eating place, the sleep-
ing, and the i nnkeeperwhi ch make this atmosphere
possible. And we have stated the context in whi ch it seems
that this pattern makes sense. Each of these three needs
to be refined, and perhaps made more preci sebut we
have the beginnings of a pattern now. . . .
274
PATTERNS WHI C H CAN BE SHARED
A great variety of patterns with this format have
been discovered.
T e n years ago a group of us began defi ni ng patterns, to
create a l anguage. 253 of these patterns are now published
in volume 2 of this series, A Pattern Language.
T h e 253 patterns range from the very large to the
very small. T h e largest ones deal with the structure of a
region, with the distribution of towns, and with the in-
ternal structure of a t own. . . . middle range patterns
cover the shape and activity of buildings, gardens, streets,
rooms . . . . and the smallest patterns deal with the
actual physical materials and structures out of which the
buildings must be made: the form of columns, vaults,
wi ndows, walls, windowsills, even the character of orna-
ment.
Each of these patterns is an attempt to capture that
essence of some situation which makes it live.
Each one is an invariant field, needed to resolve a conflict
among certain forces, expressed as an entity which has a
name, with instructions so concrete that anyone can
make one (or help to make one) , and with its functional
basis so clearly stated that everyone can decide for him-
self whether it is true, and when, and when not, to in-
clude it in his worl d.
Gradually} by hard work, it is possible to discover
many patterns which are deep, and which can help to
bring a building or a town to life.
275
T H E G A T E
They vary from culture to culture; sometimes
they are very cliferent, sometimes there are versions
of the same pattern, slightly diferent, in different
cultures.
But it is possible to discover them, and to write
them down so that they can be shared.
2 7 6
CHAPTER 15
THE REALITY OF PATTERNS
We may then gradually improve these
patterns which we share, by testing them
against experience: we can determine,
very simply, whether these patterns
make our surroundings live, or not, by
recognizing how they make us jeel.
We have seen in the last chapter that there is a
process by which a person can formulate a pattern-,
and make it explicit, so that other people can use it.
Many such patterns have been written down, in
volume 2.
But so far, there is no guarantee at all that any one
of these patterns will actually work. Each one is in-
tended to be a source of life, a generative, self-
sustaining pattern. But is it actually? How can we
distinguish patterns which work, which are deep and
worth copying, from those which are simply pipe
dreams, mad imaginings . . .
Suppose that we are trying to agree about a pattern.
How can we agree whet her it lives or not?
Or , suppose that you are reading a pattern which
someone has written down.
How can you decide whet her to make it part of your
l anguage or not?
One test says that a pattern is alive if its individual
statements are empirically true.
We know that every pattern is an instruction of the
general f or m:
context conflicting forces confi gurati on
So we say that a pattern is good, whenever we can
show that it meets the f ol l owi ng t wo empirical condi-
tions:
282
T HE REALI TY OF PATTERNS
1. The froblem is real. Thi s means that we can ex-
press the problem as a conflict among forces whi ch really
do occur within the stated context, and cannot normal l y
be resolved within that context. Thi s is an empirical
question.
2. The configuration solves the froblem. Thi s means
that when the stated arrangement of parts is present in
the stated context, the conflict can be resolved, without
any side effects. Thi s is an empirical question.
But a pattern is not alive just because its component
statements are truey one by one.
One of the funniest examples of a pattern I have ever
heard of is the "madhouse bal cony. "
Thi s is a pattern whi ch a student once invented. It
says that the balcony of any mental patient's room should
have a chest-high railing on it. T h e argument is this: on
the one hand, people want to be able to enj oy the v i e w
and this applies as much to mental patients as anyone else.
On the other hand, mental patients "have a tendency to
j ump off buildings. " I n order to resolve this conflict be-
tween forces, the railing on the balcony must be high
enough to prevent a patient from j umpi ng over, but l ow
enough so that he can enj oy the vi ew.
We l aughed for hours when we first saw this. And
yet, absurd as it is, it seems to fol l ow the f ormat of a
pattern. It has a context, problem and solution: and the
problem is expressed as a system of conflicting forces.
Wh a t is it that makes it absurd?
283
T HE GATE
The fact is that even though its individual component
statements are true, the pattern has no empirical real-
ity as a whole.
A balcony of this kind will not al l ow a mad person to
heal himself: it will not help to make the worl d more
whol e.
It is absurd because we can feel in our bones that it
woul d make no difference at all whether such a balcony
was built into the worl d or not. We know that the
problem cannot really be solved in this or any related way.
Even the fact that a pattern seems sensible, and has
clear reasoning behind it, does not mean at all that
the pattern is necessarily capable of generating life.
For exampl e, the f amous radiant city pattern of high
towers, freestanding in the landscape, was "i nvent ed" by
Le Corbusi er with great devotion and seriousness. He
believed that it woul d be possible to give every family
light, and air, and access to green, within this pattern:
and he spent many years developing this pattern, in
theory and practice.
However , he f orgot, or did not realize, that there was
one additional essential force at wor k in the syst emt he
human instinct for protection and territoriality. T h e
huge, abstractly beautiful green spaces around his high
buildings are not used, because they are too public, they
belong to too many people at the same time, and they
are under the eyes of too many hundreds of apartments
284
T HE REALI TY OF PATTERNS
hoveri ng above them. Under these circumstances, this
one f or c e a kind of animal territorial i nsti nctdestroys
this patttern' s capacity to generate life. . . .
A pattern only works, fully, when it deals with all
the forces that are actually present in the situation.
On the face of it this is a simple intellectual concept.
Wh e n we find a pattern whi ch does bring forces into
balance, then this pattern will of course begin to generate
the quality without a name whi ch is described in chap-
ter 2because it will contribute to that process in whi ch
the forces of the worl d run free. On the other hand, a
pattern al ways lacks this quality if it resolves some forces
at the expense of others whi ch it leaves unresolved.
It should be reasonably easy to identify these patterns
whi ch are alive, in these terms, and to distinguish them
from those patterns whi ch aren' t alive.
In practice, though, it turns out to be very difficult.
The difficulty is that we have no reliable way of
knowing fust exactly what the forces in a situation
are.
Th e pattern is merel y a mental i mage, whi ch can help
to predict those situations where forces will be in har-
mony, and those in whi ch they won' t .
But the actual forces whi ch will occur in a real situa-
tion, al though objectively present there, are, in the end
unpredictable, because each situation is so compl ex, and
285
T HE GATE
forces may gr ow, or die, accordi ng to subtle variations
of circumstance.
If we formulate a pattern in terms of some system of
forces, whi ch we think describes a situation, and our de-
scription of the system happens to be incomplete, then the
pattern can easily become absurd.
Ye t we have no analytical way of being sure just what
the forces are.
What we need is a way of understanding the forces
which cuts through this intellectual difficulty and goes
closer to the empirical core.
We need a way of knowi ng whi ch patterns will really
help to bring the worl d to life and whi ch ones won' t .
And we need a way of doing it whi ch is more reliable
than analytical formulation. Above all, we need a way of
doing it, whi ch is anchored in the empirical reality of what
will actually happen, without necessarily requiring com-
plex and extensive experiments whi ch are too expensive
to do.
To do this, we must rely on feelings more than in-
tellect.
For although the system of forces in a situation is very
hard to define analytically, it is possible to tell, in a holistic
way, whet her the pattern is alive or not.
T h e fact is that we feel good in the presence of a pat-
tern whi ch resolves its forces.
286
T HE REALI TY OF PATTERNS
And we feel ill at ease, uncomfortabl e, when a pattern
leaves its forces unresolved.
The pattern ALCOVE feels good to us, because we feel
the wholeness of the system there.
Ther e is an intellectual formulation of the forces which
alcoves resolve. For instance, they allow us to be private
at the edge of a communal gatheri ng, and, at the same
time, remain in touch with whatever is communal there.
But what clinches it, what makes us certain that this
formulation has some substance to it, is the fact that al-
coves make us feel good. T h e conflict is real, because the
alcove makes us feel al i ve; and we know the pattern is
complete, because we can feel no residual tension there.
The pattern T-JUNCTIONS makes us feel good, because
we feel the wholeness of the system there.
Ther e is an intellectual formulation of the forces which
T- j unct i ons resolve. A T- j unct i on creates less crossing
movements, and less conflicts for the drivers, and this
puts the pattern on a firm empirical foundation. But what
clinches it, and makes us certain that the problem is a real
one, and complete, is that we feel more comfortabl e, more
relaxed, when we are driving in a street whose junctions
are all T- j unct i ons. We know, then, that there are no
hidden crossing movements which we don' t expect; there
is no possibility of unexpected cars shooting across our
pat hi n short we feel good there; and we feel good be-
287
T HE GATE
cause the system of conflicting forces which T- j unct i ons
resolves is real, and complete.
And MOSAIC OF SUBCULTURES makes us jeel good,
because, again, we feel the wholeness of the system
there.
Agai n there is an intellectual argument, whi ch shows
that when subcultures are separated from one another by
communal land, each one can gr ow in its own way. I n
this case the system of forces is immensely intricate, and
we must wonder, indeed, if we have managed to identify
a complete balanced system of forces in this pattern.
Agai n, the certainty comes from the fact that we feel
gogd in places where this pattern does exist. In places like
the Chi nat own of San Francisco, or in Sausalito, whi ch
are vivid with their own life because they are a little
separate f rom the nearby communities, we feel good. We
feel good because we can feel, in our bones, the lack of
inhibition, the spontaneous growt h, which fol l ows its own
course in these communities, because they are uninhibited
by pressure from surrounding communities which have a
di fferent way of life.
By contrast, patterns made from thought, without
feeling, lack empirical reality entirely.
T h e madhouse balcony makes us feel nothing. Certai nl y,
we know at once, when we first hear it, that a balcony
like this will not make us feel wonderf ul . The r e is no
288
T HE REALI TY OF PATTERNS
feeling in it: and this lack of feeling is the way our knowl -
edge of its emptiness presents itself to us.
And Le Corbusier' s radiant city makes us feel worse:
it actively makes us feel bad. It may excite our intellect,
or our imagination; but when we ask ourselves how we
shall feel in a place which is really built like this, we
know again, that it will not make us feel wonderf ul .
Agai n, our feeling is the way our knowl edge of its
functional emptiness presents itself to us.
We see then, that there is a fundamental inner connec-
tion between the balance of a system of forces, and our
feelings about the pattern which resolves these forces.
It comes about because our feelings al ways deal with
the totality of any system. If there are hidden forces,
hidden conflicts, l urki ng in a pattern, we can feel them
there. And when a pattern feels good to us, it is because
it is a genuinely whol esome thing, and we know that
there are no hidden forces l urki ng there.
This makes it easier to test any given pattern.
Wh e n you first see a pattern, you will be able to tell al-
most at once, by intuition, whet her it makes you feel
good or not: whether, you want to live in a worl d which
has that pattern in it, because it helps you to feel more
alive.
If a pattern does make you feel good, there is a very
good chance that it is a good pattern. If a pattern does
289
T HE GATE
not help you to feel good, there is very little chance that
it is a good pattern.
We can always ask ourselves just how a pattern makes
us jeel. And we can always ask the same of someone
else.
Imagi ne someone who proposes that modular "aluminum
wal l panels are of great importance in the construction of
houses.
Simply ask him how he feels in rooms built out of
them.
He will be able to do dozens of critical experiments
whi ch " pr ove" that they are better, and that they make
the envi ronment better, cleaner, healthier . . . . But
the one thing he wi l l not be able to do, if he is honest
with himself, is to claim that the presence of modul ar
panels is a distinguishing feature of the places in whi ch he
feels good.
Hi s feeling is direct, and unequivocal.
It is not the same, at all, as asking someone his
opinion.
If I ask someone whet her he approves of "parki ng
garages" s ayhe may give a variety of answers. He may
say, " We l l it all depends what you mean. " Or he may
say, " The r e is no avoiding t he m" ; or he may say, " I t is
the best available solution to a difficult probl em" . . . on
and on.
290
T HE REALI TY OF PATTERNS
None of this has anythi ng to do with his feelings.
It is also not the same as asking for a person's taste.
If I ask a person whet her he likes hexagonal buildings,
say, or buildings in whi ch apartments made like shoe
boxes are piled on top of one another, he may treat the
question as a question about his taste. In this case he may
say, " I t is very i nventi ve, " or, wishing to prove that he
has good taste, " Ye s , this modern architecture is fascinat-
ing, isn't i t ? "
Still, none of this has anythi ng to do with his feelings.
And. it is also not the same as asking what a person
thinks of an idea.
Agai n, suppose I formul ate a certain pattern, and it
describes, in the problem statement, a variety of problems
which a person can connect up wi th his philosophical
leanings, his attitudes, his intellect, his ideas about the
wor l dt he n he may again give me a variety of con-
fusing answers.
He may say, " We l l , I don' t agree with your f ormul a-
tion of this or this f ac t " ; or he may say, " T h e evidence
you cited on such and such a point has been debated by
the best authorities"; or again, " We l l , I can' t take this
seriously, because if you consider its l ong term implica-
tions you can see that it woul d never do" . . .
Al l this again, has nothi ng to do wi th his feelings.
It simply asks for feelings, and for nothing else.
291
T HE GATE
Go to places where the pattern exists, and see how you
feel there.
Compare this with the way you feel in places where
the pattern is missing.
If you feel better in the places where the pattern exists,
then the pattern is a good one.
If you feel better in the places where the pattern does
not exist, or you can honestly detect no di fference, be-
tween the t wo groups of cases, then the pattern is no
good.
The success of this test hinges on a fact which I have
not said enough about so farthe extraordinary de-
gree of agreement in feofle's feelings about fatterns.
I have found that whereas people can get into the most
amazi ng and complex kinds of disagreement about the
"i deas" in a pattern, or about the philosophy expressed in
the pattern, or about the "t ast e" or "st yl e" whi ch seems
to be implied in a pattern, people who come from the
same culture do to a remarkable extent agree about the
way that different patterns make them feel.
Ta ke , for instance, the need that children have for
water. A f ew years ago I was at a meeting in San Fr an-
cisco where two hundred people met, for an afternoon, to
try to identify things that they want ed in their city. The y
met in groups of eight, around small tables, and spent
the af ternoon discussing what they wanted. At the end
of the afternoon, a spokesman f rom each group sum-
marized the things they wanted most.
292
T HE REALI TY OF PATTERNS
Several di fferent groups, quite independently, men-
tioned the fact that they wanted an opportunity for their
children to play in, and wi th, mud and waterespeci al l y
wat eri nst ead of the hard asphalt playing grounds
which parks and schools provide.
Thi s fascinated me. It happened that one of the pat-
terns in the l anguage we had been developing, POOLS AND
STREAMS, goes into great detail about the fact that
children especially, and all of us, need the opportunity to
play wi th wat erbecause it liberates essential subconscious
processes. And here, unasked for, was tenfold spontaneous
confirmation of the pattern, born directly out of people's
feelings.
Or take the question of the size of hospitals. Officials in
Sao Paul o have recently begun construction of the largest
hospital in the wo r l da hospital wi th 10, 000 beds. No w
9 out of ten peopl eprobabl y 95 out of 1 00wi l l agree
that a 10, 000 bed hospital fills them with fear and appre-
hension.
Contempl ate that simply as an empirical fact. It is an
empirical fact of an order of magni tude far vaster than
any piddling experiments and surveys which the experts
can muster.
There are few experiments, in science, where a
phenomenon is capable of generating this extraor-
dinary level of agreement.
And, yet, for some strange reason we are not yet wi l l -
ing to recognize the depth, and power and centered-
293
T HE GATE
ness of just these feelings. If the fact that Brazilian
people do not feel good when they think about this
hospital were mentioned in the legislature as a means of
shedding doubt on the experts' opinions, the legislators
woul d smile politely; it woul d even be embarrassing to
mention feelings in this situation. And yet this ocean of
shared feeling is the place where we become one with one
anot hert hi s is the source, in the end, of our agreement
about pattern languages.
It is easy to dismiss feelings as "subj ect i ve" and "unre-
liable," and therefore not a reasonable basis for any form
of scientific agreement. And of course, in private matters,
where people's feelings vary greatly from one person to
the next, their feelings cannot be used as a basis for agree-
ment.
However , in the domain of patterns, where people seem
to agree 90, 95, even 99 percent of the time, we may
treat this agreement as an extraordi nary, almost shat-
tering, discovery, about the solidity of human feelings,
and we may certainly use it as scientific.
But for fear of repeating myself, I must say once
again that the agreement lies only in peoples' actual
feelings, not in their opinions.
For exampl e, if I take people to wi ndow places ( wi ndow
seats, gl azed alcoves, a chair by a l ow windowsill looking
out onto some fl owers, a bay wi ndow . . . ) and ask
them to compare these wi ndow places with those wi ndows
in rooms where the wi ndows are flat inserts into the wal l ,
294
T HE REALI TY OF PATTERNS
almost no one will say that the flat wi ndows actually feel
more comfortable than the wi ndow pl acesso we shall
have as much as 95 percent agreement.
And if I take the same group of people to a variety of
places which have modular wal l panels in them, and com-
pare these places with places where walls are built up
from brick, and plaster, wood, paper, stone . . . almost
none of them will say that the modular panels make
them feel better, so l ong as I insist that I only want to
know how they feel. Agai n, 95 percent agreement.
But the moment I al l ow people to express their opin-
ions, or mix their ideas and opinions with their feelings,
then the agreement vanishes. Suddenly the staunch ad-
herents of modular components, and the industries which
produce them, will find all kinds of arguments to explain
why modul ar panels are better, why they are eco-
nomically necessary. And in the same way, once opinion
takes over, the wi ndow places will be dismissed as im-
practical, the need for prefabricated wi ndows discussed
as so important . . . all these arguments in fact falla-
cious, but nevertheless presented in a way whi ch makes
them seem compelling.
In short, the scientific accuracy of the patterns can
only come from direct assessment of people's feelings, not
from arguments or discussions.
These feelings which are in touch with reality are
sometimes very hard to reach.
Suppose, for instance, that a person proposes a pattern in
295
T HE GATE
whi ch wat er flows in four directions from a fountain.
If I say to this persondoes that make you feel good,
he says yes, of course, that is exactly why I do i t i t
makes me feel good.
It needs enormous discipline, to say, no, no, wait a
minute, I am not interested in that kind of glib stuff. If
you compare the situation where the water comes out in
one substantial fl ow from the pool, and can irrigate an
orchard, wi th the situation where the water trickles out
in four di recti onsand ask yourself, honestly now,
honestlywhich of the t wo makes you feel bet t ert hen
you know that it makes you feel better when the fl ow is
more substanti al i t makes more sense, the worl d be-
comes more whol e.
But it is hard to admit this, because it takes so much
hard work to concentrate attention on the feelings.
It is not hard because the feeling is not there, or be-
cause the feeling is unreliable.
It is hard, because it takes an enormous and unusual
amount of attention, to pay attention for l ong enough to
find out whi ch does actually feel better.
Yet it is only this true feeling, this feeling that re-
quires attention, this feeling that requires effort,
which is reliable enough to generate agreement.
And it is only this much deeper feeling, whi ch is con-
nected directly to the balance of forces, and to the
emergence of reality.
Once a person is wi l l i ng to take his feelings as seriously
296
T HE REALI TY OF PATTERNS
as this, and pay attention to t he mand exclude opinions
and i deast hen his perception of a pattern can approach
the quality without a name.
We see then that the concept of a balanced, pattern is
deeply rooted in the concept of feeling.
And that our feelings, when they are real feelings, pro-
vide us wi th a powerf ul way of fi ndi ng out just whi ch
patterns are balanced and whi ch ones are not.
But even so, feelings themselves are not the essence
of the matter.
For what is at stake, in a pattern which lives, is not
merely the fact that it makes us feel good, but, much
more than that, the fact that it does actually liberate a
portion of the worl d, allow the forces to run f ree: and
liberate the worl d from the imprisoning effect of concepts
and opinions.
In short, what is at stake at last, is nothing but the
quality without a name itself.
Some patterns have this quality; and others don' t. Those
whi ch do make us feel good, because they help to make us
whol e, and we feel more at one wi th ourselves in their
presence: but still it is the quality itself whi ch matters
most; not the effect whi ch it has on us.
297
T HE GATE
It is, in the end, the presence of this quality in a
pattern which makes the difference between one
which lives and one which doesn't. . . .
Ta k e as an example the relationship between pedestrians
and cars.
Conventi onal wisdom has it that pedestrians ought to
be separated from cars because it is quiet, and safe for
them.
But naggi ng reality shows us that even in towns where
they are kept completely separate, the children still run
out to play in parking lots, and people still wal k casually
along the roads reserved for cars. T h e fact is that people
take the shortest paths, and that cars are where the action
The r e is no doubt that pedestrians do need some
measure of protection from cars, for peace and quiet, and
for safety. But also, paths need to pass where the action is,
where pedestrians can meet the cars.
It is possible to deal with both of these forces at the
same time. If we put pedestrian paths at right angles to
roads, crossing them, but not entirely separate from
them, we create peace and safety, but also create places
where people on foot and cars can meet, hubs where the
action is, where the t wo systems cross.
And this pattern, whi ch does in fact resolve the con-
flict in the forces, also corresponds, of course, to just the
places where we feel most comfortable about the relation
between cars and people. Isn' t it true that in the busy part
of a city, a system of paths whi ch is entirely separate from
298
T H E R E A L I T Y OF P A T T E RNS
cars is too quiet, arti f i ci al , al most unreal ? I sn' t it, i nstead,
those paths whi ch are quiet, beaut i f ul , pedestri an paths
but whi c h do l ead up to a road, whi c h make us f eel per-
f ect l y in bal ance wi t h these f orces? As k yoursel f , if you
don' t k now a path like this in a ci ty, whe r e the road wi t h
cars on it is in si ght, crossi ng the path, but at one e n d
and ask yoursel f , whe n you are t here, if you don' t f eel
j ust ri ght .
T o this ext ent , this pat t ern, NETWORK OF PATHS AND
CARS, is based on real i ty. So l ong as the cont ext remai ns
in whi ch we have cars in the wor l d, this is the pat t ern
whi ch takes the f orces as they are and resol ves t hem,
wi t hout bias, by t reat i ng t hem exact l y as they real l y are.
It is reality itself which makes the difference.
T a k e anot her exampl e: I n the houses we built in Per u,
we based our patterns on the under l yi ng f orces we coul d
detect in peopl e' s lives. Si nce many of these f orces are
ages ol d, we wer e led to create houses whi ch have many
f eat ures in c ommon wi t h the anci ent and col oni al Pe r u-
vi an traditions. Fo r exampl e, we gave each house a
" S a l a " a special f or mal l i vi ng r oom, ri ght inside the
f ront door, f or recei vi ng f or mal guests, and a f ami l y
r oom, whe r e the f ami l y t hemsel ves mi ght l i ve, f ur t her
back into the house (see the pat t ern INTIMACY GRA-
DIENT) . An d we gave the houses l eani ng ni ches, outside
the f ront doors, whe r e people coul d stand, hal f in, hal f
outside the house, wa t c hi ng the street (see the pat t ern
FRONT DOOR RECESSES). The s e pat t erns are both c om-
299
T HE GATE
mon in traditional Peru. People criticized us strongly for
trying to go back to the past, when they said, the Peruvi an
families themselves were struggl i ng to catch up wi th the
future, and wanted houses just like Ameri can houses, so
they could have a modern way of life.
T h e issue here is not one of past or present or future.
It is a simple fact, that a Peruvi an family with a single
living room, will experience conflict whenever a stranger
visits t hemt hey try to keep their family around the
dinner table, talking and wat chi ng T V ; yet, at the same
time, they try to present the visitor with a formal way
of entering the house, not mixed up with the fami l y. And
again, if it is not possible to stand in the front door,
wat chi ng the street, many of the women will experience
a conflict between the fact that, being women, they are
expected to retire, not to be too f orward, or to sit openly
on the st reet and yet, being shut inside the house, they
want to experience some connection with the street and
the street life.
I do not j udge these facts. Th e y are simply facts about
the dynamics of being Peruvi an in 1969. So l ong as these
forces exist, people wi l l experience unresolvable conflicts,
unless the patterns are present and wi l l , to this extent, be
less able to become whol e.
And it is in the end only when our feelings are
;-perfectly in touch with the reality of forces, that we
begin to see the -patterns which are capable of gen-
erating life.
300
T HE REALI TY OF PATTERNS
Th a t is what is hardbecause so often people choose to
put their own opinions f orward, in place of reality.
I n many cases, people react to the description of these
forces by saying "i t should be otherwi se. " For instance,
ENTRANCE TRANSITION is based, in part, on the fact that
in a city street, people have a mask of street behavior,
whi ch needs to be wiped off by a transition, before a
person can relax in a private or secluded place.
One person's comment on this pattern was: Thi s fact
is bad; people should learn to be the same in the street as
they are in private places, so that we can all love one
another.
Th e comment is nice in its intent. But human beings
are not so malleable. T h e street mask is created by us, in
spite of our own volition: the fact that it comes into being
is a fundamental fact about human nature in urban
situations.
The r e is little purpose, then, in sayi ng: It woul d be
better if this force did not exist. For if it does exist any-
way, designs based on such wi shful thinking will fail.
T h e beauty of an ENTRANCE TRANSITION, and the
fact that it is capable of maki ng us feel at one wi th our-
selves, is based on thoroughgoi ng acceptance of these
forces as they really are.
Yet it is hard to give up preconceptions of what
things "ought to be" and recognize things as they
really are.
For instance, the other day a radio advertisement for the
301
T H E GA T E
boy scouts sai d: " Wh e n your boy is sitting on the street
corner wi t h other boys, this is unhe a l t hy g i v e hi m the
chance to do wha t all boys are year ni ng to d o g o i n g on
l ong hikes, fishing, and s wi mmi ng . " T h i s st at ement was
presented wi t h rel i gi ous f e r v o r a n d is, of course, a
del iberate at t empt , by bel i evers in the puri tan ethic, to
impose their concepti on of wha t a boy ought to be like,
on wha t a boy is act ual l y like. O f course, a real boy some-
times want s to go s wi mmi n g b u t somet i mes he want s to
hang out on the street corner wi t h his f ri ends, and some-
times he want s to l ook f or girl s. A person who believes
that these pursuits are " unhe a l t hy " wi l l never be able to
see the f orces whi c h are act ual l y at wor k in the boy' s
l i f e a n d wi l l never be able to use a pat t ern l anguage
wi t h any sense of real i ty.
Ho w coul d such a person r ecogni ze the real i ty of the
pattern TEENAGE COTTAGE whi c h says that a t eenager
needs a cot t age sl i ghtl y remot e f r om his parents, so as to
nurt ure the begi nni ngs of his i ndependence; or how
coul d he r ecogni ze the reality of PUBLI C OUTDOOR ROOM,
whi ch speci fi cal l y pays attenti on to the needs whi c h t een-
agers have to gat her , in urban public pl aces, a wa y f r om
their houses.
A person who believes in sl um cl earance wi l l be blind
to the real facts about the lives of people l i vi ng in sl ums.
A person wh o is convi nced that skid r ow ought to be
cl eaned up wi l l be obl i vi ous to the real f orces at wor k in a
hobo' s l i fe, because he can' t accept the exi stence of a hobo.
A person wh o is convi nced that offi ces ought to be " f l exi -
302
T HE REALI TY OF PATTERNS
bl e" will be oblivious to the real forces at work in groups
of people who are tryi ng to work.
Any preconception about the way things "ought to
be" al ways interferes with your sense of reality; it pre-
vents you f rom seeing what is actually goi ng ona nd
this will al ways prevent you f rom maki ng the environ-
ment alive. It will prevent you f rom inventing or dis-
coveri ng new patterns when you see t he mand, most of
al l i t will prevent you f rom using such patterns properly,
to create a whol e envi ronment.
In this res feet attention to reality goes far beyond
the realm of values.
Usual l y people say that the choice of patterns depends on
your opinions about what is important. One person thinks
high buildings are best; another person likes l ow ones;
one person likes plenty of space for cars, because he likes
driving fast; another one likes the emphasis to be given
to pedestrians, because he doesn' t like driving.
Wh e n we try to resolve disagreements like this, we
are led back to people's f undamental aims in l ife: to their
fundamental goals, or values. But people do not agree
about their values. So this kind of discussion still leaves us
in a position where patterns seem only to depend on
opi ni onsthe best you can say, accordi ng to this vi ew, is
that a certain pattern does or doesn' t help to satisfy a
certain goal or value. Or that some " f or ces" are " good"
and others " bad. "
303
T HE GATE
But a pattern whi ch is real makes no j udgment s about
the legitimacy of the forces in the situation.
By seeming to be unethical, by making no judgments
about individual opinions, or goals, or values, the
pattern rises to another level of morality.
Its result is to al l ow things to be al i veand this is a
higher good than the victory of any one artificial system
of values. T h e attempt to have a victory for a one-sided
view of the worl d cannot work anyway, even for the
people who seem to wi n their point of view. T h e forces
which are ignored do not go away just because they are
ignored. Th e y lurk, frustrated, underground. Sooner or
later they erupt in violence: and the system whi ch seems to
wi n is then exposed to far more catastrophic dangers.
T h e only way that a pattern can actually help to make
a situation genuinely more alive is by recogni zi ng all the
forces whi ch actually exist, and then finding a worl d in
whi ch these forces can slide past each other.
Then it becomes a piece of nature.
When we see the pattern of the ripples in a pond,
we know that this pattern is simply in equilibrium
with the forces which exist: without any mental in-
terference which is clouding them.
And, when we succeed, finally, in seeing so deep
into a man-made pattern, that it is no longer clouded
by opinions or by images, then we have discovered
a piece of nature as valid, as eternal, as the ripples in
the surface of a pond.
304
CHAPTER I 6
THE STRUCTURE OF A LANGUAGE
Once we have understood how to dis-
cover individual patterns which are
alive, we may then make a language for
ourselves, for any building task we face.
The structure of the language is created
by the network of connections among
individual patterns: and the language
lives, or not, as a totality, to the degree
these patterns form a whole.
It is clear, then, that we can discover living patterns,
and share them, and reach some reasonable degree of
confidence in their reality.
The patterns cover every range of scale in our
surroundings: the largest patterns cover aspects of
regional structure, middle range patterns cover the
shape and activity of buildings, and the smallest pat-
terns deal with the actual physical materials and
structures out of which the buildings must be made.
So far, though, we have said little about language.
In this chapter we shall see how it is possible to put
these patterns together to form coherent languages.
As we shall see, the possibility of l anguage is latent in the
fact that patterns are not isolated. But it comes out, in its
full force, when we experience the desire to make some-
thing. As soon as we want to make somet hi ngany-
t hi nga small thing like a garden seat, a large thing like
a nei ghborhoodand want to see it whol e, then we
experience the desire whi ch puts a structure on the pat-
terns, and makes l anguage out of them.
Imagine that I am going to build a garden.
Fr om chapters 10, I I , and 12, we know now that the
garden will not live, nor be a beautiful and stirring place,
unless we have a powerf ul , and deep and l i vi ng l anguage
for the gar denbef or e we start to lay it out.
Somehow, then, I must try to find, or create for my-
self, a pattern l anguage for a garden.
39
T HE GATE
One way to start a language for a garden is to get
some patterns from the pattern language we have pub-
lished in volume 2.
If you go through the patterns in that l anguage, choosing
the patterns for a garden whi ch seem relevant for you,
you mi ght, for instance, choose the f ol l owi ng list of
patterns:
HALF-HIDDEN GARDEN
TERRACED SLOPE
FRUIT TREES
TREE PLACES
GARDEN GROWING WILD
ENTRANCE TRANSITION
COURTYARDS WHICH LIVE
ROOF GARDEN
BUILDING EDGE
SUNNY PLACE
OUTDOOR ROOM
SIX-FOOT BALCONY
CONNECTION TO THE EARTH
GREENHOUSE
GARDEN SEAT
But what is it, now, which makes these patterns form
a language?
I can get these patterns from the published l anguage,
simply ticking off the ones I like, and wri ti ng them down
in the order of the l arger l anguage.
31 0
T HE STRUCTURE OF A LANGUAGE
And, of course, I mi ght also include patterns which I
have invented myself, or which my friends have told me
about mi xed in with the others.
But what makes a l anguage of t hem; and what makes
this l anguage whol e?
The structure of a pattern language is created by the
fact that individual patterns are not isolated.
T o understand this idea, fully, think about the pattern
" gar age, " and concentrate, in particular, on the garage
of a single-family house. How do you know that a par-
ticular building that you see is a garage?
Of course, you recognize it partly by the smaller pat-
terns it contains: by the fact that it is the size of a car, by
the fact that it has small wi ndows or none at all, by the
fact that it has a l arge, full-height door in front, etc.
These facts are defined by the pattern whi ch we call
" gar age. "
But the pattern of the garage and the smaller patterns
it contains are not enough to define the garage fully. If a
building with these patterns in this arrangement were
floating on a boat, you might call it a houseboat, but
certainly not a garage; if it were standing in the middle
of a field, with no road leading to it, it might be a tool-
shed, or a storehouse, but certainly not a garage.
For a building to be a garage, it has to have a drive-
way leading to it, from the street; and it will probably be
to one side of a house, not directly in front or directly be-
hi nd; and it will probably be fairly near the housewi t h
3
1 1
T HE GATE
a direct path to the house. These l arger patterns are part
of the garage pattern too.
Each -pattern then, defends both on the smaller pat-
terns it contains, and on the larger patterns within
which it is contained.
Exact l y the same is true of all the patterns in the l anguage
for a garden. Each one is incomplete, and needs the con-
text of the others, to make sense.
For example, a GARDEN WALL taken out of context is
merel y a pile of bricks. It becomes a garden wal l only
when it surrounds a garden; that is, for instance when it
helps to complete HALF-HIDDEN GARDEN or GARDEN
GROWING WILD.
ENTRANCE TRANSITION, taken by itself, is merel y a
place in the open air. Wh a t makes it an entrance transi-
tion is its position in between the front door and the
street, and its vi ew into the more distant garden; in short,
the fact that it helps to complete the l anguage pattern
MAIN ENTRANCE, and is itself completed by the smaller
pattern ZEN VIEW.
And perhaps most extreme of all, COURTYARDS WHICH
LIVE. Of course a courtyard is not a courtyard at all, un-
less it is surrounded by the buildings whi ch create it: so
it is only when this pattern helps to complete the pattern
BUILDING COMPLEX, and is itself completed by the pat-
terns BUILDING EDGE and GALLERY SURROUND, that it
becomes a courtyard at all.
31 2
T HE STRUCTURE OF A LANGUAGE
Each -pattern sits at the center of a network of connec-
tions which connect it to certain other patterns that
help to complete it.
Suppose we use a dot to stand for each pattern, and use
an arrow to stand for each connection between t wo pat-
terns. T h e n
A
B
means that the pattern A needs the pattern B as part of
it, in order for A to be compl ete; and that the pattern B
needs to be part of the pattern A, in order for B to be
complete.
If we make a picture of all the patterns whi ch are con-
nected to the pattern A, we see then that the pattern A
sits at the center of a whol e net work of patterns, some
above it, some below it.
Each pattern sits at the center of a similar net work.
And it is the network of these connections between
patterns which creates the language.
' i b
31 3
T HE GATE
Thus a l anguage for a garden mi ght have the structure
illustrated here, in whi ch each pattern has its place, con-
nected to the other patterns:
In this network, the links between the patterns are
almost as much a part of the language as the patterns
themselves.
Consider, for example, PRIVATE TERRACE ON THE
STREET and ENTRANCE TRANSITION.
So l ong as I imagine these t wo patterns as free-fl oati ng
entities, I can imagine an enormous variety of possible
houses, or gardens, which contain these patterns: and I
can imagine an enormous number of di fferent possible
relationships between these patterns.
But suppose, now, that they are connected in the l an-
guage, and that PRIVATE TERRACE ON THE STREET is a
part of ENTRANCE TRANSITION. Suddenly I imagine
people sipping drinks on the terrace, while arri vi ng guests
3 1 4
T HE STRUCTURE OF A LANGUAGE
pass the terrace, pass through the people who are sitting
there.
Or suppose, on the contrary, that I imagine PRIVATE
TERRACE ON THE STREET to be a part of the back of
INTIMACY GRADIENT instead. No w I have quite the op-
posite picture: people passing through a more somber,
more peaceful entrance transition, passing through the
house, and then comi ng out on another side, to the ter-
race, also on the street, but much more private, more
secluded.
Each time, my image of the individual patterns is
changed. And each time it intensifies them.
It is, indeed, the structure of the network which
makes sense of individual patterns, because it anchors
them, and helps make them complete.
Each pattern is modified by its position in the l anguage as
a whol e: accordi ng to the links whi ch form the l anguage.
And in virtue of its position in the whol e, each pattern
becomes especially intense, vivid, easy to visualize, and
more richly visualized. T h e l anguage not only connects
the patterns to each other, but helps them to come to life,
by gi vi ng each one a realistic context, and encouragi ng
imagination to give life to the combinations wf ych the
connected patterns generate.
But even when I have the patterns connected to one
another, in a network, so that they form a language,
how do I know if the language is a good oneP
31 5
T HE GATE
Is it complete? Should I add other patterns to it? Should I
take certain patterns out? Does it hang together? And,
above all: Wi l l it help me to generate a garden which
comes alive? We may assume that the patterns are in-
dividually alive if they follow the precepts of chapters 13
and 14. But what about the garden as a whol e?
Wi l l the l anguage al l ow me to create a wonderf ul , live
garden as a whol e? And how can I be sure of this?
The language is a good one, capable of making some-
thing whole, when it is morphologically and func-
tionally complete.
It is morphologically complete, when the patterns to-
gether form a complete structure, filled out in all its
details, with no gaps.
And it is functionally complete when the system of
patterns has that peculiar self-consistency in which the
patterns, as a system, generate only those forces which
they themselves resol veso that the system as a whol e,
can live, without the action of self-destroying inner con-
flicts.
The language is morphologically complete when I
can visualize the kind of buildings which it generates
very concretely.
It means that the general "species" of buildings whi ch
this l anguage specifies can be visualized, compl et el ynot
as a vague shadowy creature, ful l of gaps, but as a solid
31 6
T HE STRUCTURE OF A LANGUAGE
entity, whose only lack of clarity lies in the particulars.
Suppose, for instance, that I have the pattern for
HALF-HIDDEN GARDEN: but that I have no patterns
whi ch tell me what the maj or components of the garden
are. Concret el y: I have no idea how to f orm its edge; I
do not know what its main components are; I do not
know where there are any special focal points; I do not
know what happens where the garden meets the house.
Under these circumstances, I cannot really visualize the
garden at all: because I do not know enough about it.
Thi s is quite di fferent from not knowi ng how to con-
nect a particular garden with a particular house. It means
that there is a fundamental gap in my understanding of
how to put this kind of garden together: and it means that
when I get to this stage in a real case, I shall have to
scratch my head, and~start to think it out f rom the be-
ginning.
On the contrary, the l anguage for a garden is morpho-
logically complete when I can visualize the garden clearly,
as a global structure, even if I don' t yet have any specific
garden in mind.
And the language is functionally complete, when the
system of patterns it defines is fully capable of allow-
ing all its inner forces to resolve themselves.
Agai n, consider the garden. We know from chapters 6
and 7 that it is very possible that inconsistent systems of
conflicting forces can exist, and that these forces, when
not resolved internally, can gradual l y destroy the system.
31 7
T HE GATE
For instance, suppose that the garden makes no pro-
vision for the ecological interaction of trees, foundation
systems, and shade. T h e shade falls in bad places, systems
of roots start to undermine f oundat i onsand the entire
relation between building and trees is unstable, because
it causes progressive problems, whi ch wi l l , in the end,
force drastic changes on the garden, which may them-
selves then lead to other instabilities.
It is functionally complete only when all these internal
systems of forces are completely cover edi n short, when
there are enough patterns to bring all these forces into
equilibrium.
In both cases, the language is complete only when
every individual pattern in the language is complete.
For obviously, the l anguage cannot be complete as a
whol e, so l ong as any individual pattern is itself incom-
plete. Ever y pattern must have enough patterns " under "
it, to fill it out completely, morphologically. And every
pattern must also have enough patterns under it, to solve
the problems whi ch it generates.
Th u s the pattern BUILDING EDGE is morphologically
complete only when the patterns under it are enough to
create a full, developed, solid picture of the maj or global
structure of a building edge.
And it is functionally complete only when the patterns
under it, together, solve all the maj or problems, or sys-
tems of unresolved conflicting forces, which the existence
of a building edge creates.
31 8
T H E S T R UC T UR E OF A L A NGUA GE
We must therefore invent new patterns, whenever
necessary, to fill out each pattern which is not com-
plete.
T a k e BUILDING EDGE agai n, as an exampl e. T h e patterns
in the present l anguage whi ch lie di rectl y bel ow BUILD-
ING EDGE are SUNNY PLACE, GALLERY SURROUND, OUT-
DOOR ROOM, GARDEN SEAT, and ENTRANCE TRANSITION.
Ar e there any unresol ved probl ems al ong the bui l di ng
edge, in the presence of these f i ve patterns? T h i s is the
question of f unct i onal compl eteness. An d , are there any
parts of the bui l di ng edge whi ch are geomet ri cal l y un-
cl ear, and not adequat el y del i neated by these f i ve pat-
terns? T h i s is the question of morphol ogi cal compl eteness.
I t happens that the ans wer to both questions is " y e s . "
T h e r e is an unresol ved probl em. An d there is an uncl ear
geomet ri c z one. T h e r e is one part of a bui l di ng e d g e
the pl ace whe r e a l ong bl ank wal l simply runs al ong a
stretch of g a r d e n wh i c h is still probl emat i cal , because it
has no pattern cover i ng it. T h e bl ank wal l is unf r i endl y,
it wi l l create a part of the gar den, whi c h is uncomf or t abl e,
not easy to use. T h e pat t ern BUILDING EDGE itself, says, in
ver y gener al t erms, that the edge should be a defi ni te
pl ace, and that it should l ook both ways , both t owar ds the
inside of the house, and t owar ds the outside. But , in detail,
this question of the l ong bl ank wal l remai ns unsol ved. T o
solve it, it wi l l be necessary to i nvent some ne w pat t ern,
whi c h is part of BUILDING EDGE, and bel ow it in the
l anguage, and whi c h s omehow succeeds in s howi ng how
to handl e this l ong, bl ank, unf ri endl y wal l .
3
J
9
T H E GAT E
I do not know, j ust no w, exact l y wha t this pat t ern is.
But the exampl e helps to show cl earl y, that the f unct i onal
intuition whi c h tells me there is a " pr o bl e m" there, can-
not be separated f r om the morphol ogi cal i ntui ti on whi ch
tells me that there is a gap in the geomet r y.
W e keep wor ki ng at every pat t ern unti l we have a
col l ecti on of patterns bel ow it whi ch compl et el y resol ve
both f unct i onal and morphol ogi cal probl ems.
An d I compl et e the l anguage by creat i ng patterns,
and el i mi nat i ng patterns, until each pat t ern is compl et e.
But I must also make sure that the patterns below a
given pattern are its principal components.
T h e r e must not be too ma ny pat t erns underneat h a gi ven
pat t ern. Cons i der HALF-HIDDEN GARDEN this ti me.
T h e r e is one corner of the gar den whi c h wi l l be the
SUNNY PLACE; anot her pl ace wi l l be an OUTDOOR ROOM
perhaps; there is a need f or trees, f or mi ng a pl ace; there
is the charact er of the gar den as a whol e ; GARDEN GROW-
ING WILD; there is the rel ati on bet ween the gar den and
the street, in det ai l ; PRIVATE TERRACE ON THE STREET.
T h e r e is the possibility of the rel ati on bet ween the house
and the gar de n: covered perhaps by GREENHOUSE . . . .
T h e r e is the charact er of flowers in the g a r de n: perhaps
RAISED FLOWERS, and the need f or veget abl es and f rui t :
VEGETABLE GARDEN and FRUIT TREES . . . .
But not all of these need to be di rectl y underneat h
HALF-HIDDEN GARDEN. T h e reason is that some of t hem
embel l i sh each other. Fo r exampl e, GARDEN GROWING
320
T HE STRUCTURE OF A LANGUAGE
WILD, which gives the gardens global character, is itself
filled out and completed by RAISED FLOWERS and VEGE-
TABLE GARDEN; and TREE PLACES is itself filled out by
FRUIT TREES. These patterns whi ch can be "reached"
through another pattern, do not need to appear directly
below the pattern HALF-HIDDEN GARDEN.
It is essential to distinguish those patterns which are
the principal components of any given pattern, from
those which lie still further down.
If I have to make a HALF-HIDDEN GARDEN and I can
understand it as a thing which has three or four parts, I
can visualize it, and begin to create one, for myself, in my
own garden.
But if the HALF-HIDDEN GARDEN has twenty or thirty
patterns, all equally its parts, I will not be able to imagine
it coherently.
It turns out that there are just five patterns which do
have to appear immediately below HALF-HIDDEN GAR-
DEN. Th e y are GARDEN WALL, GARDEN GROWING WILD,
PRIVATE TERRACE ON THE STREET, SUNNY PLACE, and
TREE PLACES.
So, within this particular l anguage, these are the chief
"component s" of a HALF-HIDDEN GARDEN.
And this process of defining the principal components
of a given pattern is what finally completes it.
Ori gi nal l y we think, perhaps, that the main parts of a
" gar den" are l awns, and flower beds, and paths.
321
T HE GATE
But now, after careful consideration of the pattern
HALF-HIDDEN GARDEN, we begin to see that it has five
main components: GARDEN GROWING WILD, PRIVATE
TERRACE ON THE STREET, SUNNY PLACE, TREE PLACES,
and GARDEN WALL.
No w our whole functional and morphological under-
standing of the garden has changed. It is not only that we
now see the garden as made up of these particular five
new entities, whi ch changes our vision of the garden' s
f orm. Th e fact that these five patterns solve five particu-
lar problems also changes entirely, our vision of the gar-
den's function.
When every pattern has its principal components
given by the smaller patterns which lie immediately
below it in the languagey then the language is com-
plete.
And you see then what a beautiful structure a pattern
l anguage has.
Each pattern is itself a part of some l arger pat t erni t
is born out of these l arger patterns through the forces
which occur there, and the conditions which al l ow these
forces to be in harmony.
And each pattern itself gives birth to smaller patterns
whi ch, once again, through forces whi ch must also be in
harmony, gives birth to smaller patterns again created by
the conditions which put the l ower level forces into
harmony.
322
T HE STRUCTURE OF A LANGUAGE
Now we can see the full extent to which the design
of the garden lies within the language for the garden.
If you like the family of gardens whi ch your garden l an-
guage generates, then the l anguage is f i ne; but if the lan-
guage does not conj ure up an image of a marvelous place,
then there is something wr ong wi th it; and what is wr ong
can' t possibly be corrected later when you get to the pro-
cess of design. At that stage it is much too late.
We tend to imagine that the design of a building or a
garden takes a l ong ti me; and that the preparation for the
process of design is short. But in the process where the
l anguage plays its proper part, this gets reversed. Th e
preparation of the l anguage may take a very l ong time:
weeks, months, years. But the use of the l anguage, as we
shall see concretely in chapters 21 and 22 and 23, takes
no more than a f ew hours.
Essentially, this 'means that the language which you
have prepared must be judged as if it were itself a
finished garden [or a finished building).
Since the finished garden (or building) is controlled any-
way by just those patterns whi ch appear in it, then you
can tell, even before you use the l anguage, whether you
will like the places whi ch the l anguage will generate.
If the collection of patterns makes a coherent and satis-
f yi ng whol e, and requires no further insights, and no
further beauties, to complete them, then the l anguage is
3
2
3
T HE GATE
alright. But if you think of the l anguage merely as a
convenient tool, and imagine that the garden or the
building you create will become beautiful later, because of
the finesse with whi ch you handle it, but that the collec-
tion of patterns whi ch lie in the l anguage now are not
enough to make it beautiful, then there is something
deeply wr ong with the l anguage; and you must modi fy
it, until you are satisfied.
So, the real work of any -process of design lies in this
task of making up the language, from which you can
later generate the one particular design.
You must make the language first, because it is the
structure and the content of the language which
determine the design. The individual buildings
which you make, will live, or not, according to the
depth and wholeness of the language which you use
to make them with.
But of course, once you have it, this language is
general. If it has the power to make a single building
live, it can be used a thousand times, to make a
thousand buildings live.
3
2
4
CHAPTER 17
THE EVOLUTION OF A COMMON
LANGUAGE FOR A TOWN
Then finally, from separate languages
for cliferent building tasks, we can create
a larger structure still, a structure of
structures, evolving constantly, which is
the common language for a town. This
is the gate.
From chapter i6, we know how to construct an indi-
vidual language, for a specific individual building
type.
We shall see now, in this last chapter of part two,
how many of these languages can fit together, to be-
come the common language of a town.
Imagi ne, to begin wi th, that we have made a dozen lan-
guages for di fferent building tasks.
One for a house, one for a garden; one for a street;
one for a neighborhood, one for a wi ndow, one for an
office, a concert hall, one for a building with apartments
in it, one for an office building, one for a shop, one for a
public holy place, one for a river' s edge, one for a busy
crossroads in a t own.
As we make the different individual languages, we
find the patterns overlap.
For example, we find that ENTRANCE TRANSITION is part
of the l anguage for the garden, and also part of the l an-
guage for the house.
ROAD CROSSING is part of the l anguage for the street,
and also part of the l anguage for the busy crossroads.
ALCOVE is part of the l anguage for the house, and part
of the l anguage for the wor ks hopand, in some peculiar
outdoor version, perhaps also a part of the l anguage for
the river' s edge.
LIGHT ON TWO SIDES OF EVERY ROOM applies to every
habitable room, in every kind of bui l di ngso it is in al-
most all these l anguages.
3
2
9
T HE GATE
And, more subtly, we find also that different patterns
in different languages, have underlying similarities,
which suggest that they can be reformulated to make
them more general, and usable in a greater variety of
cases.
For exampl e, in the University of Or egon, we discovered
a pattern whi ch we called DEPARTMENT HEARTH. In our
work on clinics, we discovered a pattern which we called
TANGENT PATHS. In our wor k on houses in Peru, we dis-
covered a pattern called FAMILY ROOM CIRCULATION.
Al l these patterns had the same essential bundle of
relationships in common. Al l of them required a common
area at the heart of a social group, placed in such a way
that people's natural paths passed tangent to this common
area, every time that they moved in and out of the place.
So, it seemed natural to formul ate a deeper, and more
general , pattern which applied to all these di fferent
l anguages, which we called COMMON AREA AT THE
HEART.
Gradually it becomes clear that it is possible to con-
struct one much larger language, which contains all
the patterns from the individual languages, and uni-
fies them by tying them together in one larger struc-
ture.
Thi s larger l anguage is identical in structure to the
smaller languages. But in addition, it contains all these
smaller l anguages within it.
330
E V OL UT I ON OF A C O MMO N L A NGUA GE
A gr oup of us began to const ruct such a l anguage ei ght
or ni ne years ago. T o do it, we di scovered and wr ot e
do wn many hundreds of patterns. T h e n we di scarded
most of these patterns dur i ng the years, because we de-
cided t hat they wer e absur d; or because we f ound more
subtle f ormul at i ons of the same i deas; or because we f ound
that t hey di dn' t hold wat er e mpi r i c al l yor we noti ced
that t hey di dn' t account , empi ri cal l y, f or di f f erences be-
t ween the pl aces whe r e we fel t good and the places wher e
we fel t bad.
T h e 253 patterns whi c h remai n are j ust a handf ul
whi ch at this mome nt still seem val uabl e to us. T h e y f or m
the explicit l anguage, whi c h is the second vol ume of this
series, cal l ed A Pattern Language.
Our version of such a language begins with patterns
for the region ( j p i - 7 ) :
INDEPENDENT REGIONS, THE DISTRIBUTION OF TOWNS,
CITY COUNTRY FINGERS, AGRICULTURAL VALLEYS, LACE
OF COUNTRY STREETS, COUNTRY TOWNS, THE COUNTRY-
SIDE.
It has patterns for a town ($:82j) :
MOSAIC OF SUBCULTURES, SCATTERED WORK, MAGIC OF
THE CITY, LOCAL TRANSPORT AREAS, COMMUNITY OF
7000, SUBCULTURE BOUNDARY, IDENTIFIABLE NEIGH-
BORHOOD, NEIGHBORHOOD BOUNDARY, WEB OF PUBLIC
TRANSPORTATION, RING ROADS, NETWORK OF LEARN-
331
T H E G A T E
ING, WEB OF SHOPPING, MINI-BUSES, FOUR-STORY LI MI T,
NINE PER CENT PARKI NG, PARALLEL ROADS, SACRED
SITES, ACCESS TO WATER, LIFE CYCLE, MEN AND WOMEN.
It has patterns for communities and, neighborhoods
{#28-48):
ECCENTRI C NUCLEUS, DENSITY RINGS, ACTI VI TY NODES,
PROMENADE, SHOPPING STREET, NI GHT LI FE, I NTER-
CHANGE, HOUSEHOLD MIX, DEGREES OF PUBLICNESS,
HOUSE CLUSTER, ROW HOUSES, HOUSING HI LL, OLD
PEOPLE EVERYWHERE, WORK COMMUNI TY, INDUSTRIAL
RIBBON, UNIVERSITY AS A MARKETPLACE, LOCAL TOWN
HALL, NECKLACE OF COMMUNI TY PROJECTS, MARKET
OF MANY SHOPS, HEALTH CENTER, HOUSING IN BE-
TWEEN.
It has patterns for the public land inside a neighbor-
hood (#49-74):
LOOPED LOCAL ROADS, T JUNCTI ONS, GREEN STREETS,
NETWORK OF PATHS AND CARS, MAIN GATEWAYS, ROAD
CROSSING, RAISED WALK, BIKE PATHS AND RACKS, CHI L-
DREN IN THE CI TY, CARNI VAL, QUI ET BACKS, ACCESSIBLE
GREEN, SMALL PUBLI C SQUARES, HI GH PLACES, DANCING
IN THE STREET, POOLS AND STREAMS, BIRTH PLACES,
HOLY GROUND, COMMON LAND, CONNECTED PLAY,
PUBLI C OUTDOOR ROOM, GRAVE SITES, STILL WATER,
LOCAL SPORTS, ADVENTURE PLAYGROUND, ANIMALS.
332
E V O L U T I O N OF A C O MMO N L A N G U A G E
It has patterns for the private land and institutions in
the neighborhood ($175-94) :
THE FAMI LY, HOUSE FOR A SMALL FAMI LY, HOUSE FOR
A COUPLE, HOUSE FOR ONE PERSON, YOUR OWN HOME,
SELF-GOVERNING WORKSHOPS AND OFFICES, SMALL SER-
VICES WI THOUT RED TAPE, OFFICE CONNECTIONS,
MASTER AND APPRENTI CES, TEENAGE SOCIETY, SHOP-
FRONT SCHOOLS, CHILDREN' S HOME, I NDI VI DUALLY
OWNED SHOPS, STREET CAFE, CORNER GROCERY, BEER
HALL, TRAVELER' S I NN, BUS STOP, FOOD STANDS, SLEEP-
ING IN PUBLI C.
Patterns for the broad layout of the buildings in a
building complex (^95126) :
BUILDING COMPLEX, NUMBER OF STORIES, SHIELDED
PARKI NG, CI RCULATI ON REALMS, MAIN BUILDING,
PEDESTRIAN STREET, BUILDING THOROUGHFARE, FAM-
ILY OF ENTRANCES, SMALL PARKI NG LOTS, SITE REPAIR,
SOUTH FACING OUTDOORS, POSITIVE OUTDOOR SPACE,
WINGS OF LI GHT, CONNECTED BUILDINGS, LONG THI N
HOUSE, MAIN ENTRANCE, HALF- HI DDEN GARDEN, EN-
TRANCE TRANSITION, CAR CONNECTI ON, HI ERARCHY OF
OPEN SPACE, COURTYARDS WHI CH LIVE, CASCADE OF
ROOFS, SHELTERING ROOF, ROOF GARDEN, ARCADES,
PATHS AND GOALS, PATH SHAPE, BUILDING FRONTS,
PEDESTRIAN DENSITY, ACTI VI TY POCKETS, STAIR SEATS,
SOMETHING ROUGHLY IN THE MIDDLE.
333
T H E G A T E
Patterns for the building and its rooms (jp 127158):
I NTI MACY GRADIENT, INDOOR SUNLIGHT, COMMON AREAS
AT THE HEART, ENTRANCE ROOM, THE FLOW THROUGH
ROOMS, SHORT PASSAGES, STAIRCASE AS A STAGE, ZEN
VIEW, TAPESTRY OF LI GHT AND DARK, COUPLE' S REALM,
CHILDREN' S REALM, SLEEPING TO THE EAST, FARM-
HOUSE KI TCHEN, PRI VATE TERRACE ON THE STREET, A
ROOM OF ONE'S OWN, SEQUENCE OF SITTING SPACES, BED
CLUSTER, BATHI NG ROOM, BULK STORAGE, FLEXI BLE
OFFICE SPACE, COMMUNAL EATING, SMALL WORK
GROUPS, RECEPTION WELCOMES YOU, A PLACE TO WAI T,
SMALL MEETING ROOMS, HALF- PRI VATE OFFICE, ROOMS
TO RENT, TEENAGER' S COTTAGE, OLD AGE COTTAGE, SET-
TLED WORK, HOME WORKSHOP, OPEN STAIRS.
Patterns for the gardens and the -paths between the
buildings (# 159-178) :
LI GHT ON TWO SIDES OF EVERY ROOM, BUILDING EDGE,
SUNNY PLACE, NORTH FACE, OUTDOOR ROOM, STREET
WINDOWS, OPENING TO THE STREET, GALLERY SUR-
ROUND, SIX-FOOT BALCONY, CONNECTION TO THE
EARTH, TERRACED SLOPE, FRUI T TREES, TREE PLACES,
GARDEN GROWING WILD, GARDEN WALL, TRELLISED
WALK, GREENHOUSE, GARDEN SEAT, VEGETABLE GAR-
DEN, COMPOST.
Patterns for the smallest rooms and closets within the
rooms (jpr79204):
334
E V O L U T I O N OF A C O MMO N L A N G U A G E
ALCOVES, WINDOW PLACE, THE FIRE, EATING ATMO-
SPHERE, WORKSPACE ENCLOSURE, COOKING LAYOUT,
SITTING CI RCLE, COMMUNAL SLEEPING, MARRIAGE BED,
BED ALCOVE, DRESSING ROOM, CEILING HEI GHT VARI ETY,
THE SHAPE OF INDOOR SPACE, WINDOWS OVERLOOKING
LI FE, HALF- OPEN WALL, INTERIOR WINDOWS, STAIR-
CASE BAY, CORNER DOORS, THI CK WALLS, CLOSETS BE-
TWEEN ROOMS, SUNNY COUNTER, OPEN SHELVES, WAIST-
HIGH SHELF, BUI LT- I N SEATS, CHI LD CAVES, SECRET
PLACE.
Patterns for the overall configuration of construction
and materials (#205213):
STRUCTURE FOLLOWS SOCIAL SPACES, EFFI CI ENT STRUC-
TURE, GOOD MATERIALS, GRADUAL STIFFENING, ROOF
LAYOUT, FLOOR AND CEILING LAYOUT, THI CKENI NG THE
OUTER WALLS, COLUMNS AT THE CORNERS, FI NAL
COLUMN DISTRIBUTION.
Patterns for the details of construction (#214-232):
ROOT FOUNDATIONS, GROUND FLOOR SLAB, BOX COL-
UMNS, PERI METER BEAMS, WALL MEMBRANES, FLOOR-
CEILING VAULTS, ROOF VAULTS, NATURAL DOORS AND
WINDOWS, LOW SILL, DEEP REVEALS, LOW DOORWAY,
FRAMES AS THI CKENED EDGES, COLUMN PLACE, COL-
UMN CONNECTI ON, STAIR VAULT, DUCT SPACE, RADIANT
HEAT, DORMER WINDOWS, ROOF CAPS.
335
T H E GA T E
And the language finishes with patterns for details
and color and ornament ( ^ 2 3 3 - 2 5 3 ) :
FLOOR SURFACE, LAPPED OUTSIDE WALLS, SOFT INSIDE
WALLS, WINDOWS WHICH OPEN WIDE, SOLID DOORS WITH
GLASS, FILTERED LIGHT, SMALL PANES, HALF-INCH
TRIM, SEAT SPOTS, FRONT DOOR BENCH, SITTING WALL,
CANVAS ROOFS, RAISED FLOWERS, CLIMBING PLANTS,
PAVING WITH CRACKS BETWEEN THE STONES, SOFT TILE
AND BRICK, ORNAMENT, WARM COLORS, DIFFERENT
CHAIRS, POOLS OF LIGHT, THINGS FROM YOUR LIFE.
Such a language is, in principle, complex enough and
rich enough to be the language for a town.
I t covers all scales, all kinds of social institutions, all the
ma j o r ki nds of bui l di ngs, all the ma j o r ki nds of out door
space; and ways of bui l di ng whi c h are deep enough to be
used f or the f ul l vari ety of bui l di ngs in the t own.
But it is not yet fully living as a language.
Fi rst , to be l i vi ng as a l anguage, it must be the shared
vision of a gr oup of peopl e, ver y speci fi c to their cul t ure,
able to capt ure their hopes and dreams, cont ai ni ng ma ny
chi l dhood memori es, and special l ocal way s of doi ng
thi ngs.
T h e l anguage we have const ruct ed, and wr i t t en do wn,
is built on our o wn cul t ural knowl e dg e , of cour se: but it is
mor e abstract, mor e di f f use, and needs to be made con-
33
6
EVOLUTI ON OF A COMMON LANGUAGE
crete in a particular time and place, by local customs, local
climate, local ways of cooking food, local materials.
T o be the common l anguage of a people, and alive, it
has to contain much deeper s t uf f a vision of a way of
life, personal, able to make concrete people's feelings for
their parents and the past, able to connect them to a vision
of their future as a people, concrete in all their individual
particulars, the flowers which gr ow there in that place,
the winds which blow there, the kinds of factories there
are . . . .
And further, a living l anguage must be personal.
A language is a living language only when each per-
son in society, or in the town, has his own version of
this language.
For it is then not only an intellectual thing whi ch ex-
presses patterns as invariants, as rules to fol l ow, as knowl -
edge about what makes a building or a t own work right.
It is, a deeper thing, a felt thing, a thing lived through,
whi ch expresses people's innermost attitudes about their
way of life, their hopes and fears about the ways in which
they live and work t oget her a communal knowl edge of
a way of life that will be good for them.
To reach this deeper state, in which each person has a
pattern language in his mind as an expression of his
attitude to life, we cannot expect people just to copy
patterns from a book.
227
T HE GATE
I can tell you that ENTRANCE TRANSITION is a good pat-
tern. I can explain the problem to you; and define, in
great detail, the physical relationships whi ch make this
pattern. But you will not have the pattern in your own
mind, as part of your own pattern l anguage, until you
have yourself seen several entrances whi ch have this prop-
erty, seen how wonderf ul they are, compared them wi th
other entrances whi ch lack this property, and have then,
for yourself, invented your own abstraction whi ch pin-
points the difference between the entrances you like and
those you don' t.
A living language must constantly be re-created in
each -person's mind.
Even the ordinary l anguage in a person' s mind (Engl i sh,
French, what ever) is created by hi mi t is not learned.
Wh e n a baby "l earns" l anguage from his parents, or
f rom the people around him, he does not learn the rules
whi ch they have in their l anguagesbecause he cannot
see or hear the rules. He only hears the sentences which
they produce. Wh a t he does then, is to i nvent systems of
rules, for himself, rules whi ch are entirely invented, for
the first time, by him. He keeps changi ng these rules,
until wi th them, he can produce a l anguage similar to
the l anguage he hears. And at that stage, we say that he
has " l ear ned" the l anguage.
Of course, his rules are similar to his parents' rules, be-
cause they have to generate approximately the same kinds
of sentences. But in fact, the l anguage he has " l ear ned" is
a system of rules, entirely created by him, in his own
338
EVOLUTI ON OF A COMMON LANGUAGE
mind. And as he modifies his l anguage, and improves it,
deepens it, throughout his l i f ehe does it, al ways, by
creating, and i mprovi ng rules, whi ch he invents.
Just so with pattern languages.
Yo ur mind has the innate capacity to create a pattern lan-
guage. But its exact cont ent st he specific nature of the
patterns in your l anguagei s up to you. You must create
them for yourself.
Since your experience is never exactly the same as any-
one else's, the versions of the patterns which you create are
necessarily a little di fferent from the versions that each
other person finds out for himself.
Thi s does not in any way deny the fact that there are
objective, deep, invariant truths.
It merel y means that each person will, when he finds
out this truth for himself, wrest a slightly di fferent version
of this truth out for himself.
Then, as each person makes up his own language for
himself, the language begins to be a living one.
And just as there will be variations from person to person,
so too there will be even more noticeable variation from
culture to culture.
T h e variation from person to person happens because
di fferent people actually do have, to some extent, di fferent
forces in themselves, and in their lives, and they will
therefore experience di fferent configurations of forces as
l i fe-gi vi ng, or life-destroying.
339
T HE GATE
I n two di fferent cultures, the forces which exist are
even more di fferent, and there is even more opportunity
for people to experience di fferent systems of forces as
life-creating or life-destroying.
Two neighborhoods with different cultures will have
different collections of patterns in their languages.
Obvi ousl y, for instance, a Lat i n neighborhood will be
more likely to include the PROMENADE, in the l anguage
for their neighborhood, because they have the habit of
the eveni ng stroll. A culture whi ch values privacy will be
more likely to have COURTYARDS WHICH LIVE in their
houses, because they are secluded, than PRIVATE TERRACE
ON THE STREET, whi ch is more open to the street.
Different neighborhoods y just like different feofley
will quite often have different versions of the fat-
terns.
Suppose that many people have the pattern INTIMACY
GRADIENT in their l anguages. In its pure Peruvi an form,
the family has the most public room in front, the family
rooms further back, and the kitchen and bedrooms f ur-
thest f rom the street, in strict order.
In an Engl i sh neighborhood, the people may have
some version of this pattern in their l anguages: but modi-
fied, so that the kitchen comes closer to the front.
I n a neighborhood of workshops, this pattern makes
little sense in its original f orm. But even there, it may be
340
EVOLUTI ON OF A COMMON LANGUAGE
that there is some version of it, whi ch implies that the
workshops themselves have a front and a back, and that
the front part is more public, and the back more private.
And, in different neighborhoods, the -people may
have systematically different connections in their
languages.
For instance, in one neighborhood, people may have a
connection between COMMON AREAS AT THE HEART and
FARMHOUSE KITCHEN. Thi s means that for them, the
farmhouse kitchen is the center of the house, the place
where everybody comes and goes, the heart of social
life.
But , in another neighborhood, nearby, they have these
t wo patterns in their l anguage t oobut not connected.
For these people, their houses have a COMMON AREA AT
THE HEART, which is a kind of general comfortable mill-
ing space, near the f ront of the house: and the FARM-
HOUSE KITCHEN is a small, private area, towards the
back, where only close friends of the family meet.
We see then, that a language which is shared within a
town is a vast structure, far more complex than an
individual language.
Not merely a net work, but a network of networks, a
structure of structures, a vast pool of changi ng, varyi ng,
l anguages which people create for themselves as they
take on their di fferent building tasks.
341
T HE GATE
And once this kind of structure exists, we have a
living language in a town, in fust the same sense that
our common speech is living.
Thi nk about ordinary l anguage. Al l of us who speak
Engl i sh have a common l anguage, yet it is also true that
each of us has created his own l anguage for himself, in
his own brai nand that each of us has a l anguage which
is to some extent idiosyncratic. Tha t is why we can recog-
nize each person's favorite words, his style, his funny and
special ways of putting things. Ye t even though each of
us has his own l anguage, the overlap between our lan-
guages is enor mousand it is this whi ch gives us a com-
mon l anguage.
The same thing happens in genetics.
Each individual member of a given species has a slightly
di fferent set of genes on his chromosomes. If t wo indi-
viduals are closely related, there is enormous overlap in
the genesonl y a f ew are di fferent. As the overlap de-
creases, we say that the two individuals are members of
different subspecies within the species. And when the
overlap decreases below a certain threshold, we say that
they are members of altogether di fferent species.
The genetic character of a species is defined by its
gene pool.
Thi s gene pool is the collection of all the genes currently
342
EVOLUTI ON OF A COMMON LANGUAGE
possessed by all the individuals in the species. Some of
these genes are much more common than others. Th e
genes whi ch are most common define the shared character
of the speciesthe ones whi ch are less common, define in-
dividual families and strains.
Since every new individual in the species gets a com-
bination of genes taken from the gene pool (except for
very occasional mutati ons), the overall statistical distribu-
tion of genes in the gene pool stays roughly constant, but
drifts as evolution makes some genes die out and others
multiply.
Just so, a common pattern language is defined by a
pool of patterns.
Suppose that every person in society has his own personal
pattern l anguage. No w imagine the collection of all the
patterns which anyone has in his l anguage. Cal l this col-
lection of patterns the pattern pool. Some of the patterns
will occur much more often in the pattern pool than
others. T h e ones which occur most often are the ones
whi ch are shared by everyone. T h e ones whi ch occur
less often are shared by f ewer peopl ethese will probably
be peculiar to some subculture within society. And the
ones whi ch occur least often of all, these are the purely
personal patterns, which represent individual people's
idiosyncracies.
The common pattern l anguage is not any one l anguage
which any one person has in his mi ndi t is defined by
the overall distribution of patterns in the pattern pool.
343
T HE GATE
And, once people share a language in this way, the
language will begin evolving of its own accord.
Once there is a pool of patterns, and thousands of people
are taking patterns from this pool, and using them, ex-
changi ng them, replacing them, it is certain that, simply
by itself, this l anguage will evolve.
As good patterns get shared more wi del y, and bad
patterns die out, the pattern pool will gradually contain
more and more good pat t ernsand in this sense we may
say that a common l anguage is evol vi ng, and becoming
better, even at the same time that we recognize that each
person will al ways have his own personal l anguage, which
is just one versi ona unique versi onof this common
l anguage.
In this sense, even though any one person' s pattern
l anguage will al ways be unique, the overall collection of
l anguages in society will gradually drift towards a com-
mon l anguage, represented by the overall character of the
pool of patterns.
The language will evolve, because it can evolve piece-
meal, one pattern at a time.
Geneti c evolution can happen only because the genes can
mutate independently. T h e genes are independent enough,
so that a new species can evolve by a process in which one
gene changes at a time. If it were not for this, the
evolution of a complex organism woul d never have been
possible.
344
EVOLUTI ON OF A COMMON LANGUAGE
Th e key to the i mprovement of patterns is also in the
fact that it can be piecemeal. Suppose the l anguage you
have now has 100 patterns. Since the patterns are inde-
pendent, then you can change one at a time, and they
can al ways get better, because you can al ways improve
each pattern, individually. ( I f the patterns were linked,
so that as you improved one pattern, you woul d also have
to change 50 others, the system woul d be unstable, and
you could never improve it cumul ati vel y. )
Thi s means that we can define, discuss, criticize, and
improve one pattern at a time: so that we never have to
throw away all the other patterns in a l anguage just be-
cause one of them is faulty. And any pattern, created by
anyone, can fit into any pattern l anguage. In short, as
soon as someone defines a really good pattern, it can
spread and become part of all the pattern l anguages in
the worl d, wi thout regard for the other patterns whi ch
these millions of di fferent l anguages contain.
It is this one simple fact, whi ch guarantees that the
evolution of pattern l anguages will be cumulative.
As feofle exchange ideas about the environment, and
exchange patterns, the overall inventory of fatterns
in the pattern fool keefs changing.
Some patterns drop out altogether, some become rare,
some patterns multiply, some new patterns enter the
pool. Since there are criteria for deciding whi ch patterns
are good, and whi ch ones are bad, people will copy good
patterns when they see them, and won' t copy bad ones.
345
T HE GATE
Thi s means that good patterns will multiply and become
more common, while bad patterns will become rarer, and
will gradual l y drop out altogether.
Graduallyy as people modify these published lan-
guagesy add to them, erase from themy a pool of
common languages y unique to different placesy unique
to individualsy and yet broadly shared y will evolve
of its own accord.
First, the good patterns will persist; the bad ones will
drop out.
Second, since better patterns will persist, and worse
ones will drop out, the l anguages wi l l become more com-
mon. In each area, a common l anguage will evolve.
And third, a natural differentiation will occur, in
whi ch each t own, each region, each culture, adopts a
di fferent set of pat t ernsso that the great stock of pat-
tern l anguages across the earth will gradual l y get di ffer-
entiated.
Of coursey this evolution will never end.
Al t hough the process of evolution will al ways move to-
wards greater depth and greater wholeness, there is no
end to i t t her e is no static perfect l anguage, whi ch, once
defined, wi l l stay defined forever. No l anguage is ever
finished.
Th e reason is this. Each l anguage specifies a certain
structure for some envi ronment. Onc e realized, in prac-
346
EVOLUTI ON OF A COMMON LANGUAGE
tice, the very existence of that structure will create new
forces, whi ch are born for the first time, out of that
st ruct ureand these new forces will, of course, create
new problems, new conflicts, that need to be resolved by
new pat t ernswhi ch, when added to our l anguages,
will create still newer forces once again.
Thi s is the eternal cycle of development. The r e is no
hope of stilling it, and no need either. We must simply
accept the fact that in the process of evolution, there is
no final equilibrium. The r e are passing phases whi ch
approach equi l i bri umbut that is all. T h e search for
equilibrium, the brush in the dark wi th a moment of
stability, the wave whi ch hesitates a moment before it
crashes into the sea agai nt hat is the closest constancy
will ever come to being satisfied.
Yet, changing as it is, each language is a living pic-
ture of a culture, and a way of life.
T h e patterns it contains, widely shared, reflect a common
understanding about attitudes to life, about the ways that
people want to live, the way they want to rear their chil-
dren, the way they want to eat their meals, the way they
want to live in families, the way they want to move from
place to place, the way they work, the way they make
their buildings look towards the light, their feelings about
water, above all, their attitudes towards themselves.
It is a tapestry of life, which shows, in the relation-
ships among the patterns, how the various parts of
347
T HE GATE
life can fit togethery and how they can make sensey
concretely in space.
And, above all, it is not just a passive picture. It has
power in it. It is a l anguage, active, powerf ul , whi ch has
the power to let men transform themselves, and their
surroundings.
Imagi ne that one day millions of people are using
pattern l anguages, and maki ng them again. Wo n ' t it i m-
press itself then, as extraordinary, that these poems
whi ch they exchange, this giant tapestry of images, which
they create, is comi ng alive before their eyes. Wi l l it be
possible then, for people to say stonily, that poems are
not real, and that patterns are nothi ng but i mages: when,
in fact, the worl d of images controls the worl d of matter.
In early times the city itself was intended as an
image of the universeits form a guarantee of the
connection between the heavens and the earthy a
picture of a whole and coherent way of life.
A living pattern language is even more. It shows
each person his connection to the world in terms so
powerful that he can re-affirm it daily by using it to
create new life in all the places round about him.
And in this sensey finallyy as we shall seey the living
language is a gate.
348
T H E W A Y
Once we have built the gate, we can
pass through it to the practice of the
timeless way.
CHAPTER I 8
THE GENETIC POWER OF LANGUAGE
Now we shall begin to see in detail how
the rich and com-plex order of a town can
grow from thousands of creative acts.
For once we have a common pattern
language in our town
y
we shall all have
the fower to make our streets and build-
ings live
y
through our most ordinary
acts. The language
y
like a seed
y
is the
genetic system which gives our millions
of small acts the fower to form a whole.
Assumey to start withy that some version of the fat-
tern language has been adopted in a towny or in a
neighbor hoody or by a group of people or a family
who adopt it as the basis for the reconstruction of their
world.
What is the relation between this common pattern
language and the constant process of construction and
destruction which gives the town its shape?
Recogni ze, first of all, that each person in the t own has
the capacity to shape his own surroundings.
A f armer in a traditional culture " k ne w" how to
make a beautiful house for himself. We envy him, and
think that only he was able to do this because his culture
made it possible. But this power the primitive f armer had
lay in his pattern l anguage.
And if the people of the town now have a pattern lan-
guage which is whol e, they have the same power, exactly.
What ever act of building or repair is cont empl at ed
building a bench, a flower bed, a room, a terrace, a small
cottage, a whol e house, a group of houses, a remaki ng of
the street, a shop, a cafe trellis, a compl ex of public build-
ings, even the repl anning of a nei ghborhoodt hey have
the power to do it for themselves.
A person with a pattern language can design any
part of the environment.
He does not need to be an " exper t . " T h e expertise is in
the l anguage. He can equally wel l contribute to the
353
T HE WAY
planning of a city, design his own house, or remodel a
single room, because in each case he knows the relevant
patterns, knows how to combine them, and knows how
the particular piece he is worki ng on fits into the larger
whol e.
And it is essential that the people do shape their
surroundings for themselves.
A town is a living thing. Its patterns are both patterns of
action and patterns of space. And in the process of maki ng
itself, it is the patterns of activity and space, not space
alone, which are continuously built, and destroyed, and
rebuilt. For this reason, it is essential, once again, that
people do it for themselves.
If the t own' s patterns lay merely in its bricks and mor-
tar, you mi ght argue that these bricks and mortar could
be shaped by anyone.
But since the patterns are patterns of action, and the
action will not happen unless the patterns are felt, and
created, and maintained by the people whose action goes
into the patterns, there is no way the living t own can be
built by professionals, for other people to live in. T h e
living t own can only be created by a process in whi ch
patterns are created and maintained by the people who
are a part of them.
This means, theny that the growth and rebirth of a
living town is built up from a myriad of smaller acts.
354
T HE GENETI C POWER OF LANGUAGE
In a t own where the common l anguage has vanished, the
acts of construction and design are in f ew hands, and are
large and clumsy.
But once each person in the t own can shape a building
for himself, or a part of the street, or help to shape a
public building, or add a garden or a terrace to a corner
of a bui l di ngt hen, at this stage, the growt h and rebirth
of the town is the concrescence of a million acts.
It is a flux of millions upon millions of these tiny
actsy each one in the hands of the person who knows
it best} best able to adapt it to the local circumstances.
Wi t hi n this flux, the people of the t own are constantly
building, and rebuilding, tearing down, maintaining,
modi f yi ng, changi ng, and building again.
A room, a building, or a neighborhood is not made by
a single act of building, in a single day. It is the temporary
end-result of a thousand di fferent acts, extended over
time, and done by unrelated people.
But what guarantee is there that this fluxy with all its
individual actsy will not create a chaos?
How does the pattern l anguage, whi ch exists behind this
flux, steer it, and enter into it?
It hinges on the close relationship between the process
of creation and the process of repair.
355
T HE WAY
Wh e n an organism grows f rom a seed, the process of
its growt h is guided by the genetic materials. Each cell
contains the D N A , and every cell is able to take on its
part within the whol e, by f ol l owi ng the gr owt h process
which the genetic stuff defines. Somehow, because each
cell contains the same material, the cells together, gr owi ng
independently, create a whol e whi ch is complete.
And once the organism is gr own the same genetic
process guides the process of repair. If I cut mysel f, the-
same genetic process whi ch originally created me now
takes charge of the smaller process of healing the cut, and
guarantees that all the cells around the cut cooperate to
f orm a whol e again.
In fact, there is no difference between the way the
genes control the process of genetic growt h whi ch forms
the embryo, and the process of repair whi ch heals the cut.
Th e genes operate continuously, every day, and every
moment.
An organism, which seems at first sight like a static
thing, is in fact a constant flux of processes.
Cel l s are born, and die, unceasingly. Th e organism whi ch
exists today is made of di fferent materials from the
organism of yesterday. It preserves those broad invariants,
which define its character, within the flux. Ye t even
these are changi ng sl owl y, over time. So, what there is,
in fact, is a perpetual fl ux of growt h and decay, in which
the "organi sm" is not so much an object, but the char-
acter of the invariants behind the flux, whi ch is reborn
and reshaped every day.
356
T HE GENETI C POWER OF LANGUAGE
A town or building also is a constant flux of processes.
If we visit London or Ne w Yor k today, it is a di fferent
thing from what it was five years ago. As in an organism
there is a process goi ng on which shapes new buildings
constantly, destroys the old, replaces and rebuilds and
modifies the fabric.
But again, just as in an organism, there is also some-
thing which remains the samet here is an invariant con-
tinuity behind the flux, a character, a " t hi ng, " a "struc-
t ure, " which remains the same.
And it is the pattern language which, like the genes
distributed throughout the cells, makes certain that
there is this structure, this invariant permanency, in
the flux of things, so that the building or the town
stays whole.
We know, already, that the common pattern l anguage
in a t own or a communi ty defines the fundamental situa-
tions, the archetypal moments, the components of the
way of life which people want to lead.
Wh a t we shall now see is the way this pattern lan-
guage, once it is whol e, and widely spread, also maintains
the slow pulsating growt h and death of situations, build-
ings, moments, places, in the history of the t own.
Imagine the constant process of creation which is
happening in a town.
Roads being wi dened; roads closed; markets being built;
new houses; old houses rebuilt; public buildings used for
357
T HE WAY
offices; a park made on the corner of a bl ock; people
dancing and eating in the streets; street vendors catering
to them; a seat is made f rom whi ch to wat ch the street;
a girl sews cushions for her favorite corner; an orchard
blooms; old people take their canvas chairs and sit out
underneath the blossoms; a new hotel begins; a f arm-
house gets torn down; a corner where a bus stop was
becomes a place where people speak in public; the new
hotel creates a need for taxis, and the taxi company
builds a taxi rank. . . .
All this is guided by the fact that every act which
helps to shape the buildings and the towny and their
activitiesy is governed by the pattern language people
shareand governedy above ally by just that portion
of the language which is especially relevant to that
especial act.
I n short, that flux in whi ch the t own gives birth to new
activities, maintains its old ones, modifies them, changes
them, is guided by the common pattern l anguage, just as
the slow flux of a flower, while it is alive, is governed by
the seed within it.
Ho w does the l anguage do it?
Each concrete building problem has a language. The
town as an entirety has a language. And each small
building task within the town has its own language.
T h e largest possible pattern l anguage covers the t own. It
contains, as sublanguages, l anguages for the di fferent cul-
358
T H E G E NE T I C P OWER OF L A NGUA GE
tures and subcul t ures in the t own. The s e cont ai n the
part i cul ar subl anguages f or part i cul ar cl i mates or l ocal
condi ti ons, and these in t urn cont ai n the l anguages f or
i ndi vi dual nei ghborhoods. Th e s e cont ai n the l anguages
f or di f f er ent ki nds of bui l di ngs, and f or the i ndi vi dual
bui l di ng whi c h is to be built on any i ndi vi dual s i t e a nd
these once agai n cont ai n as smal l er subl anguages the
l anguages whi ch di f f er ent fami l i es, or i ndi vi dual s, mi ght
requi re f or t hemsel ves, to build thei r rooms, and gar dens,
and the vari ous i ndi vi dual little corners of the gr eat er
bui l di ngs. . . .
Her e is a l anguage f or a wi ndo w seat :
ZEN VIEW
WINDOW PLACE
BUILT-IN SEAT
FRAMES AS THICKENED EDGES
DEEP REVEALS
WINDOWS WHICH OPEN WIDE
SMALL PANES
FILTERED LIGHT
Her e is a l anguage f or a house:
THE FAMILY
HALF-HIDDEN GARDEN
TREE PLACES
WINGS OF LIGHT
MAIN ENTRANCE
INTIMACY GRADIENT
COMMON AREAS AT THE HEART
SLEEPING TO THE EAST
COURTYARDS WHICH LIVE
INDOOR SUNLIGHT
359
T H E W A Y
FARMHOUSE KI TCHEN
SEQUENCE OF SITTING SPACES
COUPLE' S REALM
HOME WORKSHOP
He r e is a l a ng ua g e f or r epai r i ng c o mmo n l and inside a
ne i ghbor hood:
ACTIVITY NODES
NETWORK OF PATHS AND CARS
GREEN STREETS
MAIN GATEWAYS
ACCESSIBLE GREEN
SMALL PUBLI C SQUARE
HIGH PLACES
DANCING IN THE STREET
HOLY GROUND
PUBLI C OUTDOOR ROOM
STREET CAFE
He r e is a l a ng ua g e f or det er mi ni ng the bounda r y of a
ci t y' s g r o wt h :
CITY COUNTRY FINGERS
AGRI CULTURAL VALLEYS
LACE OF COUNTRY STREETS
LOCAL TRANSPORT AREAS
RING ROADS
SACRED SITES
Every act of building, brings a handful of patterns
into existence.
I n a t o wn wi t h a c o mmo n l a ng ua g e , this ha ndf ul is al -
wa y s chosen f r om the same f e w hundr e d pat t er ns whi c h
3 60
T HE GENETI C POWER OF LANGUAGE
define the t owns o, as di fferent acts of building o c c ur
even though they may be far apart, or physically uncon-
nectedsti l l , gradually the same f ew hundred patterns
get created, over and over agai nand give the t own its
coherence as a fabric.
And we see, then, how each act of building, because
its -pattern language is part of the larger language for
the town, contributes to the larger process which cre-
ates the town.
Just as each gene, or group of genes within a chromo-
some, guide the growt h and repair of individual portions
of the organism, so in a town each sublanguage of the
common l anguage also guarantees the complete and
coherent emergi ng organization of the whol e.
As in the organism, there is no sharp difference be-
tween the process of construction and the process of
repair. Each process of construction helps repair some
l arger whol e, of which it is merely a part. No thing is
whole unto itself.
And the larger pattern l anguage whi ch is shared, lies
behind the flux of acts of building and repair, and makes
sure that there is a structure, an invariant permanency in
the flux of things, whi ch makes the t own stay whol e.
But the process which integrates the millions of small
acts, and makes them one, is not merely given by
the fact that all these acts are guided by parts of one
great language.
351
T HE WAY
Th e common l anguage has an integrative power which
goes far beyond this simple coordination. Thi s happens,
essentially, because each pattern is connected through the
network of the l anguage to every other pattern in the
l anguage. And this simple structural fact reflects the
more important fact that every act of building can also
go beyond its own limits, and help the other patterns in
the t own to gr ow.
Each pattern language in the larger language, can,
because it is connected to the entire language, help all
the other patterns to emerge.
Remember how each pattern in a l anguage is linked to
the patterns above it and below it. So, for example, the
pattern of PRIVATE TERRACE ON THE STREET helps
to complete the l arger patterns of the streetGREEN
STREETS, HIERARCHY OF OPEN SPACE, and COMMON
LAND. And it, in turn, is then completed by the smaller
patterns whi ch lie below it in the l anguage: OUTDOOR
ROOM, HALF-OPEN WALL, SOFT TILE AND BRICK.
Whe n we use a common -l anguage to build a pattern
into our worl d, we automatically build these l arger and
these smaller patterns too. As we build a PRIVATE TER-
RACE on the street, within the f ramework of a common
l anguage, we also try to take care to place it in a way
which helps to form the HIERARCHY OF OPEN SPACE out
on the street, which helps to f orm a GREEN STREET; and
we also complete it with the shelter of an OUTDOOR
ROOMperhaps a trellis, or a row of columns to enclose
362
T HE GENETI C POWER OF LANGUAGE
it; with a HALF-OPEN WALL whi ch helps to make the
half connection to the street; and tiles or bricks or wooden
surfaces, whi ch gradual l y wear, and show the marks of
use, and let small plants gr ow in between.
Each language tugs at the fabric of the larger lan-
guagey fulls with it other larger fatternsy and in this
fashion then helfs to re fair the larger whole.
Thus , within the l arger l anguage, it is impossible for any
act not to help to repair the l arger whol e. It is impossible
for any act of building to remain an isolated act: it al ways
becomes a portion of the flux of acts whi ch is helping to
maintain the whol e.
Th e neighborhood group who take steps to improve
their common land and streets will also help to generate
the l arger patterns for the traffic and the density and
shopping areas in the l arger town.
A man who builds a house, and has a l anguage for a
house, will also help to build the l arger street outside
his house, generate the patterns whi ch form the street
outside his house.
A child who helps to shape his room will also help to
generate the l arger patterns for the stairway and the
common space outside his room.
Even the l aying of a brick, to mend a wal l , wi l l not
only be used to mend that wal l , but will be used to help
repair the seat, the terrace, or the fireplace whi ch that
wal l helps to f orm.
363
T HE WAY
And the pattern language is the instrument by means
of which the flux which is the town perpetuates itselfy
maintains its structurey and keeps itself continuously
alive.
It guides the acts of all the individuals there in such a
way that every act of building, and each smaller act
whi ch seems more modest too, is guided by the patterns
in the l anguage whi ch are necessary to it, and gradual l y
generates these patterns, day by day, continuously, so
that this place is kept alive, continuously, by the gradual
process of creation and destruction. It is not the end
product of this process whi ch is alive, but the incessant
flux itself. The r e is no product of this process: the build-
ings and the town, whi ch live, are that incessant fl ux,
whi ch, guided by its l anguage, constantly creates itself.
We see then the enormous power which a common
pattern language has.
The process of life is marked by the continuous
creation of wholes from parts. In an organismy cells
cooperate to form organs and the body as a whole. In
a society the individual actions of the people cooperate
to form institutions and larger wholes. . . .
And in a town a pattern language is a source of
lifey above ally because it helps to generate the wholesy
from the cooperation of the individual acts.
364
CHAPTER 19
DIFFERENTIATING SPACE
Within this frocess
y
every individual
act of building is a fro cess in which sface
gets differentiated. It is not a frocess
of addition
y
in which fre-formed farts
are combined to create a whole: but a
frocess of unfolding
y
like the evolution
of an embryOy in which the whole fre-
cedes its farts
y
and actually gives birth
to them
y
by s flit ting.
Consider now a single act of building.
As we have seen, there is some language, part of
the larger language, which is specific to this act of
building, and governs it to give it order.
But how, exactly, does this language work.
I n chapters 21 and 22 we shall go through some specific
examples, whi ch show exactly how the process works to
help a person lay out the plan of a building. And then, in
chapter 23 we shall see how the building, once it is laid
out, can then be built, in a way which continues the
process of development.
However , we shall not understand these examples fully
until we first understand t wo general points about the
way a pattern l anguage works. In this chapter we shall
learn the importance of the order of the patterns in the
l anguage. In the next we shall learn the importance of
the intensity of each pattern in the sequence.
Start by remembering the fundamental truth about
the parts of any system which is alive.
Each part is slightly di fferent, accordi ng to its position in
the whol e. Each branch of a tree has a slightly di fferent
shape, accordi ng to its position in the l arger tree. Each
leaf on the branch is given its detailed f orm by its position
on the branch.
And they must have this character, because, to be
alive, they must be patterned properl y: and this means
that they must contain hundreds of di fferent patterns,
367
DI FFERENTI ATI NG SPACE
overlapped and i nterwoven at the same t i meand this
multiple patterning can only be contained in the geometry
of nature.
Ask yourself, then, what kind of process can create a
building or a flace that has this character.
Wh a t kind of process will al l ow us to cram a hundred
patterns together, in a limited space?
Or , more concret el yrememberi ng from our discus-
sion in chapter 8 that a geometry which does contain hun-
dreds of interacting patterns will al ways make every part
uni quewe may ask, again, what kind of process can
create a thing in whi ch every part is slightly di fferent.
Design is often thought of as a -process of synthesis,
a frocess of -putting together things, a process of
combination.
Accordi ng to this vi ew, a whol e is created by putting
together parts. Th e parts come first: and the f orm of the
whol e comes second.
But it is impossible to form anything which has the
character of nature by adding preformed parts.
Wh e n parts are modular and made before the whol e, by
definition then, they are identical, and it is impossible for
every part to be unique, accordi ng to its position in the
whol e.
368
DI FFERENTI ATI NG SPACE
Even more important, it simply is not possible for any
combination of modul ar parts to contain the number of
patterns whi ch must be present simultaneously in a place
which is alive.
It is only possible to make a place which, is alive by a
process in which each part is modified by its position
in the whole.
Imagi ne, for the sake of argument , that a certain FARM-
HOUSE KITCHEN consists, in essence, of a large kitchen
table, with a stove and counters round it. And suppose,
now, that in one corner there is to be an alcove. It is easy
to see that this alcove may have to be a slightly unusual
al cove; it may have to be part of the kitchen counter, or
connect up with it in some way, and perhaps have a cer-
tain specific relation even with the kitchen table.
And suppose, further, that this slightly unusual alcove
is now to contain a WINDOW PLACE and a WINDOWSILL,
a place to sit, a l ow sill. Of course, again, this wi ndow
place may have to take its character from the particular
shape which the ALCOVE has; it may have less light,
perhaps, because it is in the only corner not taken up by
cooking things; it may have a higher ceiling than usual,
because the kitchen needs more ventilation; it may have a
tiled floor, or tiles on the wal l , to counteract the steam
and water in the kitchen.
In short, each part is given its specific form by its exis-
tence in the context of the larger whol e.
369
T HE WAY
This is a differentiating process.
It vi ews design as a sequence of acts of compl exification;
structure is injected into the whol e by operating on the
whol e and crinkling it, not by adding little parts to one
another. I n the process of differentiation, the whol e gives
birth to its parts: the parts appear as folds in a cloth of
three dimensional space which is gradually crinkled. Th e
form of the whol e, and the parts, come into being
simultaneously.
The image of the differentiating process is the growth
of an embryo.
It starts as a single cell. Th e cell gr ows into a ball of cells.
The n, through a series of differentiations, each building
on the last, the structure becomes more and more com-
plex, until a finished human being is formed.
T h e first thing that happens is that this ball gets an in-
side, a middle layer, and an outside: the endoderm, meso-
derm, and ectoderm, which will later turn into skeleton,
flesh, and skin, respectively.
Th e n this ball of cells with three layers gets an axis.
T h e axis is laid down in the endoderm, and will become
the spine of the finished person.
370
DI F F E R E NT I A T I NG S PACE
T h e n this bal l , wi t h an axis, gets a head at one end.
Lat e r , the secondary structures, eyes, limbs, devel op in
rel ati on to the spinal axis and the head.
A n d so on. A t ever y stage of devel opment , ne w st ruc-
ture is laid down, on the basis of the st ruct ure whi ch has
been laid d o wn so f ar. T h e process of devel opment is, in
essence, a sequence of operati ons, each one of whi c h di f -
f erent i at es the st ruct ure whi c h has been laid d o wn by the
previ ous operati ons.
3 7 1
T HE WAY
The unfolding of a design in the mind of its creator,
under the influence of language, is just the same.
A l anguage al l ows you to generate an image of a building
in your mind, by placing patterns in space, one pattern at a
time.
At the beginning of a design process, you may have an
idea that the open space should be "more or less over
here, " and the building "more or less over there. " Neither
the pattern for "open space" nor the pattern for "bui l d-
i ng" is very precisely defined at this stage. The y are like
t wo clouds, whose size is imprecise, and with imprecise
edges. It is not even perfectly certain, at this stage, that the
cloud called "open space" will be entirely opennor that
the cloud called building will be entirely roofed. Wh a t is
happening, is that you place these t wo clouds, roughl y, at
this stage of the design, with the full understanding that
the design is accurate only to within the order of magni -
tude of the clouds themselves, and that all kinds of details
whi ch are smaller in scale, may be changed later.
Lat er in the process, you may be placing the "ent rance"
to the building. Agai n, the pattern which you call the en-
trance is a cloudy vol ume, about the right size, clear
enough so that you can pin point its location, wi th respect
to other l arger clouds, and to show its relations to the
things next to it, but no more exact than that.
And, yet another stage in the design process, you may
place a col umn. Thi s col umn has a height, and a rough
si zebut again, at the time you place it first, it has little
more. Lat er, you make the column more exact, by placing
372
DI FFERENTI ATI NG SPACE
the edges of the col umn, its reinforcing bars, its founda-
tion, and so on.
Whenever we want to make one of these vague cloudy
patterns more precise, we do it by placing other smaller
patterns, whi ch define its edge and interior.
Each pattern is an operator which differentiates space:
that isy it creates distinctions where no distinction was
before.
T h e operator is concrete and specific, insofar as it will al-
ways generate an instance of the pattern.
But the operator is quite general, because it specifies the
operation in such a way that its performance interacts with
the surroundings, to make a particular embodiment of the
pattern whi ch is unique.
And in the l anguage the operations are arranged in
sequence: so that, as they are done, one after another,
gradual l y a complete thing is born, general in the sense
that it shared its patterns with other comparable things;
specific in the sense that it is unique, according to its cir-
cumstances.
The language is a sequence of these operatorsy in
which each one further differentiates the image which
is the product of the previous differentiations.
Since the patterns are arranged in order of their morpho-
logical importance, the use of the l anguage guarantees that
a whol e is successively differentiated, so that smaller and
373
T HE WAY
smaller whol es appear in it, as a result of the distinctions
whi ch are drawn.
Whe n a pattern l anguage is properly used, it allows the
person who uses it to make places whi ch are a part of na-
ture, because the successive acts of differentiation whi ch
the patterns define, are ordered in such a way that at each
step new whol es are born, infinitely various because they
are adapted to the l arger whol es in which they sit, and
with the parts between the whol es themselves whol e, be-
cause the acts of differentiation make them so.
Here is a simple example of a balcony whose shape
grew by a differentiating process.
My house has a bay wi ndow looking out into a group of
pine trees. I decided to build a balcony out there, six
feet above the ground, at the same level as the living
room itself. Here is the sequence of decisions whi ch gave
birth to the design.
TREE PLACES: I decided to use an old pine tree as the
right-hand corner post of the extension. It is a beautiful
old tree, whi ch, as the corner of the balcony, spreads its
branches to f orm a natural umbrella.
6
374
D I F F E R E NT I A T I NG S PACE
SUNNY PLACE: I put the l ef t - hand post as f ar as it coul d
be to the l ef t , so as to catch a corner of sunshi ne that re-
peat edl y fal l s on that spot dur i ng the day.
STRUCTURE FOLLOWS SOCIAL SPACES: I deci ded to place
t wo ot her posts, not one, so that there woul d be t wo cor-
ner spaces creat ed, each a usef ul social space, wi t h a di-
amet er of about 5 f eet. On e c ol umn woul d have split
the t hi ng, and made no spaces a ny whe r e , since the corners
woul d be too l ar ge.
COLUMNS AT THE CORNERS: I wa nt e d the corners to be
l arge enough to f unct i on as social spaces, and t heref ore
l ar ger t han the cent er span. O n the other hand, since
these col umns woul d def i ne the positions of i nt ermedi at e
beams, I want ed to make the cent ral span as l ar ge as it
coul d b e a n d not wast e it by maki ng it too na r r ow. As
a resul t of these del i berati ons, I' f i nal l y settl ed on 5J/2-
3"5/
/
2- I f the post had been equal l y spaced, at modul ar
i nt erval s, the corners woul d have been too smal l to live.
PERIMETER BEAM: T h e r i ght - hand beam ran at an
angl e, so that it coul d run di rectl y i nto the tree.
375
T HE WAY
FLOOR SURFACE: T h e planks were cut, as needed, to
fit the overall shape of the balcony.
Imagine what this balcony would have been like if I
had tried to make it up from modular components.
Say, for the sake of argument, that there were some kind
of prefabricated, concrete building modules avai l abl eeach
one four feet wide.
It woul d have been impossible to use the tree, because
the modul ar piece has a prescribed way of connecti ng to a
c ol umnand there woul d, within the system, be no way
to connect it to the tree because the tree slopes at an angle
inconsistent with the modules.
It woul d also have been impossible to run the beam into
the tree, because it runs in at an angl e. It woul d have been
necessary to make some awkwar d composition of squares,
whi ch woul d destroy the simple edge between the balcony
and the bushes and trees below.
T h e wi dth of the balcony woul d have had to be either
12 or 16 feet. Th e second of these is too l arge. T h e first
37
6
DI FFERENTI ATI NG SPACE
does not make proper use of the site, and misses the patch
of sunlight on the l eft.
It woul d not have been possible to use the corners as ef-
fective places: the even spacing of the columns woul d have
made it impossible to make the corners large enough to use.
In short, this bal conyorgani c as it i scoul d not have
been built up from modul ar components.
Onl y the process of differentiation, which guaranteed
that each decision fit only the l arger decisions whi ch came
before it, and could go unhampered by the details which
were yet to come, al l owed the balcony to become a natural
living thing.
But of coursey this frocess only works because the fat-
terns in the language have a certain order.
Suppose, for instance, that I take a list of patterns for a
house, in random order. I will not be able to build a co-
herent image from them, because the different operations
will almost certainly contradict each other. T o understand
this clearly, imagine that someone is reading you a collec-
tion of patterns for a house, one at a time. Yo u are trying
to f orm a single coherent image of the house, as he goes
along. And, suppose, for a moment , that he reads out
the list of patterns in a random order. Thus :
OUTDOOR ROOM: T h e family room opens out onto a kind
of outdoor room.
ALCOVES: The r e are alcoves all around the edge of the
family room.
377
T HE WAY
CAR HOUSE CONNECTION: Th e kitchen is near the en-
trance to the house.
CHILDREN'S REALM: Th e children' s bedrooms are near
the kitchen.
FARMHOUSE KITCHEN: T h e family room and kitchen
are next to each other, with a half-open counter between
them.
You cannot read this list of statements one at a time,
in this order, and create a coherent image of a house, as
you read them. Wh e n you come to the last statement, you
will have to " go back. " By the time you have read the
first four statements, you have already placed the family
room, garden, children' s bedrooms, and kitchen in a cer-
tain relationship to one anot her and it woul d be pure
coincidence if this arrangement happened to include the
fifth pat t ernt hat kitchen and family room open into
each other.
If the arrangement you are i magi ni ng at this moment
has the kitchen and family room separated, you will have
to change i t and to change it virtually means goi ng back
to the first statement again.
The patterns will only allow me to form a single
coherent image in my mind, if the order that I take
them in allows me to build an image of a design grad-
ually, one pattern at a time.
And I shall only be able to do this if each pattern is al -
ways consistent with the total image which I have built
from all the earlier patterns in the sequence.
378
D I F F E R E NT I A T I NG S PACE
This requires that the sequence of the patterns meets
three simple conditions.
Fi rst , if pat t ern A is above pattern B in the ne t wor k of
the l anguage, then I must take A bef ore I take B. T h i s
is the most f undament al rul e. Fo r i nstance, if SITTING
ROOM is above ALCOVE in the l anguage (so that a sitting
r oom cont ai ns an al cove as a p a r t ) t h e n , obvi ousl y, I can-
not build ALCOVES into, the i mage of my SITTING ROOM,
unti l I have al ready got a rough and ready i mage of the
SITTING ROOM itself.
Second, I must take all the pat t erns i mmedi at el y
above A , as close t oget her in sequence as I can. I f
CIRCULATION REALMS and CAR HOUSE CONNECTION
are both i mmedi at el y above MAIN ENTRANCE, they both
hel p to set the stage f or the creati on of the mai n ent rance.
T h e y wi l l onl y make a coherent f r a me wo r k f or the
ent rance, in my mi nd, if they come close t oget her, so that
I can fuse t hem.
Th i r d , I must take all the pat t erns i mmedi at el y be-
l ow the pat t ern A , as close t oget her in sequence as I
can. T h u s , f or i nstance, POSITIVE OUTDOOR SPACE and
WINGS OF LIGHT, whi c h both come i mmedi at el y under
BUILDING COMPLEX, must come close t oget her. Wh e n
you put a house on a l ot you are simultaneously creat i ng
the open spaces whi c h f or m the gar den and shapi ng the
bui l di ng. Y o u cannot hel p it. On e def i nes the ot her. T h e
bui l di ng def i nes the open space and the open space def i nes
the bui l di ng. So you must use the t wo pat t erns as nearl y at
the same ti me as possible.
379
T HE WAY
We have been able to show, experimentally} that the
more a sequence of patterns meets these three condi-
tions, the more coherent a person*s image is.
Wh e n the sequence meets these conditions perfectly, any-
oneeven a so-called " l a y " personwi l l spontaneously
create a coherent image of a complete building in his
mind, as he hears the patterns one by one. Wh e n he has
heard all the patterns, he will be able to describe the com-
plete design clearly, will be able to " wa l k " another person
through it, describe what he sees f rom various angles, and
so on . . . in short, his design is coherent and complete.
On the other handy the more a sequence of patterns
violates the three conditionsy the more incoherent the
person
y
s image becomes.
For instance, if t wo patterns, both above a given pattern
in the l anguage, are wi del y separated in the sequence, it is
very likely that the relationship between these patterns
wil l be confused in the emergi ng design. Or , more ex-
treme, if a small pattern comes before a large one in the
sequence, and thus violates the first condition, it is very
likely that all the patterns whi ch lie in between the t wo
in the sequence will be dropped out of the emergi ng de-
sign, or even forgotten entirely.
And this is why a pattern language has the natural
power to help us form coherent images.
For we can al ways use our l anguage to produce a sequence
380
DI FFERENTI ATI NG SPACE
which is consistent with these three conditions. Suppose, for
instance, that you want to use a l anguage to design a house.
We know from chapter 16, that the l anguage has the
structure of a network, or cascade. Assume, for the sake
of argument, that the l anguage contains IOO patterns
whi ch are needed in the house. T o get these 100 patterns,
in the proper order, you must simply begin a kind of trip,
in which you move through the network of the l anguage,
taking one pattern at a time, movi ng roughl y downwards,
and back and forth at the same time, f ol l owi ng the three
conditions as closely as you can.
The sequences which we obtain from language> will
meet these three conditions almost automatically.
Of course, the particular sequence of patterns for any
given proj ect, will al ways be unique according to the de-
tails of the proj ect.
Thi s happens because, accordi ng to the context, the pat-
terns have slightly di fferent relations to one another . . .
which will af f ect the sequence they have when we try to
meet the three conditions. Thus , for example, in a house
on a narrow city lot, the pattern SMALL PARKING LOT will
exert a controlling influence on the desi gnand the pat-
tern must therefore come early in the sequence. I n another
house, where there is more land, this pattern can come
later in the sequence (since cars can be put almost any-
where wi thout disturbance), but other pat t ernsf or in-
stance TREE PLACESmust come earlier because they now
exert a controlling influence on the design.
But in every case, there is some sequence of patterns
381
T HE WA Y
whi ch is the most appropriate for that desi gn; and you
can get this sequence more or less directly f rom your
knowl edge of the size of the morphological ef f ect which
each pattern has, compared with all the others.
The sequence of the patterns for a designas gen-
erated by the languageis therefore the key to that
design.
For once you find the proper sequence, the power to design
coherent things fol l ows from it almost automatically, and
you will be able to make a beautiful and whol e design,
without any trouble. If the sequence is correctly f ormed,
you can create a beautiful whol e, almost wi thout trying,
because it is in the nature of your mind to do so. But , if the
sequence is not correctly f or medi f the sequence is itself
incoherent, or the patterns in it i ncompl et et hen no
amount of trying will al l ow you to create a design whi ch
is whol e.
Conventional wisdom says that a building cannot be
designed, in sequence, step by step.
But the fact is that you cannot understand the morphol ogy
of a building, or create a design whi ch has that morphol -
ogy, until you understand whi ch features are dominant,
and whi ch ones secondaryi t is fundamental to your
ability as an artist that you establish this morphological
order in your mind. I n this sense, the actual creation of
the sequence, by the artist, is one of the most crucial
382
D I F F E R E N T I A T I N G S PACE
aspects of the desi gn task. Y o u real l y underst and wha t
you are doi ng f or the fi rst ti me at that mo me nt whe n the
sequence of the pat t erns becomes cl ear to you.
Sequences are bad whe n they are the wrong sequences.
But the sequence which a language gives you works be-
cause it treats the building as a whole> at every step.
Ea c h pat t ern is a fi el d whi c h spreads t hr oughout the whol e,
and ti nges it, distorts it, patterns it. W e can take patterns,
step by step, one at a t i me, because each pat t ern moul ds
the wh o l e a n d each pat t ern can moul d the whol e whi c h
is the pr oduct of the previ ous pat t erni ngs.
In nature
}
a thing is always born, and developed
y
as a
whole.
A baby starts, f r om the fi rst day of its concept i on, as a
whol e, and is a whol e, as an embr yo, every day unti l it is
born. I t is not a sequence of addi ng parts t oget her, but a
whol e, whi ch expands, cri nkl es, di f f erent i at es itself.
A wa v e is shaped as a whol e. I t is part of the system
of waves , and is a part of a l egi t i mat e we l l - f or me d l i vi ng
whol e, as it starts, swel l s, crashes, and dies.
A mount ai n is shaped as a whol e : the crust of the earth
heaves, the mount ai n takes its f o r m; and whi l e it is g r o w-
i ng, each rock, and parti cl e of sand is also whol e : there is
not hi ng unf i ni shed in it, dur i ng the thousands of years that
bri ng it to the state we know t oday.
383
T HE WA Y
A building, too
}
can only come to life when it grows
as a whole.
As we concei ve the whol e, in our mi nds, it starts as a
whol e, and continues as a whol e, t hroughout our mental
operations; and ends as a whol e. Each mental operation
di fferenti ates it, makes it more el aborate: but treats it as a
whol e, and operates on the i mage we have of it, as a
whol e.
At every l evel , certain broad patterns get laid down:
and the details are squeezed into position to conf orm to
the structure of these broader patterns. Of course, under
these ci rcumstances, the details are al ways slightly di f f er-
ent, since they get distorted as they are squeezed into the
l arger structure al ready laid down. I n a design of this
type, one natural l y senses that the gl obal patterns are more
i mportant than the details, because they dominate the de-
sign. Each pattern is gi ven the i mportance and control over
the whol e whi ch it deserves in the hi erarchy of patterns.
And it follows
y
therefore
}
that when a pattern lan-
guage is properly used
y
it allows the person who
uses it to make places which are a part of nature.
The character of nature is not something added to
a good design. It comes directly from the order of
the language. When the order of the patterns in the
language is correcty the differentiating process allows
the design to unfold as smoothly as an opening flower.
We are ready now to find out the details of this
process of unfolding.
384
CHAPTER 20
ONE PATTERN AT A TIME
The process of unfolding goes step by
step
}
one pattern at a time. Each step
brings just one pattern to life: and the
intensity of the result depends on the
intensity of each one of these individual
steps.
Suppose now, that for a given act of building, you
have a pattern language, and that the patterns in this
language are arranged in proper sequence.
To make the design, you take the patterns one by
one, and use each one to differentiate the product of
the previous patterns.
But how exactly, does each pattern work.
A t any gi ven moment in the unf ol di ng of a sequence of
patterns, we have a partl y def i ned whol e, whi ch has the
structure gi ven to it by the patterns that come earlier in
the sequence.
An d we are now f aced wi t h the probl em of i nj ect i ng
the next pattern i nto this whol e, filling the whol e wi t h
it, i nf usi ng the whol e wi t h the structure of this pattern,
maki ng j ust those di f f erent i at i ons in the whol e, whi ch
wi l l bri ng this next pat t ern to l i fe, t hroughout the fabri c
of the whol e.
Ho w exact l y does this wor k?
Suppose, for instance, that you want to create a
WI NDOW PLACE which is alive.
St art by r emember i ng all the parti cul ar wi ndow pl aces
you have k n o wn : especially the ones whi ch wer e most
beauti f ul . Cl ose your eyes, and concent rat e on t h e ms o
that you get a direct instinctive knowl edge of the pat t ern,
rooted in your own experi ence.
An d concent rat e, also, on the parti cul ar aspects of the
WINDOW PLACE whi ch make it l i ve: the l i ght , the seat,
39
ONE P A T T E R N AT A T I ME
the wi ndowsi l l , flowers g r o wi ng outside perhaps, the
quietness and separateness, whi ch make the wi ndo w place
a " pl a c e . "
Ask yourself how this pattern would look if it were
already in the place where you are wanting it.
T o do this, you need onl y close your eyes, and i magi ne
t hat you are c omi ng t hr ough the door. I magi ne that the
r oom or pl ace whi c h you are shapi ng has a WINDOW
PLACE in it.
Y o u r knowl e dg e of the pat t ern, i nt eract i ng wi t h your
knowl e dge of the pl ace, wi l l speak to you, and tell you
j ust wha t f or m the pat t ern takes, in this parti cul ar
pl ace.
T o keep the pat t ern st rong, it is essential t hat you
don' t yet put in any ot her details. Y o u do not yet need
to i magi ne the positions of the wi nd o w panes, in detail,
because that wi l l be done by l at er patterns. Y o u do not
need to k no w, yet , wha t exact l y is the hei ght of wi nd o w-
sills, because the pat t ern LOW SILL does it f or you, l ater.
Y o u do not need to k no w, yet , j ust exact l y wha t the
cei l i ng hei ght is, t here, because the pat t ern CEILING
HEIGHT VARIETY wi l l do it f or you l at er.
T h e onl y t hi ng you must have cl ear, at this st age, is
the whol e, the space of the WINDOW PLACE itself, how
l ar ge it is, how the l i ght comes i nto it, how people sit, so
t hat they are rel ated to the l i ght and to the inside of the
r oom, and, above al l , j ust how the wi nd o w pl ace does
i ndeed def i ne a flace whi c h is i dent i f i ed, and l i ght . The s e
391
T H E WA Y
thi ngs you must k no w, because these are the ones whi ch
this pat t ern speci fi cal l y deals wi t h.
The most important thing is that you take the pattern
seriously.
T h e r e is no point at all in usi ng the pattern if you onl y
gi ve lip service to it.
For exampl e, I r emember one case wher e a man was
desi gni ng a beach house, wi t h t wo apart ment s one over the
o t he r a nd an out door stair, l eadi ng to the upstairs one.
He was tel l i ng me how the patterns he was usi ng entered
into his desi gn: and he said that the stair l andi ng at the
top of the stair was the ENTRANCE TRANSITION for the
upstairs room.
I said to hi m: Look, this little l andi ng, whi ch is maybe
three feet by f our, is not an ENTRANCE TRANSITION at all.
Y o u are cal l i ng it that and t ryi ng to pretend to yoursel f
that you have created this pattern there. But it is j ust
the top of the stair. A n ENTRANCE TRANSITION is a place
wher e the l i ght changes, a place wher e the l evel changes,
wher e you are suddenl y filled wi t h a sense of new ex-
peri ence, a change, a wi pi ng cl ean, that happens j ust be-
fore you go inside.
If you really want to make an ENTRANCE TRANSITION
there
y
at the top of the stair, you must close your
eyeSy and ask yourself: What would it be like
y
if this
was the most wonderful entrance transition in the
world?
392
ONE P A T T E R N AT A T I ME
I magi ne. I close my eyes. I see a pl ace f r om whi c h there
is a sudden vi ew, whi c h you can' t see f r om the bot t om of
the stair. I see a pl ace whi c h is f i l l ed wi t h the scent of
j asmi ne in the s ummer . I hear my step as I reach this
pl ace up on t op: it changes, because, perhaps, there is a
cr eaki ng board. So no w I begi n to i magi ne a stair whi ch
is al most closed in, by woode n f r e t wor k, the top pl aced
wi t h an openi ng whi ch l ooks out on the sea; a trellis
over head wi t h j asmi ne cl i mbi ng on i t; a seat across f r om
the door, whe r e I can sit, to smel l the br eeze; the stair
made l oosel y, so that the boards creak as I begi n to cl i mb
the stair . . . .
Now you have really done something.
N o w this ent rance transi ti on is more t han j ust a phrase;
it is an act ual l i vi ng t hi ng. But n o w, of course, it is a little
st range, perhaps. H o w do I have to build the stair, to
keep it closed in f ur t her down, wi t h f r et wor k to let l i ght
i n; how must I place the l andi ng at the top, so that it
l ooks back at the sea? I t is no l onger j ust the " t op of the
st ai rs" s omewher e. I t is a pl ace whi c h I wi l l not f or get ,
because it has a charact er. A n d this charact er is not
created wi l f ul l y, by me i t has been gener at ed, si mpl y,
by payi ng serious attenti on to the pat t ern.
Indeed
y
each fat tern, when you really do it
}
creates
an almost startling character.
Wh e n you insist, and genui nel y f or m the pat t ern, and go
393
T H E WA Y
all the wa y wi t h it, it generat es a charact er: it l ooks al -
most s t r ange; sl i ghtl y startl i ng. I t is ext r eme; you know
that someone has been wor ki ng here. I t is not bl and, but
f ul l .
For instance
y
in the second picture of this chapter
y
the
roof overhang of the SHELTERI NG ROOF is immense.
Th e r e is no lip service to the shel t eri ng roof . T h e people
who built this built it wi t h all the courage and the cer -
tainty that it wa s real l y needed. T h e r e are no hal f
measures; no compromi ses; this is a shel teri ng roof , in
f ul l .
In the third picture
y
the FI LTERED LI GHT is once again
intense.
Th i s pi cture is i mport ant , because it shows that t aki ng
patterns seriously does not requi re money. Her e, in the
simplest hut, the people inside f el t so st rongl y about
FILTERED LIGHT that they st r ung beans on stri ngs, across
the wi ndo w, in the most ext raordi nary wa y . T h e y took
the pat t ern seriously, and it creat ed, f or t hem, somet hi ng
unique.
I t is the same i ntensi ty you ma y f i nd in a rock or
tree whi ch gr ows under ext raordi nary ci rcumstances.
Wh e n a tree gr ows in a corner, wi nd f r om one angl e,
rocks beneath it, it ma y take on a wi l d charact er, intense,
created by the i nteracti on of the special ci rcumst ances,
wi t h its genes. Wh e n the patterns are gi ven their f ul l i n-
394
ONE P A T T E R N AT A T I ME
tensity, and al l owed to i nteract f reel y wi t h ci rcumstances,
we get the same.
And in the first picture > we see LI GHT ON TWO SIDES
OF EVERY ROOM as intense as possible.
Al mos t everyone has some experi ence of a r oom fil l ed
wi t h l i ght , sun st reami ng in, perhaps yel l ow curtai ns,
whi t e wood, patches of sunl i ght on the f l oor, whi ch the
cat searches f o r s o f t cushions wher e the l i ght is, a gar -
den ful l of flowers to l ook out onto.
I f you search your o wn experi ence, you can certai nl y
r emember a place like t hi s s o beauti f ul it takes your
breath a wa y to thi nk of it.
Look at the gr eat r oom f r om the Topka pi Pal ace in
this first pi cture. It is al most as t hough the room itself
is one great wi ndow. Y o u can make a room like that, if
you pay enough attenti on, are serious enough about the
wi ndows , l ook to see wher e the l i ght is, place the room,
not in some place wher e it j ust gets l i ght on t wo sides, but
in the best pl ace, wher e it can get l i ght f rom all around,
and the best and most beauti f ul l i ght there is on the site.
T h e n it can happen.
You may not believe that you can make a -place as
beautiful as that.
So, whe n you come to this pattern, LIGHT ON TWO SIDES
OF EVERY ROOM, you check in a hal f heart ed, per f unct or y
manner , to see if every room has t wo wal l s to the outside,
395
T H E WA Y
and that there are a coupl e of wi ndows , more or less in
the ri ght pl ace.
But that produces not hi ng. I t is onl y whe n you pay
attenti on, in the f ul l belief that every room you make can
be as beaut i f ul as the most beaut i f ul l i ght-f i l l ed room
whi ch you have ever s e e nt he n you are serious enough.
A t that moment it wi l l happen. Al l it requires is the wi l l .
To do it you need only let it happen in your mind.
Say to yoursel f : I am comi ng into that room, I am not
in it yet , but I come in t hrough the door, and there, to
my surprise, is the most beauti ful room that I have ever
been in. T h e pattern LIGHT ON TWO SIDES is there, as
intense and beauti f ul as in any room that I have ever
known, as intense as in the picture of the Topkapi Pal ace
. . . you say all this to yoursel f , before you wal k in
t hrough the door. An d then, wi t h your eyes closed, in
your i magi nat i on, you wal k f r om the next door room,
t hrow open the door, and step inside . . . and there it
And there it is. Suddenly
y
without your making any
conscious effort, your mind will show you how this
light on two sides is
y
in that particular place
y
as
beautiful as you have ever known it anywhere.
Do not consci ousl y try to create the pattern. I f you do
this, the i mages and ideas in your mi nd wi l l distort it,
wi l l begin to take over , and the pattern itself wi l l never
396
ONE P A T T E R N AT A T I ME
make its wa y i nto the wo r l d: instead there wi l l be a
" de s i gn. "
Ge t rid of the ideas whi ch come into your mi nd. Ge t
rid of pi ctures you have seen in magazi nes , f ri ends'
houses . . . . Insist on the pat t ern, and not hi ng else.
T h e pat t ern, and the real si tuati on, t oget her, wi l l
create the proper f or m, wi t hi n your mi nd, wi t hout your
t ryi ng to do it, if you wi l l al l ow it to happen.
Th i s is the power of the l anguage, and the reason why
the l anguage is creat i ve.
Your mind is a medium within which the creative
spark that jumps between the pattern and the world
can happen. You yourself are only the medium for
this creative spark
}
not its originator.
I r emember once, si tti ng in Ber kel ey, t r yi ng to wo r k out
a site pl an on paper, f or our houses in Per u. On e of the
LOOPED LOCAL ROADS into the site was not yet properl y
in pl ace, and we coul d not f i nd a suitable wa y of f i t t i ng
this pattern road into the desi gn, the wa y the patterns told
us t o s o I deci ded to take a wa l k around the site in my
i magi nat i on.
I sat in my chai r, in Ber kel ey, 8000 miles f r om the real
site in Li ma , closed my eyes, and began to take a wal k
ar ound the mar ket . T h e r e we r e ma ny na r r ow l anes,
covered wi t h bamboo screens to shade t hem, wi t h tiny
stalls openi ng of f t hem, and f rui t sellers sel l i ng f rui t f r om
carts. I stopped by one ol d woma n' s cart , and bought an
or ange f r om her. As I stood there I happened to be
397
T HE WA Y
f aci ng north. And then I bit into the or ange i n my
i magi nati on. And just as I bit into it, I suddenl y stopped
and asked mysel f , " No w, where is that r oa d? " And,
wi thout thi nki ng, I knew exactl y where it was, and what
its relation to the market wa s I knew it must be over
therey t owards the ri ght, f rom the direction I was f aci ng.
I knew that to be nat ural ; it must come sweepi ng t owards
the market , and touch it over there.
T h e n I stopped, and came back to my room, and my
chair, and my ef f ort to make the design. I real ized at
once that this position of the road, whi ch had come to me
so natural l y, was quite di f f erent f rom all the ones we had
been tryi ng on paper in those last f ew da y s a nd that it
was exactl y right, and satisfied the demands of all the
patterns perfectl y.
I t was the vividness of bei ng there, and biting into
that orange, that al l owed me to know, spontaneously, the
most natural place for that road to be.
You may find this way of letting patterns form
themselves
y
unusual.
T o do it, you must let go of your control and let the
pattern do the wor k. Y o u cannot do this, normal l y, be-
cause you are tryi ng to make decisions wi t hout havi ng
conf i dence in the basis for them. But if the patterns you
are using are familiar to you, if they make sense to you,
if you are conf i dent that they make sense, and that they
are prof ound, then there is no reason to be afrai d, and no
reason to be afraid of gi vi ng up your control over the
398
ONE P A T T E R N AT A T I ME
desi gn. I f the pat t ern makes sense, you do not need to
cont rol the desi gn.
You may be afraid that the design won't work if you
take fust one pattern at a time.
If you take one pat t ern at a ti me, wha t guar ant ee is
there that all the pat t erns wi l l f i t t oget her coherentl y:
5
Wh a t happens if you put the pat t erns t oget her, one at a
ti me, and t hen, suddenl y, wi t h the ni nth or tenth, you
f i nd that it is quite impossible, because there is a conf l i ct
bet ween the desi gn whi ch has emer ged so f ar, and the
next pat t ern in the sequence?
T h e great est f ear we experi ence in the process of
desi gn is that ever yt hi ng wi l l not wo r k out. An d yet the
bui l di ng wi l l become al i ve onl y whe n you can let go of
tli is f ear.
Suppose, f or i nstance, you are t r yi ng to decide whe r e
to place the ent rance to your house. As you are doi ng it,
i mages of ot her probl ems flit t hrough your mi nd. Wi l l I
be able to fi t the di ni ng r oom in if I put the ent rance
here? But on the other hand, if I put it over t here, maybe
I wo n' t have r oom to put the bed al coves in properl y;
. . . wha t shall I do? Ho w can I place the ent rance so
that all these probl ems wi l l wor k t hemsel ves out whe n
the time comes?
But you cannot create a pattern at full intensity
y
so
long as you are worrying and thinking about other
patterns
}
which you will have to deal with later in
the sequence.
399
T H E WA Y
Th i s f r enz y wi l l al ways kill the pattern. I t wi l l f orce you
to create arti fi ci al , " t hought out " conf i gurat i ons, whi ch
are stiff and lifeless. T h i s is the t hi ng whi c h prevents
people, most of t en, f r om creat i ng a pat t ern at f ul l i n-
tensity.
Suppose, for exampl e, that we are t ryi ng to build a
house whi ch has f i f t y patterns in it. I t seems al most i n-
concei vabl e that these f i f t y patterns wi l l not s omehow
conf l i ct : and it seems essential, t heref ore, to wor k out
some overal l scheme in whi ch enough compromi ses are
made to al l ow each of the patterns to be present to some
ext ent .
This frame of mind destroys the patterns.
I t destroys all possibility of life, because as soon as you
begi n to compromi se the patterns there is no life l ef t in
t hem.
But there is no need f or this f r ame of mi nd. I t is not
necessary to make compromi ses bet ween the patterns.
Wh e n you start to think about compromi ses bet ween
patterns, you are not t aki ng account of the f act that
every pattern is a rule of transformation. T h e f act that
every pattern is a rule of t ransf ormat i on means that each
pat t ern has the power to t ransf orm any conf i gurat i on by
i nj ect i ng a new conf i gurat i on into it, wi t hout essentially
di sturbi ng any essentials of the conf i gurat i on whi ch was
there bef ore.
Suppose that I want to create a MAI N ENTRANCE.
400
ONE P A T T E R N AT A T I ME
T h e char act er of the MAIN ENTRANCE as a rul e means
that I can take any conf i gur at i on, whi ch l acks this pat-
t e r n i t may be a real bui l di ng whi c h exists al ready, or a
partl y t hought out bui l di ng in my mi n d a n d appl y this
pat t ern to i t t h a t is, i nj ect a MAIN ENTRANCE into it, in
the most beaut i f ul and ext r eme f or m pos s i bl ewi t hout
di sturbi ng the essentials of wha t I have al ready.
There is no reason to be timid.
I f I am goi ng to create a beaut i f ul MAIN ENTRANCE,
there is no point in wor r y i ng about whet her I wi l l l ater
be able to create a beaut i f ul ENTRANCE TRANSITION
there.
A t this time whe n I i nj ect MAIN ENTRANCE into the
desi gn, I need thi nk onl y about the MAIN ENTRANCE
pat t ern, in all its ful l ness and e x t r e mi t y i n the certai n
knowl e dge that whe n I come, l at er, to the ENTRANCE
TRANSITION pat t ern, I wi l l once agai n be able to i nj ect
that pat t ern, in all its ful l ness and ext r emi t y also.
The order of the language will make sure that it is
possible.
Fo r as we have seen in Chapt er 1 9, the order of the
l anguage is the or der whi c h the pat t erns need to operate
on one anot her to creat e a whol e. I t is a morphol ogi cal
or der , similar to the or der whi ch must be present in an
evol vi ng embr yo.
An d it is this ver y same order whi c h also al l ows each
401
T HE WA Y
pattern to devel op its ful l intensity. Wh e n we have the
order of the l anguage ri ght, we can pay attention to one
pattern at a time, wi t h f ul l intensity, because the inter-
ference between patterns, and the conflicts bet ween pat-
terns, are reduced to almost not hi ng by the order of the
l anguage.
Within the sequence which the language defines, you
can focus on each pattern by itself
y
one at a time
y
certain that those patterns which come later in the
sequence will fit into the design which has evolved
so far.
You can pay full attention to each pattern; you can
let it have its full intensity.
Then you can give each pattern just that strange
intensity which makes the pattern live.
402
CHAPTER 2 1
SHAPING ONE BUILDING
From a sequence of these individual fat-
terns, whole buildings with the character
of nature will form themselves within
your thoughts, as easily as sentences.
We are ready
}
now> to see just how a sequence of
patterns can create a building in our minds.
It happens with surprising ease. The building al-
most "makes itself " just as a sentence seems to when
we speak.
And it can happen as easily within an ordinary
person's mind
y
or in a builder's mind. Kveryone
}
builder or not> can do this for himself
}
to make a
building live . . . .
Assume, to start wi t h, that we have a l anguage for a
house.
Look at the patterns in the order they come in, one
at a time.
Add nothi ng, except just what the patterns demand.
Sl owl y, you wi l l f i nd that an i mage of a house is
gr owi ng in your mi nd.
Here are the rough notes I wrote down during the
week it took to design a small cottage in this way.
I decided to build a smal l cot t age/workshop at the back
of our office. A place l arge enough to live i n; a place
where guests mi ght st ay; and a place where someone
could live and wor k, as a workshop; and a place we
coul d rent to a fri end, when it wasn' t occupied by one of
us.
The r e is a l arge house in f r ont ; another cottage be-
hi nd; an old gar age; outside stairs l eadi ng to the upstairs
of the l arge house. I decided that it woul d not be practical
407
T H E WA Y
to spend mor e t han about $3000 f or mat eri al s. A t $8 per
square f oot f or mat er i al s ( I k n e w t hat we wo ul d bui l d it
oursel ves so l abor wa s to be f r e e ) , we coul d bui l d a cot -
t age of 400 square f eet .
He r e is the l a ng ua g e I chose f or the bui l di ng:
WORK COMMUNI TY
THE FAMI LY
BUILDING COMPLEX
CI RCULATI ON REALMS
NUMBER OF STORIES
HOUSE FOR ONE PERSON
SOUTH FACING OUTDOORS
WINGS OF LI GHT
CONNECTED BUILDINGS
POSITIVE OUTDOOR SPACE
SITE REPAIR
MAIN ENTRANCE
ENTRANCE TRANSITION
CASCADE OF ROOFS
ROOF GARDEN
SHELTERING ROOF
ARCADE
I NTI MACY GRADIENT
ENTRANCE ROOM
STAIRCASE AS A STAGE
ZEN VIEW
TAPESTRY OF LI GHT AND DARK
FARMHOUSE KI TCHEN
BATHING ROOM
HOME WORKSHOP
408
S HA P I NG ONE B UI L DI NG
LIGHT ON TWO SIDES OF EVERY ROOM
BUILDING EDGE
SUNNY PLACE
OUTDOOR ROOM
CONNECTION TO THE EARTH
TREE PLACES
ALCOVES
WINDOW PLACE
THE FIRE
BED ALCOVE
THICK WALLS
OPEN SHELVES
CEILING HEIGHT VARIETY
The first thing was repair.
T h e exi sti ng cot t age is di sconnect ed. T h e gar age is a bit
derel i ct ; the trees and grass at the ver y back need pr uni ng
and are ver y ov e r g r own. Abo v e al l , the people wh o live
upstairs in the mai n house, and at the back, have no over -
all sense of connect i on to one anot her . Al s o the most
beaut i f ul part of the g a r d e n wh i c h f aces south, and is
under the l ocust tree, is unused, because there is not hi ng
near it or ar ound it, and no paths whi c h nat ural l y go to
and f ro to make it nat ural l y usable.
To solve all these problems
y
I tried
y
-first of all
y
to
make a building which created SOUTH FACING OUT-
DOORS and POSITIVE OUTDOOR SPACE.
409
T H E WA Y
For SOUTH FACING OUTDOORS, I i magi ne a nice big t er-
race, out t owar ds the back of the mai n house, in the sun.
I f we put it to the south and west of the cot t age, it wi l l
be in the other openi ng in the trees, and get pl enty of
sun: a good place to wor k, make t hi ngs; perhaps we can
put a wor kbench out there in good weat her ; and a place
for a couple of chairs and a table, wher e we can sit and
have a dri nk. We need to spend a day on the site, wat c h-
i ng the sun, to i denti f y the exact places wher e sun falls
(SUNNY PLACE) ; it is tri cky, because the sun comes
t hrough the trees, onl y in a f e w special places, and we
must be very accurat e about pl aci ng t hem.
Al l this puts the cot t age as f ar to the nort h as possible.
T o f or m POSITIVE OUTDOOR SPACE, I also place the
bui l di ng wel l back into the site, so that it l eaves a wel l -
f or med space bet ween the gar age and the trees in f ront .
I n that position, there is space f or a cot t age, r unni ng
nort h-sout h, up to about 13 feet wi de, and up to about
25 feet l ong. As f ar as connect i ons wi t h the exi sti ng cot-
tage are concerned (BUILDING COMPLEX, CONNECTED
BUILDINGS), there is no bat hroom in the existing cot t age,
so it wi l l be a great hel p if we build a bat hroom whi ch
the t wo cottages share. T h e r e is a nat ural place for that,
ri ght bet ween the t wo buil dings.
Next NUMBER OF STORIES, CASCADE OF ROOFS, SHEL-
TERING ROOF, ROOF GARDEN give me the overall shape
of the building.
Mos t l y it wi l l be a one-story bui l di ng; but we want to
try the t wo- st ory st ruct ure, and it woul d be nice to have
4 1 0
S HAP I NG ONE B UI L DI NG
a sl eepi ng l of t upstairs. T h i s t wo- st or y part of the t hi ng
shoul d nat ural l y be to the nort h end, so that it f or ms a
ROOF GARDEN to the south. Gi v e n its position, it makes
sense to thi nk of this sl eepi ng l of t as about 8 x 1 3 , openi ng
out to a flat roof to the south, over the one- st ory part of
the cot t age. T h i s goes some wa y t owar ds creat i ng the
CASCADE OF ROOFS. SO that our nei ghbor to the nort h
doesn' t have a hi gh wal l ri ght next to his gar den, it makes
sense to i magi ne l owe r , al cove roof s f al l i ng of f to the
nort h. A n d the same s omewher e to the south perhaps,
and the same agai n whe r e the ent rance is, there mi ght be
an ent rance porch. T h i s wi l l make a number of l ower
roofs, l ow enough to t ouch, ar ound the edge of the
bui l di ng (SHELTERING ROOF and CASCADE OF ROOFS).
Within this overall shape, CIRCULATION REALMS and
WORK COMMUNI TY tell me how to complete the site.
CIRCULATION REALMS is not ;ood; and the connect i on
O '
wi t h the mai n house needed by WORK COMMUNITY isn' t
good enough. T h e mai n troubl e is this. T h e r e are t wo
paths to the back: one up the dr i v e wa y ; the other t hrough
dark bushes. T h e one up the dr i ve way is O K , but not a
di rect connect i on; and the mai n house' s back porch goes
out si deways to i t; not di rect. T o make the connect i on
cl ear, and the ci rcul at i on, we wi l l open up the back of the
back porch, so it connect s di rectl y to the t errace of the
cot t age. I t wi l l onl y be a f e w feet, then, f r om the back
porch, to the cof f ee, umbrel l a, chairs, wo r k b e n c h o r
what ever else we put on the t e r r a c e a nd it wi l l be
nat ural to go back and f or wa r d all the ti me. W e can l ay
4 1 1
T H E WA Y
tiles into the earth, to make the connect i on; also, l ooki ng
at the bushes under there, whi ch are so d a r k we need to
prune t hem, cut out dead wood, to bri ng more l i ght in
al ong that path. W e mi ght even prune out enough dead
wood so grass coul d g r o w there, wi t h the trees j ust
standi ng in the grass.
SITE REPAI R tells me exactly what to protect around
the building.
T h e tree to the north is down, as our nei ghbor wa nt e d;
in exchange we wi l l , I hope, be able to build up to his
f enc es i nc e he can get sunl i ght all the wa y onto his
l awn now. I t is a shame cut t i ng down a t ree: but the
trees at the back g r e w too t hi ck; one less and the others
wi l l g r ow st r onger ; and, most i mport ant , it helps to
repair his garden to the north of ours, by gi vi ng him
SOUTH FACING OUTDOORS too.
I n cl eari ng the site, the little apple tree next to the
gar age seems more beauti ful than ever ; and the wi l d
onions, wi t h their whi t e flowers, g r owi ng around the
f oot of it are l ovel y. W e have put stakes around t hem, to
protect them whi l e we bui l d: they get t rampl ed ver y
easily (SITE REPAIR).
Combi ni ng SITE REPAIR and ROOF GARDEN, I i magi ne
the roof gar den about ei ght or nine feet up, beauti f ul l y
f r amed and surrounded by the l owe r branches of the
trees to the east and wes t : on the site I stake out the
rough position of this roof gar den, so that it wi l l fit j ust
ri ght into the trees.
4 1 2
S HA P I NG ONE B UI L DI NG
Now I begin more carefully to work with my eyes
closed
}
to imagine how the patterns will be
y
in their
best
y
most natural
y
most simple form
y
as the building
comes to life.
MAI N ENTRANCE gives me the approach to the
buildingy and the position of the entrances.
T h e r e are t wo way s to approach the c ot t a ge e i t he r f r om
our back porch of the mai n house, or up the dr i veway.
Wh e r e is the ent rance and wha t is it l i ke, to make these
t wo approaches wor k? I n both cases I come across the
t errace in f r ont , to reach the ent rance. I had ori gi nal l y
t hought of an ent rance wi t h a porch or ar c ade: but it
seems too dark in t here. A s I close my eyes, I see a f r ont
door, st andi ng f or wa r d a little f r om the mai n r oom of
the cot t age, j ust behi nd the brambl e bush, and next to
the acaci a whi c h is still st andi ng. I i magi ne a smal l seat
on ei ther side of it: a nat ural pl ace to sit in the sun: and
the ent rance f r ame el aborat ed, perhaps car ved or pai nted,
not muc h, j ust sl i ght l y, perhaps bul gi ng f or wa r d. Si nce
I k now that the bat hr oom wi l l be to the back, next to
the exi sti ng cot t age, t owar ds the nor t h, and I assume
there wi l l be a short arcade connect i ng the t wo bui l di ngs,
and gi vi ng access to this bat hr oom, I am not certai n of
the rel ati on bet ween the mai n ent r ance and this arcade
behi nd it. Al s o I am not certai n if the ent rance is at a
sl i ght angl e, to f ace the dr i v e way mor e, or if it f aces due
west . Bef or e I t hought it ought to be west , but cl eari ng
the site has made the di agonal seem possible. Some how
it seems nat ur al t hat it shoul d occupy the little di agonal
4 1 3
T H E WA Y
bet ween the apple tree and the acaci a. T h e r e is also the
question of the stair. Shall it be near the ent rance goi ng
uppe r haps even out door s or shall it be back in the far
corner, t ucked a wa y (OPEN STAIRS, STAIRCASE AS A
STAGE)?
INTIMACY GRADIENT and INDOOR SUNLIGHT give me
the overall layout of the inside.
INTIMACY GRADIENT doesn' t mean much in such a smal l
bui l di ng, except perhaps for the f ol l owi ng ideas, ( i ) A
smal l seat or wi ndow seat inside the f ront door, ( 2 ) the
stairs far enough back so it is a secl uded " b e d " area, and
( 3 ) the stairs pl aced so a person can go out to the bath-
room wi t hout c omi ng t hrough the f ront d o o r i n other
wor ds a ki nd of back access out to the little arcade that
gi ves onto the bat hrooms, INDOOR SUNLIGHT tells me that
mai n usable spaces are t owar ds the terrace, t owar ds the
gar age, t owar ds the mai n ho us e a nd that the north side,
over t owar ds our nei ghbor, is kept for dark closets, stor-
age. I t may make sense to pl ace a whol e r ow of st orage
al coves over on the nort h s i det hi s wi l l also help to ac-
complish NORTH FACE. Th i s may i ncl ude ki tchen count er
and stove if they are added l at er.
STAIRCASE AS A STAGE, ZEN VIEW, TAPESTRY OF LIGHT
AND DARK give me the -position of the stair to the
upstairs.
St andi ng inside the mai n room of the cot t age, it feels as
if the stair coul d go up on the side opposite the ent rance.
4 1 4
S HAP I NG ONE B UI L DI NG
Th i s makes the most sense; it helps f or m the r oom then,
and its roof , whi ch wi l l stick up sl i ghtl y, at the back of
the roof t errace, and f or m a nice angl e wi t h the t wo-
story s e c t i ona beaut i f ul cor ner , f aci ng sout h- west , nice
to sit in to enj oy the roof . T h i s means that the stair wi l l
go up, perhaps t owar ds a wi ndo w at its top, whi c h looks
out onto the nei ghbor' s gar den to the nort h ( n e w v i e w
the onl y place f r om whi c h one can see out that wa y ) ;
and gi ves LIGHT TO WALK TOWARDS. Ot he r aspects of
TAPESTRY OF LIGHT AND DARKt he r e shoul d be l i ght ,
wher e the back area ( ki t chen ar ea) opens out to the
door whi c h l eads to the a r c a de pe r ha ps a little f ount ai n
or court t here, f or mi ng l i ght , i nvi t i ng us out t owar ds the
smal l exi sti ng cot t age. An d , of course, f r om inside the
mai n r oom, l ooki ng out t owar ds the f ront door ont o the
t errace, is also l ooki ng t owar ds l i ght.
ARCADE tells me how to connect the building to the
cottage west of it.
A s f ar as the little arcade at the back is concer ned, be-
t ween the " ki t c he n" and the old cot t age, wi t h the bat h-
room of f it, I tal ked to Susie in the cot t age: we l ooked at
the wi ndo w of her bedr oom, wher e I had hoped to make
a door, and it was cl ear to both of us that if we made a
door, it woul d rui n the inside of the r o o mi t is so smal l ,
the second door woul d make it like a corri dor. So, I sug-
gested we l eave the wi ndo w f r ame wher e it is, and put a
step on the inside, and t wo steps on the outsi de, like a
stile. W e wi l l put a casement wi ndo w in the wi ndo w
f r ame, perhaps make the wi ndowsi l l 3 i nches l o we r ; and
4 1 5
T H E WA Y
she can get out to the bat hroom by goi ng over the stile,
down the t wo steps, i nto the arcade.
SLEEPING TO THE EAST helps shape the roof in detail
y
because of light.
I l ooked at the l i ght t hrough that wi ndow. T h e r e is a
danger our ne w cot t age may take away her mor ni ng
l i ght ; so we wi l l pl ace bamboos to mar k the roof line
we expect , and move it, unti l there is still pl enty of
mor ni ng l i ght c omi ng in t hr ough that wi ndow. Looki ng
out of the same wi ndo w, it seems mor e i mport ant for the
roof to pitch to east and west , wi t h gabl e ends at north
and south, so that the slope of it al l ows the l i ght to come
do wn easier into her cot t age. T h e gabl e end makes more
sense f or the sl eepi ng l of t a n y wa y i t can open di rectl y
onto the roof gar den (SHELTERING ROOF).
ENTRANCE TRANSITION shows me how to arrange the
area in front of the building.
I haven' t been caref ul enough wi t h this pa t t e r nha v e l ef t
it a little too late. I have been t hi nki ng about the possibility
of a TRELLISED WALK, or TRELLIS a ny wa y , to hel p close
of f the terrace to the south, and hel p protect it a little
f r om the big house to the south. T h i s wi l l also make the
terrace more of an OUTDOOR ROOM, and hel p to make the
di rect connect i on to the house more i mport ant than the
one f r om the dr i veway. So I close my eyes, and i magi ne
comi ng up the dr i veway, passing under a j asmi ne- covered
trellis, whi ch ties into the gar age, t hr ough i nto the bri ght er
4 1 6
S HA P I NG ONE B UI L DI NG
l i ght of the t errace, whi c h f or ms a ki nd of ant er oom to
the mai n ent rance. T h e n this whol e t errace becomes a
ki nd of room. T h e trees, whi ch f or m its corners, also help
to emphasi ze its charact er as an OUTDOOR ROOM.
FARMHOUSE KITCHEN gives me the character of the
main room inside.
Ev e n t hough the cot t age wi l l be a wor kshop, and pl ace to
l i ve, it makes most sense to think of the inside as a FARM-
HOUSE KI TCHEN, wi t h a bi g table in the mi ddl e, chai rs
around it, one l i ght hung over the cent er, a couch or ar m-
chai r of f to one side . . . . Wh e n I start to i magi ne this,
and i magi ne ent eri ng it, I real i ze that it is mor e i mport ant
than I real i zed to keep it back, sl i ght l y, f r om the door, to
make somet hi ng out of the ENTRANCE ROOM that lies be-
t we e n e v e n t hough, in a bui l di ng as smal l as this, this
ENTRANCE ROOM ma y be shrunk to al most not hi ng. I
i magi ne c omi ng in, bet ween t wo seats, into a gl az ed pl ace,
wi t h l i ght c omi ng i n, and then passing t hr ough a second
door way, perhaps a LOW DOORWAY, into the mai n room
of the FARMHOUSE KI TCHEN proper.
CONNECTION TO THE EARTH and TERRACED SLOPE help
me to complete the way the building's outer edge is
formed.
O f course, the t errace gi ves the connect i on to the eart h.
But I have been t r yi ng to i magi ne how to make the edge
of the t errace, whe r e it meet s the eart h. I f the t errace i t -
self is made of tiles (l ai d either in eart h, or g r o u t n o t
4 1 7
T H E WA Y
sure y e t ) , the edgi ng coul d be a SITTING WALLb u t that
seems too f ormal , too e nc l os e dor perhaps better, it coul d
be made by a simple concret e bl ock surround. Th i s seems
a bit stark. I close my eyes, and see the slight step, wi t h
blocks that are filled wi t h r ock- gar den flowersthese
f or m the edge except at the f ew parti cul ar places wher e
there is an actual step to the path beyond.
T h e slope of the gr ound is not enough to need a TER-
RACED SLOPE; but there is a defi ni te fall of a f ew inches
f r om the back of the site to the f ront of the terrace. W e
decide to place a nat ural step, al ong the cont our line,
wher ever it makes s ens es o that we do as little earth
movi ng or filling as possible, and the house sits j ust the
wa y the land is.
As far as the connect i on to the earth :oes, there are still
O '
t wo big unanswered questions. Wh a t exact l y happens
around the little apple tree to the south? And what exact l y
happens al ong the west wal l of the bui l di ng, bet ween the
ent rance area and the bathi ng room arcade? I t is possible
that the place under the acacia tree mi ght be bl ocked com-
pletely by a WINDOW PLACE whi cl i f or ms part of the en-
trance, or falls j ust inside it. I n this case one woul d not
be able to wal k al ong this edge of the bui l di ng, and coul d
onl y get to the bat hroom arcade by goi ng i nto the build-
i ng. No t sure if this is r i ght ; perhaps too tight.
WI NDOW PLACE and ENTRANCE ROOM fix the detailed
arrangement of the entrance.
I n order to make progress on all this, we we nt out to the
site, and l ooked around, t ryi ng to i magi ne all this more
4 1 8
S HAP I NG ONE B UI L DI NG
concret el y on the l and. W e started especially wi t h the
f ront door. Shoul d it be angl ed, to face the t errace, or
f aci ng west ( i nt o the acacia tree, or f aci ng south, t owar ds
the g a r a g e ) ? Al t ho ug h f aci ng south is less di rect than
whe n it is angl ed, it seems be s t i t creates a sl ight sense of
ENTRANCE TRANSITION, doesn' t al l ow such a compl ete
vi ew of the inside f r om the t e r r a c e i t uses the little apple
tree very ni cel y, to one si de: and it l eaves the WINDOW
PLACE to the west , j ust perf ect , inside the f ront door,
hel pi ng to f or m the ENTRANCE ROOM. W e staked it out,
wi t h seven- f oot - hi gh stakes, so we coul d begin to feel its
presence. T h e r e is a need to prot ect the apple tree and
wi l d oni ons, f r om t r a mpl i ng s o it wi l l be nat ural to make
a l ow wal l at a sl ight angl e, perhaps c ur vi ng out , to f or m
the approach i nto the do o r t hi s wi l l make the FRONT
DOOR BENCH.
ALCOVE then generates a further differentiation of the
inside room.
No w we stood inside the r oom, l ooki ng t owar ds the door,
t owar ds the count er area at the back, to make the actual
shape of the room wor k out j ust ri ght. T h e WINDOW
PLACE to the ri ght of the door wor ks beaut i f ul l y. Anot he r
ALCOVE to the l ef t of the door, on the l ef t - hand side of
the apple tree seems j ust ri ght too.
Now STAIRCASE BAY shows us how to stake out the
four corners of the stair, so that we get a realistic look
at its effect on the room.
4 1 9
T H E WA Y
I i magi ne it ver y steep ( 7 feet hori zont al run, for a cl i mb
of 8 ' 6 " ) , and no more than about 2 feet wi de s i nc e it
onl y l eads to a sl eepi ng l of t . W e f i x the top of the stair, by
knowi ng that the back count er of the ki tchen wi l l come in
three feet f r om the nort h face of the bui l di ng, and that the
upper story wi l l rise di rectl y f r om that line. I f the sl eepi ng
l of t is 7 feet nort h-t o-sout h, enough for a b e d a n d the
stair comes up inside it, wi t h a 3 f oot l andi ng at the top,
this al l ows us to fix the top of the stair 6 or 7 feet south
of the property l i ne a nd the bot t om of the stair 14 feet
south of the line. Wh e n we l ook at the stair, it bl ocks the
south-east al cove a bi t s o we splay the al cove, around the
apple tree, to connect it better wi t h the mai n room. A
splay of t wo feet makes an enor mous di f f erence. W e stake
it out also, and i magi ne a wi ndow in it, l ooki ng west
t owar ds the apple tree (WINDOW PLACE) .
THI CK WALLS helps me define the inside edge of the
farmhouse kitchen.
No w, st andi ng in the mi ddl e of the room whi c h is to be
the FARMHOUSE KITCHEN, I i magi ne anot her seat or
closets under the stai r; perhaps a wi ndo w under the stair
too, l ooki ng t owards the gar den to the east: smal l wi ndows
over the count er to the nort h whi c h f or ms the mai n THICK
WALL. Ta l k i n g about the second story, we real i ze that the
l oad of its southern wal l wi l l fal l ri ght over the vaul t
whi ch f or ms the FARMHOUSE KITCHEN: it wi l l probabl y
need a rib in the mi ddl e, and this rib can gi ve us a nice
cent er to the r oom, a place to hang a l i ght (POOLS OF
LI GHT) .
420
S HA P I NG ONE B UI L DI NG
CEI LI NG HEI GHT VARI ETY completes the upstairs and
the downstairs.
T h i s pat t ern is satisfied al most aut omat i cal l y by wha t has
gone bef ore. Fo r the mai n r oom, I i magi ne one big vaul t ,
perhaps 8 ' 6 " hi gh at the cent er. T h e back wal l , wher e the
ki t chen count er is, the mai n al cove to the south, and the
wi ndo w pl ace by the door, all spri ng of f the peri met er
beam, whi c h wi l l be at about 6 ' 6 " g o i n g do wn to 5 ' 6 "
or 5 ' o " . Upst ai rs the sl eepi ng r oom is l ow a ny wa y , under
the r oof ; and it has a still l ower secti on, over to the west ,
wher e the bed is in an al cove whi c h has no mor e than a
4 ' 6 " to 5 ' o " cei l i ng.
All in all
}
the design took about a week of continuous
on-and-off thought.
I mul l ed each gr oup of patterns, in t urn, as the notes
s how. Somet i mes, I spent as muc h as an hour t hi nki ng
about one pat t ern. I n these cases, I di dn' t act i vel y thi nk
how to do the pattern for an hour. I did all ki nds of other
thi ngs, drove the car, pl ayed musi c, ate an appl e, wat er ed
the gar den, et c. , wai t i ng f or the pat t ern to f or m itself in
my mi nd, by t aki ng on shape appropri ate to this part i cul ar
site and probl em. I n ma ny of these cases, I got the key i n-
si ght by wal ki ng into the desi gn, so f ar as it was compl et ed,
and then aski ng, wha t woul d I see over there if this pat-
tern I am now t hi nki ng about wer e in the bui l di ng? Ve r y
of t en, the ans wer came al most i mmedi at el y. But it onl y
came if I was real l y there, coul d touch and smel l what
was around me.
42 1
T HE WA Y
And I never made a drawing of the building.
T h e design was done compl etel y in my mi nd.
Onl y in the fluidity of your mi nd can you conceive a
whol e. As the design unfol ds, and the new patterns are
brought into play, accordi ng to the order of the l anguage,
the entire design has to shift and resettle itself in your
mi nd wi th every new pattern. Eac h new pattern in the
sequence transforms the whol e design created by the pre-
vious pat t ernsi t transf orms it as a whol e, it shakes it up,
and realigns it.
Thi s can only happen if the design is represented in an
utterl y fluid medi um; it cannot happen in any medi um
where there is the slightest resistance to change. A dr aw-
i ng, even a rough dr awi ng, is very r i gi di t embodies a
commi t ment to details of arrangement far beyond what
the design itself actual l y calls f or whi l e it is in an embry-
onic state. Indeed, all the external media I k no ws a nd,
cl ay, drawi ngs, bits of paper l yi ng on the floorare
all far too rigid in this same sense. T h e onl y medi um
whi ch is truly fluid, whi ch al l ows the design to gr ow and
change as new patterns enter it, is the mi nd.
Representati on there is fluid: it is an i mage, yet an
i mage whi ch contains no more than essenti al sand it can
change, almost of its own accord, under the t ransf orm-
i ng i mpact of a t hought about a new pattern. Wi t hi n
the medi um of the mi nd, each new pattern transforms
the whol e design, almost by itself, wi thout any special ef -
fort.
422
S HA P I NG ONE B UI L DI NG
Imagine trying to build sentences by shuffling words
around on a piece of tracing paper.
Wh a t terrible sentences. T h e act of speech is a spont ane-
ous, and i mmedi at e response to a situation. T h e more
spontaneous it is, the more di rectl y rel ated to the situation,
and the mor e beauti f ul . T h i s spontanei ty is gover ned by
the rules of Engl i s h whi ch are disciplined and or der ed;
but the use of these rul es, and the creati on of a totality
f r om t hem, takes place in the i mmedi acy and fluidity of
your o wn mi nd.
Just so wi t h a pattern l anguage. T h e pat t erns are dis-
ci pl i ned; and the order of the l anguage is disciplined. But
you can onl y use these patterns in that order if you are
wi l l i ng to combi ne the discipline they gi ve you, wi t h the
spontanei ty and i mmedi acy of di rect experi ence. Y o u can-
not create a desi gn by pat c hwor k, on pieces of t raci ng
paper. Y o u can onl y create it, as if it wer e a real experi -
ence of a real bui l di ng: and that you can onl y do in your
mi nd.
It is only in the mind's eye, eyes shut, not on paper,
that a building can be born out of the vividness of
actual experience.
I n the cot t age I have been descri bi ng, we even built the
bui l di ng wi t hout the use of dr a wi ng s s i mpl y by st aki ng
out the bui l di ng, as I saw it in my mi nd' s eye, and then
usi ng a pattern l anguage for const ruct i on, in the wa y de-
scribed in chapt er 23.
423
T HE WA Y
Of course, this little experimental building is still im-
mensely far from the great beauty and simplicity of
the houses which are shown at the beginning of the
chapter.
It wi l l take years more of experi ment, wi th ways of build-
i ng, before we can do that.
It is too loose, too i nf ormal , the construction patterns
whi ch control its detailed shape are not harmoni ous
enough, not disciplined enough . . .
Ye t still, this buil ding has just the begi nni ng of a spirit, a
hint of a touchi ng quality, whi ch is at least a f ew steps
down the road.
Anyone can use a language to design a building in
this way.
No matter who does it, the buildings whi ch are made
like this wi l l be ordi nary and natural , because each part
in the design is f ormed by its position in the whol e.
It is a primitive process. T h e primitive f armer spends
no time " des i gni ng" his house. He thinks briefly where
and how to build it, and then sets about buil ding it. T h e
use of the l anguage is like that. T h e speed is the essence.
It takes time to l earn the l anguage. But it takes no
o o
more than a f ew hours or days to design a house. If it
takes l onger, you know it is tri cky, " desi gned, " and no
l onger organi c.
And it is just like English.
425
S HAP I NG ONE B UI L DI NG
Wh e n I speak Engl i s h, the sent ences f or m t hemsel ves in
my mi nd as fast as I can say t hem. An d this is true of pat -
tern l anguages also.
T h e qual i ty that makes a bui l di ng feel as t hough it has
been there f or a thousand years, the qual i ty that makes it
feel that it has f l owed like wr i t i ng f r om a pen, comes al -
most aut omat i cal l y whe n I rel ax my mi nd, and let the
l anguage generat e the bui l di ng f reel y there.
I still remember the -first time I used a pattern lan-
guage in this way. I found myself so completely
caught up in the process that I was trembling. A hand-
ful of simple statements made it possible for my mind
to flow out and open
y
through themand yet
y
al-
though the house which came was made by me
y
born
of my feelings
y
it was at the same time as though the
house became real
y
almost by itself
y
of its own voli-
tion
y
through my thoughts.
It is a fearsome thing
y
like diving into water. And
yet it is exhilaratingbecause you aren
y
t controlling
it. You are only the medium in which the patterns
come to life
y
and of their own accord give birth to
something new.
426
CHAPTER 22
SHAPING A GROUP OF BUILDINGS
In the same way
y
groups of people can
conceive their larger public buildings
y
on
the ground
y
by following a common pat-
tern language
y
almost as if they had a
single mind.
We know from chapter 21 y that an individual ferson>
can create a building in his mind simfly by letting a
sequence of patterns generate ity on the site.
Now we go one stef furthery and see how a grouf
of feofley also on a sitey and with a common lan-
guagey can use the same frocess to design a larger
building.
It is often said that no group of people can create a work
of art, or anything whi ch is whol e, since di fferent people
pull in di fferent directions, and make the end product a
compromise whi ch has no strength.
T h e use of a shared pattern l anguage solves these prob-
lems. As we shall now see, a group of people who use a
common pattern l anguage can make a design together
just as wel l as a single person can within his mind.
Here is an exam fie of a clinic.
I t is a psychiatric clinic to serve a rural population of about
50, 000 in Cal i forni a. Th e building has about 25, 000
square feet of internal space, and sits on a piece of land
whose area is about 40, 000 square feet in the middle of
an existing hospital. T h e building was designed by a team
whi ch included the director of the clinic ( Dr . Ryan, a
psychiatrist), several of his staff members who had years
of experience worki ng with patients, and t wo of us f rom
the Cent er for Envi ronment al Structure.
Again the frocess begins with a pattern language.
43
2
S HA P I NG A GROUP OF BUI L DI NGS
W e sent Dr . Ry a n a sequence of patterns f r om the pri nted
pat t ern l anguage whi c h we t hought mi ght be usef ul .
W e asked hi m to pick those he t hought wer e r el evant ;
get rid of those whi c h wer e i r r el evant ; and asked hi m to
add what ever special patterns or new " i de as " whi c h seemed
to be mi ssi ng, i ncl udi ng, of course, those special parts or
" pat t er ns " specific to a cl i ni c. Tho s e ne w ones whi c h he
added are mar ked wi t h asterisks bel ow.
Af t e r our first discussion, we had a l anguage of some
f ort y pat t erns:
BUILDING COMPLEX
NUMBER OF STORIES
SHIELDED PARKING
MAIN GATEWAY
CIRCULATION REALMS
MAIN BUILDING
PEDESTRIAN STREET
*ADULT DAY CARE
* ADOLESCE NT DAY CARE
*CHILDREN' S DAY CARE
^OUTPATIENT
*I NPATI ENT
^ADMINISTRATION
^EMERGENCY
FAMILY OF ENTRANCES
SOUTH FACING OUTDOORS
WINGS OF LIGHT
POSITIVE OUTDOOR SPACE
HALF-HIDDEN GARDEN
HIERARCHY OF OPEN SPACE
433
T H E W A Y
COURTYARDS WHI CH LIVE
CASCADE OF ROOFS
SHELTERING ROOFS
ARCADES
PATHS AND GOALS
PEDESTRIAN DENSITY
I NTI MACY GRADIENT
COMMON AREAS AT THE HEART
ENTRANCE ROOM
TAPESTRY OF LIGHT AND DARK
FARMHOUSE KI TCHEN
FLEXI BLE OFFICE SPACE
SMALL WORK GROUPS
RECEPTION WELCOMES YOU
A PLACE TO WAI T
SMALL MEETING ROOMS
HALF- PRI VATE OFFICE
LI GHT ON TWO SIDES OF EVERY ROOM
BUILDING EDGE
OUTDOOR ROOM
THE SHAPE OF INDOOR SPACE
CEILING HEI GHT VARIETY
Gradually this language changed.
A s mor e discussion t ook pl ace, peopl e' s i deas about the pat -
t erns whi c h the cl i ni c shoul d cont ai n, c ha ng e d. T h e y de-
ci ded t hat I NPATI ENT wa s uni mpor t ant , si nce the near by
hospi tal wo u l d t ake car e of ov e r ni ght pat i ent s. T h e n it
t ur ned out t hat the cl i ni c needed a si ngl e ar ea f or oc c u-
434
SHAPI NG A GROUP OF BUI LDI NGS
pational t herapyand that this woul d become the MAIN
BUILDING.
Dr. Ryan decided that there ought to be a GREEN-
HOUSE as fart of this MAIN BUILDING: patients could
help plants to grow> and then transplant them into the
gardens y and look after the gardens.
T h e n the discussion of the GREENHOUSE made the HALF-
HIDDEN GARDENS seem much more important, and they
became an essential part of the conception of the building.
Lat er, when we realized the importance of the CHIL-
DREN'S HOME, a place at the entrance of the clinic where
parents could leave children while they were being treated,
we introduced STILL WATER, and a FOUNTAIN where the
children could play and splash about.
The r e was some debate about COMMUNAL EATING;
finally it was agreed that this pattern should be included,
because the advantages of staff and patients eating lunch
together regularly seemed so essential. Onl y the fact that
each person should cook for the others in turn was not
included, since it seemed impractical.
Every aspect of the clinic's lifey was discussedy and set-
tledy in the medium of patterns.
T h e l anguage has the medi um in whi ch people worked
out their disagreements, and in whi ch they built a com-
mon picture of the building and the institution as a whol e.
Usual l y people have a great deal of trouble when they
435
T HE WA Y
try to define the future of an institutionbecause they
have no l anguage, no medium, in whi ch they can forge
their definitions, no way of gradual l y building up agree-
ments, no way of gradual l y settling disagreements.
But with the pattern l anguage as a base, the group of
people gradual l y come to see themselves, and their activi-
ties, and their envi ronment, as one t hi ngas a whol e.
And finallyy when everyone agreed about the 'pattern
languagey we were ready to begin design.
At this stage, the people who were goi ng to run the clinic,
had a shared vision, a vision not only shared in its inten-
tions, in its broad outline, but shared in the details too.
As a communi ty, they knew, now, just exactly what they
wanted, how it was goi ng to work, what kind of places
there woul d be in it, . . . everythi ng in short, they
needed to know, in order to begin design.
Then we began the design itself.
It took a week, Monday to Fri day, out on the site itself,
wal ki ng around parked cars and obstacles, overcoats against
the fog, wal ki ng, wal ki ng all day l ong, cups of cof f ee,
crazy danci ng around, as the building took shape, chalk
marks on the ground, stones to mark corners. People
wondered what on earth we could be doing out there in
the fog, wal ki ng around, all day l ong, for so many days.
We began with BUILDING COMPLEX.
436
S HA P I NG A GROUP OF BUI L DI NGS
T h e first pat t ern. W e sat, at fi rst, around a table in the
nearby heal th cent er. Ho w is this part i cul ar clinic goi ng to
ref l ect the bui l di ng compl ex pat t ern? T h e pattern requi res
that any bui l di ng be made up of visible component s, whi c h
correspond to social gr oups: a n d i f the compl ex is at a
l ow de ns i t y t hat the component s act ual l y be separate,
connect ed by arcades and passages.
Fi rst of all, Dr . Ry a n said, I see ma ny many little cot -
tages, each one i ndi vi dual and personal . H o w ma ny do
you see? We l l , perhaps 30 separate cot t ages.
T h e enti re bui l di ng compl ex wi l l have 24, 000 square
f eet. I poi nt out that if there are 30 cot t ages, each one
wi l l on the aver age have about 800 square f e e t pe r haps
25 X 3 0 a n d that some of t hem wi l l be even smal l er.
T h i s di dn' t sound ri ght. T h e r e was some discussion a mo ng
the st af f . T h e n he said, We l l , let' s perhaps say 6 or 8
separate bui l di ngs, cl ustered and connect ed, but i denti f i -
able and separate.
Wi t h this idea cl ear in our mi nds, we we nt out doors to
the site itself.
Next we placed the MAIN GATEWAY and MAIN EN-
TRANCE to the building complex.
O O
(J '
437
T HE WA Y
Al l the next patterns we took on outdoors. We wal ked out
into the f og in overcoats, and looked around. I asked: Sup-
pose there is one mai n entrance to this buil ding compl ex:
Whe r e is it? Cl ose your eyes; i magi ne; wher e do you
see it?
Is it al ong the mai n street? Is it on the corner?
Dr . Ry a n said: I see it half way al ong the dri veway whi ch
leads back from the road to the mai n hospital. I asked
then: We l l , let's decide exactl y where it is. T h e pattern
says that it must be immediate and visible f rom all possible
lines of approach. If it is in this position, then there are
t wo lines of appr oachone f rom the main road, wal ki ng
back; one f rom the hospital parki ng lot, if you have driven
in, parked your car, and are now wal ki ng f or war d, to-
wards the road agai n. Le t us go to both these places and
try to i magi ne the best position for it.
Fi rst, all six of us stood at the road end of the dri veway,
and looked back. I wal ked to the hal f way point and said:
Imagi ne that I am at the ent r ancei s it ri ght now? I
moved a f ew f e e t no w? moved again . . . now? T h e y
said stop, go back, f or war d a bi t t her e was very great
agr e e me nt and I made chalk marks at the nearest and
the farthest points. T h e y were about 10 feet apart onl y,
in a total l ength of 200 feet.
T h e n we went to the other e ndt he parki ng lot, and
did the same thing. Agai n I made chal k marks whi ch now
showed the best place to make the entrance feel good, for
someone comi ng to it f rom his car. T h e t wo sets of chal k-
marks were about ten feet apart: less than the size of the
entrance itself.
438
SHAPI NG A GROUP OF BUI LDI NGS
No w the position of the main entrance was fi xed. I ex-
plained that we woul d now mark that, and that from
now on it woul d be a given about the desi gnt hat we
woul d no l onger think about movi ng the entrance, in view
of later t hi ngsbut woul d let the design grow outward
f rom this decision. A little f r i ght eni ngwhat if things
don' t work out?
Next> with the main entrance fixed, we started to de-
fine the CIRCULATION REALMS.
I explained that this pattern required a single, simple
pedestrian area, opening directly f rom the mai n entrance,
and, further, a series of individual pedestrian realms open-
ing off this one main path.
We stood at the main entrance, and wondered how
this mi ght be.
At the far end of the site, opposite the entrance, were
four magni f i cent trees. It seemed natural, then, to make
the main path go down towards those trees. And, wi th
several small buildings opening off this path, some to the
l eft, some to the right, it was easy to imagine a series of
smaller paths, more or less at right angles to the main
path, opening off it.
439
T HE WAY
Within the circulation realms we placed the MAIN
BUILDING.
Thi s pattern calls for a main building, in any group of
buildings, to act as a heart and focus: and requires that this
building have paths, tangent to it, wi th vi ews into the in-
side, so that everyone who moves about the building com-
plex is connected to it all the time.
We spent some time discussing what part of the clinic
might most naturally function as a main building. Fi nal l y,
we agreed that the so-called occupational therapy building
wh e r e patients do various kinds of creative w o r k
woul d make the best " hear t , " and decided to make a l arge
building, wi th a specially high roof, right in the middle,
for this reason.
Theny outside the main building} an ACTIVITY NODE.
440
S HA P I NG A GROUP OF BUI L DI NGS
I f there wa s to be a node of act i vi t y, inside the bui l di ng
compl ex, it seemed nat ural to pl ace it j ust at one of the
pl aces whe r e the mai n " s t r eet " is crossed by t wo of the
wi de " s t r e e t s " a nd whe r e several i mport ant bui l di ngs
meet ar ound it. W e deci de to open this crossi ng, to have a
f ount ai n t here, and to make doors f r om the mai n bui l di ng,
f r om the admi ni strati on bui l di ng, and f r om the child care,
wi t h chi l dren pl ayi ng, all open i nto this node.
Around the activity node at the key points in the cir-
culation realmsy we placed RECEPTION, ADMINISTRA-
TION, OUTPATI ENT, ADULT DAY CARE, ADOLESCENT DAY
CARE, CHILD DAY CARE.
N o w we pl aced the vari ous di f f er ent bui l di ngs on the site.
Dr . Ry a n had rat her cl ear ideas al r eady, about the posi-
tions of these bui l di ngs. He showed us whe r e he f el t they
4 4 1
T HE WAY
ought to go, and we discussed it, wal ki ng about the site.
One question came up. Ther e were goi ng to be t wo
outpatient t e a ms Dr . Ryan had placed them to the right,
after, just after the entrance, since that seemed to him the
most natural position for the buildings whi ch woul d be
used most.
Since there were to be two teams, each with its own
identifiable place, we thought about the circulation realms.
We all stood at the activity node and asked ourselves how
they mi ght be placed, so that they woul d be clearly dif-
f er ent so that a patient woul d know whi ch one was
"hi s. "
Several of the staff stood, with their eyes closed, and
suggested that if there were a courtyard, and the t wo
teams opened off this courtyard to the left and right re-
spectively, it woul d be clear and simple.
Theny in a special placey near the main entrancey COM-
MUNAL EATING.
Dr . Ryan agreed that the process of sharing food is one of
the most fundamental in any human group. We discussed
the various ways this mi ght happen, and mi ght help pa-
tients to become more emotionally stable.
He and the chief administrator for the clinic decided
finally that it woul d make most sense to place a kind of
cafe, in the first garden on the left, attached to the library
and administrative services, visible f rom the activity node
and fountain at the main cross roads within the proj ect.
442
S HA P I NG A GROUP OF BUI L DI NGS
Now} within the individual building areas, we made
SOUTH FACING OUTDOORS, WINGS OF LIGHT, POSITIVE
OUTDOOR SPACE.
N o w came the most di ffi cul t part of the process. A t this
stage, we had some r ough idea of whe r e the vari ous bui l d-
i ngs we r e ; and some r ough i dea of the mai n paths and
mov e me nt bet ween bui l di ngs. N o w came the mome nt
whe n the act ual position of bui l di ngs, and the shape of
the out doors had to be f i xed. T h i s is al ways one of the
most di ffi cul t moment s in the l ayout of a l ar ge gr oup of
bui l di ngs. It is tense, and rat her ner ve- r acki ng. Unt i l this
has been accompl i shed, there is a di agr ammat i c qual i ty
about the t hi ngs whi c h have been laid out : the people
wal ki ng about ask t hemsel ves whe t he r there is any act ual ,
concret e wa y of l ayi ng out the bui l di ngs whi c h gi ves t hem
sensible shapes, and whi c h gi ves sensible shapes to the out -
door spaces too.
As al ways, ever yone became rat her nervous. I ndeed, in
this parti cul ar case it was especially di ffi cul t. W e spent an
af t er noon, not knowi ng exact l y how to ar r ange the bui l d-
i ngs, we nt home and slept on i t; and the next mor ni ng,
f i nal l y f ound a way of doi ng it whi c h made it seem simple
and wor kabl e.
I t hought , fi rst, that every gar den, in bet ween the
bui l di ngs, needed to be cupped t owar ds the south. T h i s
made the l ef t - hand side and the r i ght - hand side, whi ch
so f ar had seemed symmet r i cal , become asymmet r i cal .
I t was compl i cat ed by the f act that all these gar dens, or
court yards, needed to be connect ed to the mai n pedestri an
443
T HE WA Y
pat hs o there woul d be vi ews of flowers, and trellises,
glimpsed f rom the path, inviting people into the back
spaces.
We real i zed, finally, that the double ef f ect of the con-
nection to the main path, and the f aci ng south, coupled
wi th the idea of buildings whi ch were not too wi de at any
point, so that they woul d give natural dayl i ght in all
rooms, led us to a series of rough T- s haped buildings
placed to the north of south-f aci ng courtyards. At this
moment , when, for the first time, we had a l ayout of the
building space and open space, we finally knew we had
a group of buildings that could be built.
The r e was no doubt, at any stage, that these patterns
woul d make somethi ng buildable. But it is wort h record-
i ng the f act that to those people who had not seen the
process in action, it was very remarkabl e that the issue
woul d resolve itself wi thi n the loose and slightly i rregul ar
array of buildings whi ch had been f ormed so far.
Thi s is an extreme exampl e of the f ear, the fear of
pl ungi ng in, whi ch people must live wi t h, whe n they are
l etting their l anguage generate a buil ding for t hem. It
was onl y because of the conf i dence that it woul d wor k,
that everyone al l owed the thi ng to remai n fluid for so
l ong.
444
SHAPI NG A GROUP OF BUI LDI NGS
Of course, it woul d have been possible to make some
formal arrangement of buildings and open spacesome
formal geometrical arrangement, very much earlier. A
formal arrangement woul d have guaranteed that there
woul d be some feasible way of placing the buildings.
But it woul d have killed the spirit of the buildings.
It woul d have killed that subtle, rambling balance of
coherence, and incoherence, whi ch comes f rom the fact
that every building is unique, accordi ng to its position in
the whol e.
Within the individual buildings> and in the proper
places in the circulation realmsy we placed a FAMILY
OF ENTRANCES.
Fi nal l y, to make these buildings coherent, not only wi th
respect to space and vol ume, but also for the person
comi ng in, we dealt with the family of entrances: the pat-
terns whi ch require that, in some fashion, all the di fferent
buildings have similar entrances, whi ch are clearly visible,
as a group, f rom the main pedestrian paths, and whi ch
are also members of a fami l y, so that one sees " at a gl ance"
the way they span the ful l range of possible entrances.
We wal ked about the site, whi ch was by now laid out
in some detail, with chalk marks and stones, and asked
ourselves what we woul d like to see from various points,
what we woul d like to see of entrances. I talked about the
FAMILY OF ENTRANCES pattern: then asked everyone to
stand in di fferent places in the site, wi th their eyes closed.
Imagi ne, now, that the FAMILY OF ENTRANCES pattern is
solved as perfectly as you can i magi nei t is ideal; it is
445
T HE WAY
what you dream about when you think this pattern is
there, in the most beautiful form it can be.
One person suggested "a whol e lot of porches"; each
one with its seats, so people can wait, outdoors, for ap-
pointments, a couple of steps up. Ni ce wood c ol umns
each one sticking f orward from its respective building.
At this stage, the basic layout of the building complexy
as a complex} was complete.
T h e decisions which had been made on behalf of the clinic
"as a whol e" were finished. No w it was time to go down
into the details of the individual buildings, and the indi-
vidual gardens.
For this, we asked the doctors and the staff to rearrange
themselves, in small groups, each one responsible for one
of these bui l di ngsso that the individual buildings could
be designed by the people who knew most about them,
most about what was goi ng to happen there.
Now different specialists on the clinic staff worked out
the details of each different building.
446
SHAPI NG A GROUP OF BUI LDI NGS
T h e doctors concerned most with children designed the
building for child treatment and for adolescent treatment;
the social workers most concerned wi th outpatients de-
signed the outpatient areas; the administrative officer of
the clinic designed the administrative building.
The director of the clinic himself designed the details
of the large central building.
He put child care at one end, just inside the entrance, so
that playing children woul d be visible, and children comi ng
to it woul d feel comfortable and unafraid (as specified by
VISIBLE CHILD CARE). He placed a l arge greenhouse at
one end of the main social hall, with the idea that patients
could learn to take care of plants, and might, in the end,
take care of all the plants in the clinic's gardens (OCCU-
PATIONAL THERAPY). He made alcoves inside the main
social hall where small groups can gather to talk (FAMILY
ROOM ALCOVES agai n) ; and an arcade outside, al ong the
main street, to create a zone of social space neither en-
tirely private nor entirely public (as directed by ARCADE).
Each fart of the building was designedy in detaily by
a frocess like the frocess described in chafter 2 r.
T h e patterns whi ch affected the design included, for ex-
ample, SHORT CORRIDORS, whi ch explains how l ong corri-
dors in buildings make people feel i nhuman; RECEPTION
WELCOMES YOU, whi ch says that a building for patients
should not have a formal reception counter, but instead a
more i nformal arrangement with comfortable chairs, a
447
fireplace, and coffee where people can be made to feel at
home; FARMHOUSE KITCHEN, a pattern most relevant to
houses, whi ch shows how a kitchen with a big table is one
of the most comfortable places for communal di scussi on
this pattern was used in three of the day treatment pro-
grams; FLEXIBLE OFFICE SPACE, whi ch requires a l arge
number of small workrooms and alcoves, instead of the
continuous open workspaces typical of modern office build-
ings. FAMILY ROOM ALCOVES, also most often used in
houses, shows how small, low-ceilinged alcoves off the
edge of l arger rooms give people a chance to sit alone, or
in twos, and be quiet, wi thout l eavi ng the l arger group al-
together.
And we see then how a group of people can design a
complex building.
448
Once they agree about the l anguage, the actual emergence
of the form is simple and fluid. Wh e n a group of people
try to do something together, they usually fail, because
their assumptions are different at every stage. But with a
l anguage, the assumptions are almost completely explicit
from the start.
Of course they no l onger have the medi um of a single
mind, as an individual person does. But instead, the group
uses the site "out there in front of t hem, " as the medium
in whi ch the design takes its shape. People wal k around,
wave their arms, gradual l y build up a common picture of
the building as it takes its shapeand all, still, without
maki ng a drawi ng.
And, it is for this reason, that the site becomes so much
more important for a group.
The site speaks to the peoplethe building forms it-
selfand people experience it as something receivedy
not created.
449
SHAPI NG A GROUP OF BUI LDI NGS
And they are able to visualize the building, right before
their eyes, as if it were already there.
Th e idea that "ordi nary" people cannot visualize a
building is completely false.
Th e building grows, and comes alive, before their very
eyes.
A f ew sticks in the ground, or stones, or chalk marks,
are enough to bring the image to mi nd.
And then the building can be built directly f rom these
marks.
Of coursey this buildingy like the experimental build-
ing in chapter 21 y is still immensely shallower than
the buildings in the photographs which start this
chapter.
It has great beauty of layout. But in its details of con-
struction it still falls far behind. Indeed, in its construction
it is completely spoiled.
For reasons outside our control, it was necessary that
this particular building, once laid out, was then "detai l ed"
by ordinary processes. It was taken to the drawi ng board,
by people who had not laid it out, far from the site, and
given mechanical " dr a wn" details, quite inappropriate to
its design . . . until it became, in the end, no di fferent
f rom a thousand ordinary buildings of our time.
In short, it was almost destroyed, because it was not
built in the right way. At first I hesitated, I was not sure
whet her to write this, or whet her to include the picture,
because it is so sad and so depressing. But then I realized
how essential it is to include it: because many people may
451
T HE WAY
be wi l l i ng to lay out a building in the way I have de-
scribed, and will then try to get it built from drawi ngs.
The life, pulse, substance, subtlety of the building can
only be retained, if it is built, in the same way that
it has been designedby a sequential and linguistic
process, which gives birth to the building slowly, in
which the building gets its final form during the ac-
tual process of construction: where the details, known
in advance as patterns, get their substance from the
process of creating them, right there, exactly where
the building stands.
In short, a building laid out by a pattern l anguage process,
and whi ch comes to life because of it, will die again, quite
certainly, when it is built, unless the process of construc-
tion is the sameunl ess, that is, the same spirit which
generated rooms that are just right, entrances where they
should be, light comi ng f rom the right directions . . . is
carried on into the details, and also shapes the columns,
and the beams, the wi ndow frames, the doors, the vaults,
the colors and the ornament as wel l .
In the next chapter we shall see how such a construc-
tion process works.
Yet even this clinic, crude as it is in its construction,
already touched the hearts of the people who laid it
out.
452
SHAPI NG A GROUP OF BUI LDI NGS
In earlier chapters I have described, in theory, why the
active use of a l anguage is so important to a person. It is
because it is the only process in whi ch he is able to make
his picture of the worl d solid and act ual hi s feelings are
embodied in the active concrete manifestation of his l an-
guage : he feels his worl d as whol e; it comes f rom within
him, and is then around him, physically.
In the case of this clinic, we observed this process in
fact.
Dr. Ryan told us, after his clinic was built, that this
one week he sfent with us, shaping the building, was
the most important week he had spent in five years
the week in which he had felt most alive.
No w, years later, seeing the building made r eal even
though he has since mov e dhe remembers that week,
standing in the fog, maki ng chalk marks on the ground as
we laid the building out, talking about the place for the
entrance, the place for the greenhouse, the places where
people could sit, the fountain, the small gardens, the
rooms, the arcadeshe remembers this week as the best
week in five years of his worki ng life.
The simple process by which people generate a living
building, simply by walking it out, waving their arms,
thinking together, placing stakes in the ground, will
always touch them deeply.
453
T H E W A Y
It is a moment when
y
within the medium of a
shared language
y
they create a common image of their
lives together
y
and experience the union which this
common process of creation generates in them.
454
CHAPTER 23
THE PROCESS OF CONSTRUCTION
Once the buildings are conceived like
this
y
they can be built
y
directly
y
from a
few simple marks made in the ground
again within a common language
y
but
directlyy and without the use of drawings
Suppose noWy that you have the layout of a building
doney according to the processes described in the last
two chapters. It happensy as we have seeny with very
great ease.
Now we come to the actual building of the building.
Agai n, just as before, the process is sequential. Onl y now
the patterns operate not on a mental image, but on the
building itself, as it is being built. Each pattern defines an
operation, which helps to differentiate, and to complete,
the building as it gr ows: and when the last patterns are
introduced into the gr owi ng fabric, the building is com-
plete.
Agai n, the patterns operate upon the whol e: they are
not parts, whi ch can be addedbut relationships, which
get imposed upon the previous ones, in order to make
more detail, more structure, and more substanceso the
substance of the building emerges gradual l y, but al ways as
a whol e, at each stage of its growt h.
Supposey to start withy that we have used a pattern
language to lay out a rough scheme of spaces for a
building.
And suppose that we have captured this rough scheme on
paper, with a rough pencil sketch, or on the ground, with
stakes, or sticks and stones.
In order for the building to be alivey its construction
details must be unique and fitted to their individual
circumstances as carefully as the larger parts.
459
T HE WAY
Thi s means that, like the larger parts, the details must be
very careful l y shaped according to their position in the
l arger whol e; and, although similar parts will have a
similar shape, no two of them will ever be exactly identical.
Look, for instance, at these drawi ngs. Accordi ng to the
room you start wi th, the exact spacing of the columns,
and hence the exact size of panels whi ch f orm the walls,
are di fferent in each case.
a
T h e rooms become alive because the details of the
col umn spacing fit the whol e. Any kind of irregularity in
the room can be accommodated without trouble by the
building process. T h e exact size and spacing of the build-
ing details, governed by the process, adapts itself to the
nature of the room.
The details of a building cannot be made alive when
they are made from modular farts.
Suppose, for example, that the building system contains a
panel whi ch is four feet wi de, whi ch fits together with
other panels on a four-foot grid. None of the many rooms
whi ch I have described could be built, exactly as they are,
with these four-foot panels.
T o build these rooms out of modul ar four-foot panels,
each of the rooms woul d have to be made into a perfect
square, sixteen feet by sixteen feet.
460
T HE PROCESS OF CONSTRUCTI ON
m cm ran
The modular panels tyrannize the geometry of the
room.
If the builder wants to build the room from modular four-
foot panels, he must change the size of the rooms, and
change their shape, to fit his panels.
In such a building system, it is impossible for a person
to create a plan whi ch reflects the l arger subtleties of site
or plan. Each plan will al ways be chopped and disfigured
to make it fit the building details.
And the beautiful variety in whi ch a hundred or a thou-
sand rooms can all be roughl y fi fteen by sixteen, yet no
two alike, is destroyed, and replaced by an endless repeti-
tion in whi ch hundreds and thousands of rooms are ex-
actly and identically the same.
Andy for the same reasony the details of a building
cannot be made alive when they are drawn at a draw-
ing board.
T h e details of a building cannot be alive when they are
specified in the form of worki ng drawi ngs, because these
drawi ngs al ways assume, for the sake of simplicity, that
the various manifestations of a given part are all identical.
T h e person who draws a worki ng drawi ng cannot
461
T HE WAY
draw each wi ndow, or each brick, di fferentl y, because he
has no basis for knowi ng the subtle differences whi ch will
be required. These only become clear when the actual
building process is already under way. So he draws them
the same, because he has no reason, sitting at the draw-
ing board, to make them different. But if the builder
builds according to a detailed drawi ng, and is constrained
by his contract to make the building exactly like the draw-
ing, he then makes the detail identical, to fol l ow the draw-
i ngand in the actual building this becomes dead and
artificial.
To make the building livey its patterns must be gen-
erated on the sitey so that each one takes its own shape
according to its context.
Consider, for instance, patterns like COLUMNS AT THE
CORNERS or FINAL COLUMN DISTRIBUTION, whi ch give
the proper col umn spacing so that columns act as stiffen-
ing for walls, in the most efficient way.
T o create these patterns correctly the builder makes a
room by placing columns at the four corners, and then
places extra columns along each wal l , at equal spacings, to
make col umn intervals somewhere between 4 and 6 feet,
according to the l ength of the wal l , then finally places a
beam al ong each wal l , over the columns.
Thi s process is an active representation of the patterns.
Each time the process is used, it will create a slightly dif-
ferent confi gurati on, accordi ng to the plan of the room.
462
T HE PROCESS OF CONSTRUCTI ON
Each room will have in it the skeleton of col umns and
beams whi ch embody the same patterns. Ye t , no two
rooms will have exactly the same panel sizes.
It is essential, therefore, that the builder build only
from rough drawings: and that he carry out the de-
tailed patterns from the drawings according to the
processes given by the pattern language in his mind.
Thi s is commonpl ace in nature. Wh e n the spider builds its
web, the process is standardi zed; but the parts which are
created are all di fferent. Each web is beautiful, unique,
perfectly adapted to its situation. Y e t it is created by a
standard process: and there is just one process. It is very
simple. Y e t this simple process interacts in an infinite
variety of ways wi th di fferent circumstances to produce
di fferent particular webs.
And just so in the building process which I shall now
describe. T h e individual processes are standardized, and
very simple. But the actual parts whi ch are produced are
infinitely vari oust hey are infinitely di fferent manifesta-
tions of the patterns whi ch the processes define.
The process for making vaults is standardbut the
individual vaults which it produces are unique.
Wh a t is standard is the process of weavi ng the basket,
placing the strips of wood, coveri ng wi th cloth, stiffening
the cloth with resin, coveri ng the resined cloth wi th light-
463
T HE WAY
wei ght concrete . . . . But the actual product, whi ch this
process produces, is di fferent each time, accordi ng to the
local circumstances.
And the process for making columns is standardbut
again the individual columns which it produces are
unique.
Nai l i ng the boards, placing in position, nailing on the
beam, filling with concret et hese are standard operations.
But each col umn that is made this way is di f f er ent i t is
made by a di fferent person, and reflects that fact. Perhaps
one is carved, another colored in an individual f as hi on
and each is in a di fferent position, has di fferent connec-
tions to its surroundings, and is therefore di fferent because
of that.
For concreteness' sake, I shall now give a sequence of
construction processes which will produce a building
in this manner.
Of course, the sequence of these processes is just an ex-
ampl e: it depends on a particular combination of materials.
But some similar sequence, with just the same increasing
definition, starting rough, and getti ng more precise as the
building gets finished, is necessary.
First, stake out the corners of the ground-floor rooms
and spaces.
T o make sure these stakes are right, it is often helpful to
464
T HE PROCESS OF CONSTRUCTI ON
use large stakes, bamboos, or old bits of wood, so that
several people can visualize the exact form and size of the
rooms, their relation to one another, and their relation to
the outdoors around them.
Wher ever there are outdoor spaces near the bui l di ng
terraces, paths, entrances, balconies, arcades, trellises,
garden wal l s . . . stake them out as wel l , so that you can
feel the indoors and the outdoors together.
It is very l i kel yal most cert ai nt hat you will modi fy
the building as you have so far conceived it. T h e stakes
are so vivid that you will almost certainly begin to see all
kinds of subtlety, whi ch you could not imagine before,
now that the stakes and rooms are actual, right out there
on the ground.
Modi f y the position of the stakes, a foot here, a foot
there, until they are as perfectly placed as you can i mag-
i ne; and until the layout of the rooms seems just exactly
right.
Erect the comer columns, and flace stiffening columns
as nearly as possible at equal sfacings, within the
framework given by the corner columns.
For buildings of di fferent heights, and on di fferent stories,
also accordi ng to the building height, these intermediate
columns need di fferent spacing, because the forces comi ng
down are di fferent. But on any one floor, their spacing
will be roughly constant.
However , on any one floor the spacing is only roughly
constant. T h e di fferent rooms have walls of different
lengths. Because they are not modul ar, the spacing of the
465
T HE WA Y
intermediate col umns is relaxed and natural ; it will vary
with the exact spacing of the corner columns.
Tie the columns together, with perimeter beams.
These beams then form the upper edge of every room.
The y make it possible to visualize the space of the rooms,
very cl earl y; make it possible to put the wi ndow frames
and door frames in position: and most important, they
provide the tension ri ng around each room, whi ch forms
the basis for the springing of the individual vaults.
Make the beams lowest around alcoves; higher around
ordinary rooms, and highest of all round the big and
public rooms.
Thi s will start the process of creating ceiling height vari-
ety. T h e alcoves can have beams as l ow as 5 feet or 5 feet
6 inches; the ordinary rooms as l ow as 6 feet or 6 feet 6
inches; the larger rooms perhaps 7 feet, 8 feet, or 9 feet.
In every case, the belly of the vault will add a foot or
two, or even more to the perimeter beam hei ght; accord-
i ngl y, the extent of the addition will depend on the span
of the vault.
Put in the window frames and door frames.
Yo u already have some rough idea of where you want to
put the doors and wi ndows, f rom the conception of the
building you have worked out in your mind.
But now the f rami ng of the rooms is up, you can see
466
T HE PROCESS OF CONSTRUCTI ON
just exactly where the openings should go. By mocki ng
them up, with bits of rough wood, you can modi f y them,
and adjust them, until they create the perfect relationship
between the inside and the outside, give the right views,
the right amount of light in the right places, the right sill
heights, the right heights for the doors, and the proper
breakdown of the l arger openings into smaller ones.
Now weave the baskets which will form the basis for
the vaults above each room.
Each room, no matter what its shape, can be roofed with
a simple vault, whose f ormwork can be woven out of thin
strips of flexible wood, perhaps a foot apart. Thi s basket
can adapt to all the small irregularities of the room, and
can even go round corners if it has to.
And shape the belly of the vaul t; it can be shaped to
give each room just the ceiling height it needs. For struc-
tural reasons the vault needs to be roughl y one sixth of
the span of the vault. But this sixth is quite variable, and
you may now adjust the exact curve of the vault to make
the room feel right for just exactly what you will be
doing there.
Put in the walls between the columns and, the window
jrames.
These wal l s can be made of any simple sheet material:
tiles, wood planks, hol l ow blocks, l i ghtwei ght sheets and
boards, cut and placed so they fill the gaps between the
columns and the wi ndow frames.
467
T HE WA Y
NLake half vaults for the stairs, so that each stair goes
up at just the proper angle in the bays reserved them.
Th e stairs can be conceived as lying over vaults. T h e steps
will be filled in, over a vault, or pair of vaults, or series of
vaulted arches. These vaulted arches now give the posi-
tion of the stair.
Trowel the concrete for the vaults onto the basket
forms, and fill the walls to make them solid.
Usi ng perhaps a l i ghtwei ght fill, or ul tral i ghtwei ght con-
crete, trowel a one-inch vault onto the basket forms, after
they have been covered wi th a simple cloth and sti ffened;
and do the same to fill the wal l s and columns in one
contiguous mass, so that the building becomes three di-
mensionally rigid.
Now start the second story} by the same procedure as
the first.
Place col umns in between the col umns whi ch are already
there, and place the bottom of these columns-on the vault,
where they will be filled by the fill whi ch forms the
floor.
Fill the floor, to make it horizontal.
Make up the form boards on the outside of the building,
to contain a horizontal floor, poured in above the vaults:
and fill the space wi th voi dsj ugs, bottles, anythi ng
468
T HE PROCESS OF CONSTRUCTI ON
whi ch will f orm roughly spherical bubbles in the concrete,
and reduce its mass without reducing strength.
Compl ete the second story, just as you have built the
first, then do the same for the third if there is one.
Make the terraces and seats and balconies around the
building.
Tr e a t them as part of the building, and yet part of the
earth.
Use terracotta tiles set simply in the earth, with small
plants gr owi ng in between them. Set the tiles in mud, keep
it as wet as possible, so that it grips the tiles quite firmly,
and yet makes the possibility of movement over time, and
al l ows plants to grow between them.
Pl ant a f ew small f l oweri ng plants, between the tiles,
while you are setting them in mud, so that after a f ew
months, the flowers are yel l ow, purple . . .
Build individual doors and windows, as cheafly as you
can, but each one shafed, and subdivided right accord-
ing to the frame.
Because of the procedure you have fol l owed in laying out
the wi ndow frames and door frames, all the openings in
the building are of slightly di fferent sizes now.
Thi s is essential, and it means that you cannot use stan-
dard doors or wi ndows.
Now nail together simple doors and wi ndows. Th e y
can be made of simple planks just nailed and gl ued to-
469
T HE WAY
gether, with small strips to f orm the subdivision of the
wi ndow panes, routed out to make a rebate for the wi n-
dow glass.
Carve decorations in the panels round the doors, and
in the other -places where you want some emphasis or
gaiety.
Yo u can carve simple scrolls and lines and hearts and dots
into the boards whi ch f orm the outside of the walls. Lat er,
when the building is almost finished, you can fill these
scrolls with plaster.
Paint the walls white; leave the columns visible.
Plaster in between the splines of basketwork whi ch formed
the basis for the vaults, and are still visible below the vault
in every room.
Plaster in the ornaments whi ch you have carved.
Plaster into the holes in those panels where you have cut
holes for ornaments.
Oi l the wood; wax the floor.
Then, finally, the finished building will have a rhythm
of the same patterns repeated hundreds and thousands
of times, but different every time that they occur.
The r e are not only roughl y equal columns, col umn spac-
ings, arcades, wi ndows, doorways, dormers, roofs, and ter-
races. Thes e are the larger patterns which repeat. But
there is also a weal th of mouldings, tiles, drips, gutters,
472
T HE PROCESS OF CONSTRUCTI ON
panels, bri ckwork, edges, door sills, ornaments, small strips,
small squares, small cornerstones, col umn heads, column
feet, rings cut in columns, braces, bracing details, nail-
heads, handles, spacers, sparsely placed, just where they
have to be, but visible, so that the building is completed by
these smallest structures, and formed by the rhythm of
their almost regul ar irregularities.
The -picture on the last two pages shows an example
of a building built like this.
Thi s picture is a photograph of a model . T h e building is
a four-story apartment building, for twenty-seven families,
in which the members of each family have designed their
own apartments, using a common pattern l anguage, and
the building is then intended to be built, floor by floor,
using the system of columns and beams and vaults which
I have just described.
Al t hough the model is very rough, it is already possible
to see the way that the discipline of the construction process
interacts with the informal ity of plan, to produce a build-
ing which goes much further towards the quality without
a name, than either of the actual buildings shown in chap-
ters 21 and 22.
A building built like this will always be a little looser
and a little more -fluid than a machine-made build-
ing.
Its doors and columns, wi ndows, shelves, wal l panels, ceil-
ings, terraces, and balustrades are shaped exactly to take
473
T HE WAY
their part in the l arger whol e: they fit it perfectly. And
because they fit it perfectly, they are therefore a little
rougher in appearance than the slick machine-smooth
quality of buildings made of factory materials.
But the beauty of the building lies in the fact that it
is whole.
T h e essential thing is this. Each process (gi ven by a pat-
tern) takes the configuration whi ch has been produced by
the previous processes, and adapts itself to them. No matter
where the col umns are, the process of weavi ng a vault can
form the vault according to the position of the columns.
No matter where the edges of a wi ndow are, the process
of maki ng a wi ndow forms the wi ndow and its panes
according to the size and shape of the wi ndow f rame.
And it is this whi ch makes the building whol e.
The building, like the countless buildings of tradi-
tional society, has the simplicity of a rough pencil
drawing. Done in a few minutes, the drawing cap-
tures the wholethe essence and the feeling of a
horse in motion, a woman bendingbecause its parts
are free within the rhythm of the whole.
And just so with the building now. It has a certain
roughness. But it is full of feeling, and it forms a
whole.
474
CHAPTER 24
THE PROCESS OF REPAIR
Next, several acts of building, each one
done to repair and magnify the product
of the previous acts, will slowly generate
a larger and more complex whole than
any single act can generate.
We know now, how a single act of building works.
We know that any person can lay out a building for
himself; that any group of people can do the same;
and we know how the builders can then carry out a
process of constructiony which will make a unified or-
ganic wholey out of the stakes marked on the ground.
Now we shall see how several acts of buildings in
a roWy will generate an even m<ore coherent and more
complex wholey piecemealby making sure that every
act contributes to the order of the previous acts.
In theory, according to chapter every act of building
is, with respect to its l arger context, an act of repair: a part
of the much l arger process in which several acts together
generate the l arger wholes from which a building complex
or a town is made.
But so far we have not had an opportunity to see this
cl earl ybecause in chapters 1 9- 23, we have been paying
attention to the individual creative act as an act whi ch
makes something new.
Now we shall change the focus, and pay attention to
each act, as an act of repair, within the l arger whol e.
No building is ever perfect.
Each building, when it is first built, is an attempt to make
a self-maintaining whole confi gurati on.
But our predictions are invariably wrong. People use
buildings differently from the way they thought they
woul d. And the l arger the pieces become, the more serious
this is.
479
T HE WAY
T h e process of design, in the mi nd' s eye, or on the site,
is an attempt to simulate in advance, the feeling and
events which will emerge in the real building, and to
create a configuration which is in repose with respect to
these events.
But the prediction is all guesswork; the real events
which happen there are al ways at least slightly di f f erent;
and the l arger the building is, the more likely the guesses
are to be inaccurate.
It is therefore necessary to keep changing the build-
ings, according to the real events which actually hap-
f>en there.
And the l arger the compl ex of buildings, neighborhood, or
town, the more essential it is for it to be built up gradual l y,
from thousands of acts, self-correcting acts, each one im-
proving and repairing the acts of the others.
Suppose, for instance, that some corner of your house
is not as alive as you would like.
For example, suppose I look at the house, and realize that
its garden is not worki ng properly as a HALF-HIDDEN
GARDEN, because although it is to one side of the house,
there is not enough protection between the garden and
the street. It needs some kind of wal l .
And, suppose that I go further, and bearing in mind
that work needs to be done to mend the barrier between
the garden and the street, I examine the garden f rom the
point of view of PRIVATE TERRACE ON THE STREET, and
480
T HE PROCESS OF REPAI R
find it missing. Suppose that I decide I need to build a
small brick terrace, at one side of my house, where it
touches the garden, in order to repair this gap.
Now, if I have already decided that some kind of wal l
needs to be built to protect the garden where it is too
open, it is only natural that I shall try to make the miss-
ing terrace in some way to link it to this missing wal l .
I n short, when I get a chance to start mendi ng the
garden, I can mend both these defective patterns with the
same act of building. And the repairs I make are not just
"repai rs, " but new designs, complex themselves, between
the crevices of the first design.
Or suppose that you have built a small laboratory
building.
It has a kitchen, a library, four labs, and a main entrance.
Yo u want to add a fifth laboratory to it, because you need
more space.
Don' t look for the best place right away. First, look at
the existing building, and see what is wr ong with it. The r e
is a path where tin cans collect; a tree whi ch is a beautiful
tree, but somehow no one uses it; one of the four labs is
al ways empty, there is nothing obviously wr ong with it,
but somehow no one goes there; the main entrance has
no places to sit comf ort abl y; the earth around one corner
of the building is being eroded.
Now, look at all these things whi ch are wr ong, and
build the fifth lab in such a way that it takes care of all
these problems, and also does, for itself, what it has to do.
481
T HE WAY
Can you see how rich and various the -parts of the
building will be, when they are built like thisP
Th e fi fth lab will be unique, unlike any of the other labs.
But not because you try hard to make it subtle, or beauti-
ful, or arty. It comes about in the most obvious way, just
because you are trying to be practical. Can you imagine
how hard it is to make one little laboratory, about 20 feet
by 20 feet, repair all these di fferent problems at once?
It isn't impossible; but to do it you will have to stretch
it here, extend it there, give it a special wi ndow here to
make the tree more useful, bring the path around it here,
to make the path with the tin cans less deserted, give it a
door at just this spot, to help create a pleasant corner in
the entrance, where people can wait.
So the richness and uniqueness of this little addition
comes about in the simplest and most practical way pos-
sible. It happens almost by itself, just because you pay
attention to the defects in the present building, and try
hard to repair them.
Each act of building, which differentiates a fart of
space, needs to be followed soon by further acts of
building, which further differentiate the space to make
it still more whole.
Thi s is commonplace in nature: and indeed, it is just this
whi ch al ways manages to make the parts of nature whol e.
Consider the leaves on a tree. A t first sight it seems as
though the leaves are solid, and the air between the leaves
482
T HE PROCESS OF REPAI R
is merely space. But the air between the leaves is as much
a part of nature as the leaves themselves: it takes on shape
as strongly as the leaves themselves; and like the leaves, it
is given its shape by the influences whi ch work on it.
Each leaf has a shape which is determined by the need
for strength, the growt h of the material, and the flowing
of the sap within the leaf. But the air between t wo leaves is
given its shape as definitely. If the leaves are too close to-
gether, the air between the leaves cannot act as a channel
for the sunlight whi ch the leaves need; and there may not
be enough breeze there to ventilate the leaves; if the leaves
are too far apart, the distribution of the leaves on the
twigs and branches is inefficient, and the tree will not get
enough sunlight to support it. Ever y part you look at is
not only whol e itself, but is part of a larger whol e, has
whol es around it, and is itself made up entirely of wholes.
Thi s is essential to the way that nature works: ajid all
of it is generated by the processes of successive differentia-
tions, each one helping to fill gaps, and mend gaps in the
whol e.
When things are first built, the gaps between the parts
are often left unwhole.
In the kind of worl d we have today, almost half the places
in a building or a t own are places "i n bet ween" the places
where you are meant to be.
T h e dark narrow space between t wo houses, the corner
of the kitchen which no one can reach, the area between
the train tracks and the next door i ndustrythese are
483
T HE WAY
obvious examples of places whi ch are literally forgotten
and left over.
And there are more remarkable examples, where spaces
are actually intended to be l eft over.
Thi nk about streets with cars parked on them, park-
ing lots, l ong corridors, wai ti ng rooms, the path between
the front door and the street, the garage, the closet under-
neath the stairs, the bathroom, the wi ndowl ess front hall
of the house. Al l these places are made with the mistaken
notion that you are only there in limbo, in between the
moments of l i vi ngas if they were way-stations between
the f ew places where you are actually meant to be alive.
But these gaps must be healed and made as whole as
the parts on either side of them.
I n a t own or building which is whol e, there are no places
like this; and in a life which is truly lived, there are no
moments like this either.
In a life whi ch is truly lived, there are no moments
which are "i n bet ween" or "out of l i f e" ever y moment
is lived fully. T h e Zen master says " whe n I eat, I eat;
when I drink, I dri nk; when I wal k, I wal k. " A build-
i ng or a t own whi ch is alive, has the same quality.
In a building or a town which is alive, and whi ch sup-
ports a life that is fully lived, there are no places whi ch
are way-stations between moments of l i vi ng; every place
is made in such a way that life can be savored fully there.
Ever y square inch of it has some valuable purpose, and is
able to support some moment in a person's life whi ch is
484
T HE PROCESS OF REPAI R
truly l i ved; and for this reason every part of it is whol e,
and every place between t wo whol es is also whol e.
Sl owl y, as the "process of repai r" repairs the gaps be-
tween the wholes, the structure becomes complete and
whole at every level.
This goes vastly beyond the normal conception of
repair.
In the commonpl ace use of the word repair, we assume
that when we repair something, we are essentially trying
to get it back to its original state. Thi s kind of repair is
patching, conservative, static.
But in this new use of the word repair, we assume^ in-
stead, that every entity is changi ng constantly: and that at
every moment we use the defects of the present state as
the starting point for the definition of the new state.
When we repair something in this new sensey we as-
sume that we are going to transform it, that new
wholes will be borny thaty indeedy the entire whole
which is being repaired will become a different whole
as the result of the repair.
In this sense, the idea of repair is creative, dynami c, open.
It assumes that we are constantly led to the creation of
new wholes, by paying attention to the defects in the exist-
ing wholes, and trying to repair them. It is still true that
every act helps to repair some l arger, older whol e: but the
repair not only patches i t i t also modifies it, transforms
485
T HE WAY
it, sets it on the road to becoming something else, entirely
new.
In this framework, we gain an entirely new view of
the process through which a sequence of acts of build-
ing generates a whole.
Broadl y, what is happening is that there is, at each stage
in the life of any part of the envi ronment, a wholeness
whi ch is specific to that moment in its life: and that each
new act of building, provided that it is done with an eye
to maki ng the overall whol e still more whol e, more alive,
will transform that whol e, and gradually give birth to
new wholes.
In this sense, then, the idea of repair explains both how
we can mend past defects in things, and also, at the same
time, how it is possible to make and remake the worl d
so that the cooperation of a number of acts of building, in
sequence, also creates whol es which are complete, and live,
at every moment of their hi st oryyet al ways gi vi ng way,
under the process of repair, to even newer wholes, which
once again remake themselves at the next stage of the
repair.
In order to see this clearly, let us imagine that there
is somewhere a building complex, growing, over time.
Step by step, each act of building, which contributes to the
growi ng whol e, is also helping to repair or heal what is
already there.
486
T HE PROCESS OF REPAI R
Specifically, let us imagine a cluster of houses growi ng
over time.
Each house starts with a small beginningno more
than a family kitchen, with a bed alcove of one endy
and a kitchen counter.
Al l in all, no more than 300-500 square feet to start wi th.
Theny for the first few yearsy people add 100-200
square feet more each year.
First a bedroom perhaps; another bedroom; a workshop;
a garden terrace; a full bathing room; arcades and
porches; studio; a bigger sitting room, wi th a big fire-
place; a garden shed.
And, at the same time common things are also built;
they plant an avenue of trees; a small gazebo; a shared
outdoor room; paving on the paths; closed-in garages; a
communal workshop; a small fountain or a swi mmi ng
hole. . . .
As the buildings reach maturityy the increments get
smaller.
A bench; a sitting wal l ; a roof over the entrance; a rail-
i ng on the terrace upstairs; a stair to the roof ; a fishpond;
a bay wi ndow; an extra gat e; a vegetable patch; a wal l of
shelves; a garden seat around the trunk of a growi ng
tree. . . .
487
T HE WAY
Yet, at the same time, collectively, the houses begin
to generate the larger -patterns which define the
cluster.
Each person begins to work with his neighbor, first the
neighbor on one side, then the neighbor on the other side;
and together they try to make the space between their
houses beautiful. Of course, they start by eliminating obvi-
ous conflicts between wi ndows, or cases where one house
takes the sun from the next man' s garden; but they also
work out details accordi ng to the detailed patterns. For
example, one group of people decide to make their en-
trances open off a small public garden, with a tree, and
wi th an outdoor seat in it, facing the sun, and from whi ch
people can watch the distant outdoor room, accordi ng to
the patterns TREE PLACES, OUTDOOR SEATS, and PUBLIC
OUTDOOR ROOM.
For exampl e, path shape says that a path must be a
place, almost like a room, partly enclosed, wi th a middle,
so that people feel comfortabl e staying there, not just
wal ki ng through. Each group uses this pattern to make
the paths outside their houses better shaped, with seats in
them; and the pattern even affects the exact shape of the
boundary wal l where the houses meet the paths.
And PRIVATE TERRACE on THE STREET says that each
house should have a private terrace, near its living room,
but placed so that one can see from it, into the public
space of the street beyond, and wave or say hello to some-
one out there, wi thout the privacy of the house being dis-
488
T H E PROCES S OF R E P A I R
The slow growth of twelve houses
489
T HE WAY
turbed. Each of the houses has such a terrace, and the
way these terraces make the paths alive is also discussed
and modified by the groups at each end of the cluster.
Slowly, at every level, the arrangement of wholes
becomes so dense that there are no gaps between the
wholes: every part, and every part between two parts,
is whole.
I n a house, the garden round the house is positive; the
boundaries between house and garden are again positive.
T h e thickness of the walls is positive; the wal l whi ch
forms the enclosure between terrace and garden is again
positive; a seat; while the interior walls are places too
(shelves, niches, etc. ) ; within the house, each room is, of
course, a pl ace; but to make this happen, each one has
wi ndows on two si deswi t h the result that the rooms take
on an odd configuration, in pl anand every place be-
tween the rooms is once again a place. At the level of con-
struction, we find the same. Ever y room has its corners
marked by columns. Each column is again a visible co-
herent t hi ng; the col umns which stand free have places
around them once agai n; each col umn is itself made in
such a way that the places where it joins other entities, are
once again entities: columns and feet.
So, the houses get their form, both as a group, and
separately, as individuals, from the gradual accretion
of a number of small separate acts.
490
T HE PROCESS OF REPAI R
T h e bench is placed, to form a terrace; the extra bed-
room helps to shield the garden f rom the nei ghborhood;
the paving on the path is placed in such a way that it de-
fines the entrance transition, and forms a contour edge to
help prevent erosion from the rain; the extra gr owi ng
house helps to form common land outside the house; the
avenue of trees helps to form a park in the common l and;
the garages not only shield the cars, but also help to form
gateways into the cluster. . . .
Ever y small act helps not only to increase the space, but
also makes a contribution to the larger patterns whi ch are
needed there.
And finally, the common character of the cluster of
houses grows, without control, simply from the accumul a-
tion of the individual actsbecause each act of building is
conceived as something whi ch not only does good for it-
self, but also has the obligation to help generate the whol e.
In chapter ig I argued thai an organic whole could
only be created by a differentia ting process.
I explained that only a process of differentiation, because
it defines the parts within the whol e, can generate a natu-
ral thi ng; because only this kind of process can shape parts
individually, according to their position in the whol e.
We see now that there is a second, complementary
process which produces the same results, but works
piecemeal, instead.
491
t h e w a y
Wh e n a place grows, and things are added to it, gradu-
ally, being shaped as they get added, to help f orm l arger
patterns, the place also remains whol e at every s t agebut
in this case the geometri c vol ume of the whol e keeps chang-
ing, because there is an actual concrete aggregat i on of
matter taki ng place.
This process, like the simple differentiating process, is
able to make wholes in which the parts are shaped ac-
cording to their place.
But this process is still more powerful: because it
can make groups of buildings which are larger and
more complex.
And it is more powerful, above all, because it leaves
no mistakes: because the gaps get filled, the small
things that are wrong are gradually corrected, and
finally, the whole is so smooth and relaxed, that it
will seem as though it had been there forever. It has
no roughness about it, it simply lies there stretched
out in time.
492
CHAPTER 25
THE SLOW EMERGENCE OF A TOWN
Finally, within the framework of a com-
mon language, millions of individual
acts of building will together generate a
town which is alive, and whole, and un-
predictable, without controlthis is the
slow emergence of the quality without a
name, as if from nothing.
Finally, then, we come to the town itself.
We have seen how a few dozen acts of building,
done within a common -pattern language can gradually
generate a whole-, and that the larger -patterns which
are needed to d.efme that whole, can be created piece-
meal, by the slow concrescence of the individual acts.
Now we shall see how this same process can be ex-
tended to a town.
For we now face at last, the deepest and the most f ar-
reachi ng postulate of a l l na me l y : the postulate that it is
possible for all the large-scale order of a t own to be cre-
ated purely by means of i ncremental piecemeal acts.
The first thing to recognize is that for any system as
vast as a town there is a fundamental problem.
Wh e n a single human mi nd or a group of mi nds together
conceive a building, they natural l y conceive it as a whol e,
and its parts then fall into place, both to support that
whol e, and also to be whol e themselves.
But when a t own gr ows, it does not gr ow in one hu-
man mi nd, nor in any coherent group of minds. A t own
is made f rom millions upon millions of individual acts of
building. Ho w can we be.sure that the t own wi l l be whol e,
and not a rambl i ng, i ncoherent chaos, if it is built f rom
millions upon millions of individual acts?
The question is: can the structure emerge, simply
from the spontaneous interaction of the parts?
496
t h e s l o w e m e r g e n c e o f a t o w n
Ca n it be created by a free process, in whi ch the people
locally do what they want , and still create the whol e
successfully?
Or must it be planned, by a hidden hand, according to
a blueprint or a master plan?
Mus t there be some kind of control , some kind of totali-
tarian order, imposed f rom above, whi ch restricts the f ree-
dom of the individual acts, and forces them into a l arge-
scale order?
To put this question in perspective I should like to
compare it to a question that arose in the early years
of biology: "How does an organism get formed?"
Consi der your hand, for exampl e. Hol d it out in front of
you. Do you realize that this compl ex shape, this intricate
structure of bones, and muscl e, f i ngers, thumb, f i nger-
nails, joints, wri nkl es, subtle curves, has come into being
entirely wi thout the background of a blueprint or a master
plan?
Do you real ize, f ul l y, that the cooperation of the cells
has f ormed this hand, guided only by certain rules whi ch
direct the gr owt h of the individual cells, accordi ng to their
interaction wi th each other?
O r consider another exampl e. As I look out of my wi n-
dow, I see a f ew square yards of flowering bushes there,
wi th grass bel ow them, a tree or t wo sticking out, and a
f e w other plants, sitting among them.
497
t h e w a y
If I look at these bushes, down to the details of their
individual leaves, the grains of soil bel ow them, the twi gs,
the petals of the flowers, the insects sitting on the leaves,
the gaps between the leaves, where l ower leaves are open
to the sky, do I have to believe that there is a hidden de-
signer who has created this?
At first biologists thought that there must be a hidden
designer.
T h e y believed that this miracle could not be happening
wi thout somethi ng to guide it, a spiritual master plan,
whi ch told the cells just where to place themselves. Up
until the seventeenth century, some biologists even be-
lieved that every cell in a man contained a little man, who
was a model of the l arger man.
Yet now it has become clear that the organism is
formed purely by the interaction of its cells, guided
by the genetic code.
Recent l y, our experi ments have begun to make it clear
that this seemi ng miracl e is not a miracl e of gui dance f rom
above, but instead a miracle of subtle organi zed coopera-
tion bet ween the parts: that the gr owi ng cells al one, com-
muni cati ng wi th each other, and guided only by the in-
structions pr ogr ammed into them by the genetic code, act
correctl y, wi t h respect to one another, in such a way that
they create an entirely individual whol e, not predictable in
detail, but recogni zabl e in species.
498
t h e s l o w e m e r g e n c e o f a t o w n
And this is true for a town too.
A t one time people believed that a t own had to be planned
by a pl anner who made a plan or blueprint. It was said
that if the order of the t own is not created f rom above,
there wi l l just not be an order in the t own. And so, even
in spite of the most obvious evi dence of all the beautiful
t owns and villages built in traditional societies wi thout
master plans, this belief has taken hol d, and people have
al l owed themselves to give up their f reedom.
As in biology, though, it is becomi ng clear now that the
structure of a t own can be woven much more deeply,
more intricately, f rom the interaction of its individual acts
of building wi thi n a common l anguage, than it can f rom
a blueprint or a master pl anand that indeed, just like
your hand, or like the bush outside my wi ndow, it is best
generated by the interaction of the rules whi ch govern
the construction of the parts.
Let us see in detail, how a process of interacting rules
can work to generate a town.
T h e essential fact, whi ch makes it possible, is that the pat-
terns are not generated, suddenl y, compl etel y, but that in-
stead each l arger pattern comes into being as the end
product of a l ong sequence of tiny ac t s and that these
tiny acts themselves have the power to create the pattern,
if they are repeated often enough.
This is commonplace, in the growth of an organism,
499
t h e w a y
where all the larger patterns are generated, merely as
the end products of tiny, daily transformations.
At any gi ven moment , in a gr owi ng organi sm, there is no
sense of the " e nd" or of the final " g o a l " of gr owt h.
The r e is, instead, a process of transf ormati on, whi ch is
able to take the present state of the organi sm, and move
it slightly, in the next minute of g r o wt h i n such a way
that when the same process is then repeated in the minute
after that, and in the minute after that, sl owl y, i nexorabl y,
the necessary patterns come into be i ngnot accordi ng to
some pl an, but as the product of a sequence of t ransf orm-
i ng steps.
In detail, this happens through the action of certain
chemi cal fields, created by the hormones. The s e fields en-
courage and inhibit gr owt h, at di f f erent parts of s pa c e
and this di fferenti al gr owt h process then sl owl y generates
the gr owi ng whol e. Accor di ng to the state of these fields
at any one moment , the gr owt h process creates a certain
minute gr owt h, whi ch transforms the existing structure,
just slightly, accordi ng to a certain rule.
As the gr owt h occurs, the chemi cal fields change, so
that the " s ame" transf ormati on, gui ded by the " s ame"
rule, has a slightly di f f erent ef f ect each time that it occurs.
Thus , the repeated application of the transf ormati on,
gui ded by the changi ng concentrati ons in the chemi cal
fields whi ch tell the organi sm how close it is to reachi ng
equilibrium, guide it t owards the finished pattern. But the
finished pattern is merel y the end product of the succession
of tiny transformati ons.
500
t h e s l o w e m e r g e n c e o f a t o w n
And just this too, must also happen in a town.
I n this case, the " chemi cal f i el ds" are repl aced merel y by
people' s consciousness of the l arger scale patterns, whi ch
provi de the rules of gr owt h. I f people have agreement s
about these l arger scale patterns, then they can use their
knowl edge of the patterns, and the degree to whi ch these
patterns have been attai ned, or not, to gui de the gr owt h
and the assembl y of the smal l er patterns.
Sl owl y, under the i mpact of this gui dance, the sequence
of smal l -scal e t ransf ormat i ons wi l l , of its o wn accord,
create the l ar ger patterns, piece by pi ece: wi t hout any indi-
vi dual person necessarily knowi ng j ust exactl y how or
wher e these l arger patterns wi l l be in the finished t own.
Here, for example, is the way a process of this kind
can generate a very large-scale pattern like CI TY COUN-
TRY FINGERS.
A t any gi ven moment , the actual boundary bet ween the
t own and the count ry is a r ough uneven curve. Suppose
that the t own gi ves local communi t i es i ncenti ves whi ch
encourage g r owt h j ust wher e these curves bul ge out-
war ds ; whi ch inhibit g r owt h on the outside of the places
wher e the curve bul ges i nwar ds ; and whi ch even en-
courage destructi on of buildings, and the re-creati on of
open space, on the inside of the curve at those points wher e
the curve bul ges i nwards.
Unde r the i mpact of these i ncenti ves, the bul ges wi l l
g r o w out war ds gr adual l y, to f orm city f i nger s; the " no n-
501
t h e w a y
bul ges" wi l l stay wher e they are, or even g r o w backwar ds
into the t own, mai nt ai ni ng and creat i ng count ry fingers.
O f course, in any one mont h, the actual progress is
mi nut e. But that does not mat t er. Unde r the i mpact of
this g r owt h process, sl owl y but i nexorabl y the pattern
CITY COUNTRY FINGERS wi l l come into bei ng.
On a slightly smaller scale
}
the same can happen in a
community to generate a PROMENADE.
Suppose, f or i nstance, that there is the begi nni ng of a
PEDESTRIAN STREET, and PROMENADE, bet ween some
corner wher e there is an ice cr eam store, and another
corner wher e people gat her in the eveni ngs.
T h e communi t y now makes it cl ear to all the i ndi vi dual
nei ghborhoods around the line bet ween these points, that
they want to encourage the r emoval of car traffi c f r om
the paths al ong this line, that they hope to see ne w c om-
muni t y acti vi ty l ocate al ong the emer gi ng promenade, and
so on.
Ea c h nei ghborhood, then, f ormul at es the wa y in whi ch
it can best help the sl ow creati on of this PROMENADE
502
t h e s l o w e m e r g e n c e o f a t o w n
and does so because it wi l l gain f r om the i ncenti ves whi ch
the l arger communi t y has in its power . Suppose, for i n-
stance, that one nei ghborhood sees that it can create a
GATEWAY, wi t h a little SPORTS ACTIVITY, in a place wher e
the PROMENADE mi ght run. I n order to create the pr om-
enade, they wi l l then build a path, whi ch mi ght run in
that part of the c ommuni t y, past the gat eway whi ch l eads
into the nei ghborhood, and past the table tennis and l ocal
sports. Gr adual l y the promenade emer ges f r om the piece-
meal ef f ort s of the di f f erent nei ghborhoods.
patterns in a local neighborhood.
Consi der t wo patterns whi ch are the responsibility of the
l ocal nei ghborhood: MAIN GATEWAYS and NETWORK OF
PATHS AND CARS.
T h e nei ghborhood creates i ncenti ves, whi ch encourage
the house clusters, and wor k communi t i es, and indi-
vi dual house owner s, to generat e these patterns sl owl y,
piece by piece.
503
t h e w a y
On e year the people in one cl uster take down a back
f ence, and so a path begi ns f r om one street, in bet ween
t wo clusters, and t owar ds anot her street. T h e next year,
a di f f erent cluster connects its c ommon area wi t h that
same path, act i ng agai n under the i ncenti ve whi ch the
local nei ghborhood provi des, knowi ng that it is in their
interest to help to generat e this l ar ger pattern whi ch the
nei ghborhood needs.
Anot her year, the people who live near the boundary of
the nei ghborhood decide to bri dge the street wi t h a pair of
small bui l di ngs whi ch help to f or m a gat e. O f course, they
build this gate in such a wa y that it connects wi t h the
emer gi ng NETWORK OF PATHS AND CARS. T h i s act does
not, in itself, f or m a compl ete MAIN GATEWAY. But the
nar r owi ng of the street, wher e the t wo bui l di ngs have
been built, is cl earl y the precursor of a gat e: other smal l er
acts wi l l f ol l ow it, to make the gat e way compl et e. But
meanwhi l e, the nei ghborhood has al l owed these t wo buil d-
i ngs to encroach on the street, because they recogni ze that
the nar r owi ng whi ch this creates wi l l gradual l y move that
corner of the nei ghborhood t owar ds a state wher e it does
504
t h e s l o w e m e r g e n c e o f a t o w n
have the MAIN GATEWAY whi ch the whol e nei ghborhood
needs in that quarter.
Each of these processes requires a large group, and a
group of smaller groups.
I n this respect they are j ust like the exampl e of the houses
and the cluster of houses, in chapter 24, but ext ended to a
l arger scale. Th e r e the houses act i ndi vi dual l y, to generat e
the l arger patterns whi ch the cluster needs. An d j ust so
here, but at a l arger scale. T h e clusters act t oget her, to
generat e the patterns whi ch the nei ghborhood needs. T h e
nei ghborhoods act t oget her to create the patterns, whi ch
the communi t y needs. An d the communi t i es act t oget her
to generat e the patterns whi ch the whol e t own needs.
In order for these processes to cover the whole struc-
ture of a town, it is therefore necessary that the town
be made up from a hierarchy of groups and land,
each one responsible for its own patterns.
A t the l owest l evel , each i ndi vi dual person owns his own
private space: and is responsible for hel pi ng to create the
patterns there, accordi ng to his needs.
A t the second l evel , the f ami l y has its o wn l and, and
its own c ommon space: and the wor kgr oup has the same.
T h e families, and wor kgr oups are responsible f or all the
l arger patterns whi ch are needed f or the c ommon space.
A t the third l evel , each cluster of families or wor k-
groups is a wel l def i ned l egal e nt i t y a l egal l y def i ned
505
t h e w a y
g r oupwhi c h owns its own l and (that l and, whi ch the
families all use together, but whi ch is not private to any
one of t h e m) a n d is responsible for all the patterns
whi ch are needed there.
At the fourth l evel , the nei ghborhood, made up of clus-
ters, is again a wel l - def i ned, l egal l y constituted human
group, whi ch once agai n owns that land whi ch is com-
mon to its member s l ocal roads, local parks, local ki nder-
gar t ens but does not own the smal l er common l ands the
clusters own. T h e nei ghborhood, as a group, is responsible
for the patterns in its common l and.
A t the next level agai n, there are communi ti es, made
up of nei ghborhoodsagai n wel l def i ned, and l egal l y con-
st i t ut edagai n wi th their own common l and, i ncl udi ng
l arger roads, l arge public buildings, again responsible for
all those patterns whi ch are needed to serve the whol e
communi t y.
Fi nal l y, at the l evel of the t own, there is agai n a l egal
entity, again owni ng common l a ndno t owni ng, as a
t own does now, all the streets, all the parks, but owni ng
onl y those whi ch are specifically used by ever yonet he
very l argest ones, and responsible for just those l argest pat-
terns whi ch are needed in this l argest common l and.
And, in order for the larger patterns to come into be-
ing, piecemeal, from the aggregation of the smaller
acts, it is then necessary that each group is made re-
sponsible for helping the next larger group, create the
larger patterns which the larger group requires.
506
t h e s l o w e m e r g e n c e o f a t o w n
Thu s , whe n a person f orms his room he is gi ven specific
i ncenti ves to hel p to f orm the l arger patterns of the house-
hol d, or workshop in whi ch his room is pl aced: so COM-
MON AREAS AT THE HEART, A ROOM OF ONE'S OWN,
INTIMACY GRADIENT, BUILDING EDGE, POSITIVE OUT-
DOOR SPACE, WINGS OF LIGHT, wi l l gr adual l y emer ge.
And whe n member s of a f ami l y build or modi f y their
house, they are gi ven specific i ncenti ves by the cl uster,
whi ch makes t hem responsible for i mpr ovi ng the envi ron-
ment around t he ma bo v e t hem, bel ow t hem, and to the
side: so BUILDING COMPLEX, CIRCULATION REALMS,
HALF-HIDDEN GARDEN, SMALL PARKING LOTS, SHIELDED
PARKING, FAMILY OF ENTRANCES, wi l l gr adual l y emerge
under the responsibility of the cl uster.
Wh e n each cluster modi f i es its overal l f or m, or builds
on, it is responsible to the nei ghborhood for bri ngi ng
l arger nei ghborhood patterns into bei ng: NEIGHBORHOOD
BOUNDARY, MAIN GATEWAY, GREEN STREET, POOLS AND
STREAMS, CHILDREN' S HOME, LOOPED LOCAL ROAD, HOME
WORKSHOPS, SCATTERED WORK, QUIET BACKS. T h e
nei ghborhood can gi ve money, or other i ncenti ves, to en-
courage those smal l acts whi ch help to bri ng these l arger
patterns into bei ng.
T h e l arger communi t y can in the same wa y gi ve both
money and permission to those nei ghborhoods whi ch help
bri ng its even l arger patterns into bei ng: ACCESSIBLE
GREEN, PARALLEL ROADS, PROMENADE, SHOPPING
STREET, MOSAIC OF SUBCULTURES, SUBCULTURE BOUND-
ARY, SCATTERED WORK, SACRED SITES, HOLY GROUND,
HEALTH CENTERS. Ag a i n, these l ar ger patterns wi l l
507
t h e w a y
emer ge accordi ng to the vol unt ary cooperati on of the
nei ghborhoods, ECCENTRIC NUCLEUS, DENSITY RINGS . . .
Ev e n at the ver y l argest l evel s, the regi on and the city
can provi de incentives whi ch wi l l encourage these com-
muni ti es to modi f y their own i nternal structure in a way
whi ch helps the l argest patterns to appear: NETWORK OF
TRANSPORTATION, RING ROADS, CITY COUNTRY FINGERS,
LOCAL TRANSPORT AREAS, AGRICULTURAL VALLEYS,
ACCESS TO WATER, SACRED SITES . . .
Under these circumstances, it is certain that every fat-
tern will appear at the level where it is needed.
T h e smal l patterns produced di rectl y by the individual s,
and repeated over and agai n. T h e l arge patterns generat ed
i ndi rectl y, by the gr adual i ncrement al repetition of the
smal l er patterns.
But it is never certain just exactly where a given pat-
tern will appear.
Nor is it certai n j ust what f orm any one pattern wi l l take,
in any one parti cul ar pl ace.
W e k no w, ahead of time, wha t general f or m it has.
But we do not know its exact f or m, its exact di men-
sions, its detailed charact er, until it has g r o wn to mat uri t y
be c a us e it f orms itself, in the process of g r owt h, and
it is onl y gr owt h itself, in response to the details of its sur-
roundi ngs, whi ch can shape it correct l y.
I n this sense it is like the nat ural order of an oak tree.
508
t h e s l o w e m e r g e n c e o f a t o w n
The -final shape of any one particular oak tree is un-
predictable.
Wh e n the oak tree grows, there is no blueprint, no
master plan, whi ch tells the t wi gs and branches where
to go.
We know in general that it wi l l have the overall f orm
of an oak, because its gr owt h is gui ded by the pattern
l anguage for an oak tree (its genetic code) . But it is un-
predictable, in detail, because each small step is shaped by
the interaction of this l anguage wi t h external forces and
condi t i onsrai n, wi nd, sunl ight, the composition of the
earth, position of other trees and bushes, the thickness of
the leaves on its own branches.
And a town which is whole, like an oak tree, must be
unpredictable also.
T h e fine details cannot be known ahead of time. We may
know, f rom the pattern l anguage whi ch is shared, what
kind of t own it wi l l be. But it is impossible to predict its
detailed plan : and it is not possible to make it gr ow accord-
i ng to some plan. It must be unpredictable, so that the
individual acts of building can be free to fit themselves to
all the local forces whi ch they meet.
T h e people of a t own may know that there is goi ng to
be a main pedestrian street, because there is a pattern
whi ch tells them so. But , they cannot know just where
this main pedestrian street wi l l be, until it is al ready there.
T h e street wi l l be built up f rom smal l er acts, wher ever
509
t h e w a y
the opportunity arises. Wh e n it is finally made, its f orm is
partly gi ven by the history of happy accidents whi ch let
the people build it al ong wi th their own more private acts.
The r e is no way of knowi ng, ahead of time, just where
these accidents wi l l fall.
This process, exactly like the emergence of any other
form of life, alone produces a living order.
I t is a process by whi ch the small acts of individuals, al-
most random, are sieved and harnessed, so that what they
create is orderl y, even though the product of confusi on.
It creates order, not by f orci ng it, nor by imposing it
upon the worl d ( t hrough plans or drawi ngs or compo-
nents) : but because it is a process whi ch draws order f rom
its sur r oundi ngsi t al l ows it to come together.
But of course, by this means far more order can come
into being, than could possibly come into being through
an i nvented act.
It is vastly more complex than any other kind of
order. It cannot be created by decision. It cannot be
designed. It cannot be predicted in a plan. It is the
living testament of hundreds and thousands of people,
making their own lives and all their inner forces mani-
fest.
And, finally, the whole emerges.
5 1 0
CHAPTER 26
ITS AGELESS CHARACTER
And as the whole emerges, we shall see
it take that ageless character which gives
the timeless way its name. This character
is a specific, morphological character,
sharp and precise, which must come into
being any time a building or a town
becomes alive: it is the physical embodi-
ment, in buildings, of the quality with-
out a name.
If you follow the way of building which I have de-
scribed in the last twenty-five cha-pters, you will find
that the buildings which emerge will, gradually, and
of their own accord, take on a certain character.
It is a timeless character.
Look at the drawi ngs of buildings whi ch people have
made wi th pattern l anguages on this page and the next.
T h e y could be Roman, Persian, f rom Mohe nj o Dar o,
f rom medi eval Russia, Icel and, Af r i ca. T h e y mi ght be
fi ve hundred years old, or fi ve thousand, or built five
thousand years f rom now.
Wi l l y nilly, wi thout real i zi ng what they are doi ng,
wi thout knowi ng its si gni fi cance, people make buildings
whi ch are far more like the countless t owns and build-
ings of vanished cultures, and ages past, than any of the
buildings whi ch are being built today.
5 1 7
t h e w a y
In short, the use of languages does not just help to
root our buildings in reality; does not just guarantee
that they meet human needs; that they are congruent
with forces lying in themit makes a concrete differ-
ence to the way they look.
I n order to make this precise, let me distinguish t wo di f f er-
ent morphol ogi es.
J
5 1 8
i t s a g e l e s s c h a r a c t e r
I magi ne that we sort the bui l di ngs of the wor l d into
o o
t wo piles. I n the one pile, all those tradi ti onal bui l di ngs,
built f or t housands of years, in tradi ti onal societies all over
the wor l d. An d , in the ot her pile, all those bui l di ngs built
in the last hundr ed years, built by total i tari an t echnol ogy,
by i ndust ry.
Al t houg h the bui l di ngs and t owns in the fi rst pile have
vast vari ety of di f f er ent f o r ms br i c k houses, st r aw huts,
stone vaul ts, ti mber f r ami ng, t hat ched roofs, l og cabins,
piled dry stone wal l s, stone col umns, steep roof s, flat
roof s, arched wi ndows , strai ght wi ndows , bri ck, wood,
stone, whi t e, bl ue, br own, yel l ow, nar r ow streets, wi de
streets, open compounds, closed cour t yar ds s t i l l c om-
pared wi t h the ot her pile, they have somet hi ng in c om-
mon.
I t is a parti cul ar morphol ogi cal charact er. An d whe n
bui l di ngs are made in the f r a me wo r k of the timeless wa y ,
they al ways have this special charact er.
This character is marked, to start with, by the fatterns
underlying it.
L o w bui l di ngs; open stairs l eadi ng to upper floors; l ong
tables wi t h c ommunal eat i ng; roof s whi ch are either
pi tched or domed, l arge and visible, or else used as t er-
races; r ooms pl aced so that the l i ght floods into t hem f r om
t wo sides at l east; gar dens made f or smel l i ng and t ouch-
i ng flowers not onl y seei ng t he m; wa t e r still and mo v i ng ;
arcades al ong the edge of bui l di ngs; porches bet ween
bui l di ngs and gar dens ; smal l public and pri vate squares,
wi t h arcades and porches at their edges ; gal l eri es at upper
5 1 9
t h e w a y
l evel s; col umns at the corners of the rooms and spaces;
ceilings at di f f erent heights accordi ng to the i nti macy of
rooms; small alcoves at the edges of rooms; trellises
covered by roses and vines; wal l s in whi ch possessions and
decorations show the character of life and fill the room
with it; communi ti es separated f rom one another by sub-
stantial boundaries, so that each one can live in its own
way, unhampered by the others.
It is marked by greater differentiation.
If we compare these buildings wi th the buildings of our
present era, there is much more vari ety, and more detail:
there are more internal di fferences among the parts.
The r e are rooms of di f f erent sizes, doors of di f f erent
widths, col umns of di f f erent thickness accordi ng to their
place in the building, ornament s of di f f erent kinds in dif-
ferent places, gradients of wi ndow size f rom floor to floor.
The r e are small rooms opening off l arger rooms; there
are swel l i ngs at the places where paths meet ; there are en-
l argement s of the connection wher e a col umn meets a
beam; there are more hi ghl y di fferenti ated kinds of wood
pieces in a wi ndow to correspond to the divisions between
l arge panes and small.
Tho u g h most of the rooms are rect angul ar, there are a
f ew ci rcul ar, or elliptical, or odd-shaped pol ygons, mixed
in among the rectangl es. Edges between adj acent places
are al ways places in themsel ves; they have a thickness,
there is a crinkled character to t hem; it is almost never
just a plane between t wo voids. The r e are openings be-
520
i t s a g e l e s s c h a r a c t e r
t ween each space, and the ones next to it; and these open-
ings occupy a fairly l arge percentage of the wal l area.
Cur ved lines and surfaces are rare; but they occur occa-
sionally, at points of emphasis. Col umns are thick, and
often bunched or clustered. Paths are of ten slightly
crooked, streets of ten taper, wi th mi nor bends in them.
T h e character is marked, in short, by greater di fferences,
and greater di fferenti ati on.
But it is marked, above all, by a sfecial balance be-
tween "order" and "disorder."
The r e is a perfect balance bet ween straight lines and
crooked ones, between angl es that are square, and angl es
that are not quite square, between equal and unequal
spacing. Thi s does not happen because the buildings are i n-
accurate. It happens because they are more accurate.
T h e similarity of parts occurs because the forces whi ch
create the parts are al ways more or less the same. But the
slight roughness and unevenness among these similarities,
comes f rom the fact that forces are never exactly the
same.
T h e straight lines whi ch are roughl y straight, wi l l come
about because the boundary of a space must al ways have a
space whi ch is alive on both sides of it. A curved wal l
f orms a concavi ty of space on its outside, whi ch tends to
destroy space. But the straight wal l s are not perfectl y
straight, because there is no reason for them to be perfect.
T h e angl es whi ch are roughl y square wi l l come about
because f ew angl es in a room, or in the edges of an out-
521
t h e w a y
door area are comf ortabl e if they are acute. But they are
not perf ectl y square, because there is no reason for them
to be perf ect.
And it is marked
y
in feeling, by a sharpness and a
freedom and a sleepiness which happens everywhere
when men and women are free in their hearts.
A f ew cups and glasses standi ng on a rough table, a hand-
ful of flowers freshly picked f rom the garden on the table,
the notes of an old piano, children pl ayi ng in the corners.
It is not necessarily complicated. It is not necessarily
simple.
Ma ny students, when they first try to create this char-
acter, create a tortured intricacy. But this is almost the
opposite of the true character. I n order for a place to have
this character, it does not necessarily have hundreds of
small angl es, f unny corners, and so on. Someti mes, it is
perfectl y regul ar.
It comes simply from the fact that every part is whole
in its own right.
I magi ne a prefabricated wi ndow whi ch sits in a hole in a
wal l . It is a one, a uni t; but it can be lifted directly out
f rom the wal l . Thi s is both literally true, and true in feel -
ing. Li t eral l y, you can lift the wi ndow out wi t hout doi ng
damage to the fabric of the wal l . And, in your i magi na-
522
i t s a g e l e s s c h a r a c t e r
tion, the wi ndow can be removed wi thout disturbing the
fabric of what surrounds it.
Compar e this wi th another wi ndow. Imagi ne a pair of
col umns outside the wi ndow, f ormi ng a part of the wi n-
dow space. T h e y create an ambi guous space whi ch is part
of the outside, and yet also part of the wi ndow. Imagi ne
splayed reveals, whi ch help to f orm the wi ndow, and yet,
also, wi th the light reflected off them, shining in the
room, they are also part of the room. And imagine a
wi ndow seat whose back is part of the wi ndow si l l not
a seat l eani ng against the wi ndow sill, but a seat whose
back is indistinguishable f rom the wi ndow sill, because it is
continuous.
Thi s wi ndow cannot be lifted out. It is one wi th the
patterns whi ch surround i t; it is both distinct itself, and
also part of them. T h e boundaries between things are less
mar ked; they overl ap wi th other boundaries in such a way
that the continuity of the worl d, at this particular place,
is greater.
This character emerges whenever any fart of the
world is healed.
Each thi ng is made of parts, but the parts overlap and
i nterl ock to such an extent that the oneness of all things
becomes more marked. The r e are no gaps between the
parts, because each gap is just as much a part itself. And
there are no clear divisions between levels in the structure,
because, to some extent, each part reaches down, and is
conti nuous and i ntegral wi th smal l er units of structure,
523
t h e w a y
whi ch, once agai n, cannot be lifted out, because their
boundaries overl ap, and are continuous wi th l arger units.
It is therefore the most fundamental mark of health
and- life in our surroundings.
Unde r the gui dance of a process whi ch al l ows whol es to
f orm, at the scale of individual persons, families, gardens,
trees, forests, wal l s, kitchen counters, each part becomes
whol e in its own terms, because it is adapted to the l arger
wholes whi ch it is part of , and because it is adapted to the
smaller whol es whi ch are a part of it.
T h e n the worl d becomes onet her e are no r i f t s
because each part is part of l arger whol es, and smal l er,
there is a conti nuum of order, whi ch leaves the parts indis-
tinct, and uni fi ed.
Outwardly this character reminds us of the buildings
of the fast.
Y o u can see this in the historical plans at the begi nni ng
of the chapter. T h e y too, all have this inner rel axati on.
T h e y too all have the balance of order and di sorder; the
gentle rectangl es, distorted slightly, wher ever the building
or the land require it; they too all have the subtle balance
of small spaces and open spaces; the unity whi ch happens
when each part, inside or outside, is a part wi th its own
solid shape; they all have the slightly rambl i ng, i nnocent
appearance, whi ch shines through the tighter order, and
lets us feel at peace.
524
i t s a g e l e s s c h a r a c t e r
If we think carelessly, we may imagine that this gentl e
character comes merel y f rom the fact that they were made
i nf ormal l y, wi thout machines, and sl owl y over time. But
the truth is that this character arises in these buildings not
because of history, or because the processes whi ch built
them were so primitive. Thes e buildings have this char-
acter because they are so deep, because they were made
by a process whi ch al l owed each part to be entirely one
wi th its surroundi ngs, in whi ch there is no ego l ef t, only
the gentl e persuasion of the necessities.
Yet this character cannot be generated by a ferson
yearning for the ancient fast.
It happens simply because when you understand as much
about the forces whi ch surround us as a person wi th a
l i vi ng pattern l anguage does, and when you build accord-
i ng to these forces, the kinds of buildings whi ch you
make, are simply more like ancient buildings than like
modern ones.
Wh a t looks at first sight like an accidental quality whi ch
marks the t owns and villages of the past, turns out to be
the most f undament al physical property of the worl d we
live in.
It is simfly the character of buildings which re-fleet
the forces in them froferly.
T h e prismatic buildings of our own time, the buildings
built with the simple geomet ry of cubes, and circles,
525
t h e w a y
spheres, and spirals, and rectangl es; this geomet ry is the
na'i've order, created by the childish search for order. We
happen to think of this order as the proper order for a
building, because we have been taught to think so; but we
are wr ong.
T h e proper order for a building or a t own, whi ch
comes about when buildings are correctl y fitted to the
forces in them, is a much richer order, wi th a far more
compl ex geomet ry. But it is not merel y rich and compl ex;
it is also very specific. And it wi l l show itself, under any
circumstances, where buildings are actually correct. Wh e n -
ever anyone manages to make a building whi ch is alive,
it wi l l have this specific character, because that is the only
character whi ch is compatible wi th life.
When I myself first started, to make buildings with
this character, the character amazed me.
A t first, I was afraid that I must really be a conservative
at heart and that I was unconsciously striving to re-make
the past.
But then I read a passage in an ancient Chi nese paint-
i ng ma nua l t he Must ard Seed Gar den manual of paint-
i ng whi c h made the situation clear to me.
T h e wri ter of that manual describes how, in his search
for a way of painting, he had discovered for himself the
same central way that thousands of others like him had
also discovered for themselves, t hroughout the course of
history. He says that the more one understands of paint-
i ng, the more one recogni zes that the art of painting is
526
i t s a g e l e s s c h a r a c t e r
essentially one way, whi ch will al ways be discovered and
rediscovered, over and over agai n, because it is connected
wi th the very nature of painting, and must be discovered
by anybody who takes pai nti ng seriously. T h e idea of style
is meani ngl ess: what we see as a style (of a person or of
an age) is not hi ng but another individual ef f ort to pene-
trate the central secret of painting, whi ch is gi ven by the
T a o , but cannot itself be named.
T h e more I learn about t owns and buildings, the more
I feel the same thing to be true. It is true that many of
the historic styles of buil ding have some quality in com-
mo nt he y have it not because they are old, but because
man has, over and over agai n, approached the secret
whi ch is at the heart of archi tecture. I n fact, the principles
whi ch make a building good, are simple and di r e c t
they f ol l ow directly f rom the nature of human beings, and
the l aws of nat ur e and any person who penetrates these
l aws will, as he does so, come closer and closer to this
great tradition, in whi ch man has sought for the same
thi ng, over and over agai n, and come al ways to the same
conclusions.
And it is because this same morfhology, underlying
all things, will always arise in the endthat the time-
less way of building is a truly timeless one.
As you learn to make buildings more and more alive, and
thereby more and more true to their own nature, you wi l l
inevitably approach this timeless character.
Thes e are the f orms whi ch man has f ound, time and
527
t h e w a y
time agai n, as he approaches the heart of building. As the
acts of building in a communi t y are governed more and
more by a common pattern l anguage, they wi l l more and
more closely create and re-create the body of timeless
f orms whi ch have been part of archi tecture since society
began.
The timeless character of buildings is as much a fart
of nature as the character of rivers, trees, hills, flames,
and stars.
Each class of fhenomena in nature has its own char-
acteristic morfhology. Stars have their character;
oceans have their character; rivers have their char-
acter; mountains have their character-, forests have
theirs; trees, flowers, insects, all have theirs. And
when buildings are made froferly, and true to all the
forces in them, then they too will always have their
own specific character. This is the character created by
the timeless way.
It is the fhysical embodiment, in towns and build-
ings, of the quality without a name.
528
T H E
K E R N E L OF
T H E W A Y
And yet the timeless way is not
complete, and wi l l not f ul l y generate
the quality without a name, until we
leave the gate behind.
CHAPTER 27
THE KERNEL OF THE WAY
Indeed this ageless character has
nothing, in the end, to do with lan-
guages. The language, and the processes
which stem from it, merely release the
fundamental order which is native to us.
They do not teach us, they only remind
us of what we know already, and of what
we shall discover time and time again,
when we give uf our ideas and opinions,
and do exactly what emerges from our-
selves.
From what you have read so far, it may seem as
though the life of buildings, and the timeless char-
acter they have when they are living, can be created
simfly by the use of pattern languages. If the people
have a living language, it seems that what emerges
from their acts of building will be alive; it seems as
though the life of towns can be created simfly by the
use of languages.
And yet, we wonder, can it be so simfle? Can any
process really generate the nameless quality which
stirs the heart of nature? Can any theory be so
powerful?
These doubts are right. There is a kernel at the
center of the timeless way, a central teaching, which
I have not described till now.
T h e essence of this kernel is the fact that we can only
make a building live when we are egoless.
Imagi ne, for example, blue tiles, white fountains, birds
nests under the arcades, yel l ow paint, fresh scrubbed
woodwork.
Ornament s around the edge of the roof, red flowers in
bushes round the entrance, great wi ndows, filled with
cushions, flower pots where seedlings are gr owi ng, and a
broom hangi ng on the wal l . . . pinnacles high in the
sky, the vaults of the building catching the light of the
sun, deep shadows in the alcoves around the building
edge.
Th e beauty of this place, the quality in it whi ch touches
us, the thing whi ch makes it live, is, above all, that it is
carefree, that it is innocent.
535
T HE KERNEL OF T HE WAY
This innocence will only come about when people hon-
estly forget themselves.
It goes without saying that the vast steel and glass and
concrete structures of our famous architects do not have
this quality.
It goes without saying that the mass-produced develop-
ment houses, built by big developers, do not have this
quality.
But it is true that even the more " nat ur al " architects,
like Frank Ll oyd Wr i g ht and Al var Aal to, also do not
reach this quality.
And it is also true that the " f unk y " relaxed hippy-style
architecture, with irregular redwood facades, and old-
fashioned country style interiors, also does not reach this
quality.
These places are not innocent, and cannot reach the
qualify without a name, because they are made with an
outward glance. T h e people who make them make them
the way they do because they are trying to convey some-
thing, some image, to the worl d outside. Even when they
are made to seem natural, even their naturalness is calcu-
l ated; it is in the end a pose.
In case you think that I am simply refecting my own
time} and searching for the past, I should like to tell
you about two places I know which are entirely of the
twentieth century, and have this innocence.
One is a fruit stand, on a country road, not far from here.
It is a simple shelter, made of corrugated iron and pl y-
536
T HE KERNEL OF T HE WAY
woodi t has no purpose whatsoever, except to protect
the fruit.
Th e other is the deck of a fishing boat, in the Nort h
Sea. It is a simple diesel-engined fishing boat, perhaps a
40-foot boat. Thr e e Danish brothers fish from it. In one
corner there is al ways a huge pile of empty beer bottles,
perhaps 3 or 4 feet hi gh; they drink continuously while
they are at sea, and while they are in port.
These t wo places have a little of the innocence and ego-
lessness whi ch is necessary to the quality wi thout a name.
And why? Because the people who made them simply do
not care what people think of them. I don' t mean that
they are defi ant: people who defiantly don' t care what
other people think of them, they still care at least enough
to be def i ant and it is still a posture. But in these t wo
cases, the fruit stand, and the deck of that fishing boat,
the people did not care what other people t hought ; and
they also do not care about the fact that they don' t care.
It means nothing to them. The y only do exactly what
they have to do to take care of their situation.
And of course, there are larger examfles too.
A concrete yard, or a steel mill, where there is no de-
sire to impress at all, only the need for things to work
. . . that sometimes has this quality. A farmyard has it
often, for the same kind of reason; or a new cafe, where
the owners have too little money to do anythi ng to i m-
press, and concentrate only on maki ng their customers
feel comfortabl e, in the true sense, with a bare mi ni mum.
And, of course, sometimes it exists too in a house, built
537
T HE KERNEL OF T HE WAY
some time ago, the flowers round the house, the rambl ing
trellises, looked after with patience, hidden behind a great
wal l , not visible, done only out of love and out of the de-
sire to live, and let the roses bloom.
To make a building egoless, like this, the builder must
let go of all his willful images, and start with a void.
Archi tects sometimes say that in order to design a build-
ing, you must have "an i mage" to start wi th, so as to give
coherence and order to the whol e.
But you can never create a natural thing in this state of
mind. If you have an i deaand try to add the patterns to
it, the idea controls, distorts, makes artificial, the work
which the patterns themselves are trying to do in your
mind.
Instead you must start with nothi ng in your mind.
You are able to do this only when you no longer fear
that nothing will happen, and you can therefore afford
to let go of your images.
At first, when you are still unsure that a pattern l anguage
will genuinely generate forms inside your mind, you hang
on, tight, to all the images you have, because you are afraid
that wi thout them, there may be nothing left. Once you
learn that the pattern l anguage and the site together, wi l l
genuinely generate form inside your mind, f rom nothing,
you can trust yourself to let go of your images entirely.
For a person who is unfree, the l anguage seems like mere
538
T HE KERNEL OF T HE WAY
information, because he feels that he must be in control,
that he must inject the creative impulse, that he must sup-
ply the i mage whi ch controls the design. But once a person
can relax, and let the forces in the situation act through
him as if he were a medi um, then he sees that the l an-
guage, wi th very little help, is able to do almost all the
work, and that the building shapes itself.
Thi s is the importance of the void. A person who is free,
and egoless, starts wi th a void, and lets the l anguage gen-
erate the necessary forms, out of this void. He overcomes
the need to hold onto an image, the need to control the
design, and he is comfortabl e with the void, and confident
that the l aws of nature, formul ated as patterns, acting in
his mind, will together create all that is required.
At this stage, the buildings life will come directly
from your language.
A man who is not afraid to die, is free to live because he is
open to what happens next, and is not al ways killing it by
trying to control it.
In the same way, the l anguage and the building it creates
begins to come to life when I begin to be relaxed about
what happens next. I can work within the order of the
l anguage, without worryi ng about the patterns whi ch are
comi ng later, because I am sure that, no matter what hap-
pens, I will al ways be able to find a way of bri ngi ng them
into the design, when I come to them. I don' t need to take
precautions in advance. Wh y am I so sure that I can al-
ways find a way of bringing in the smaller patterns? Be-
539
T HE KERNEL OF T HE WAY
cause I don' t care what shape the finished building or its
details haveprovi ded only they are natural. I have no
preformed mould that I am trying to pour the patterns
into; I don' t mind how strange, how curious, the building
turns out to be, if I can only satisfy the patterns.
Sometimes a wi l l ow tree whi ch grows in an awkwar d
corner of a garden, ends up bul gi ng and twisted, as it fits
itself to the forces in the garden. But it is no less natural,
no less free. If a building which I make turns out bulging
and twisted it will be no less free than the wi l l ow tree.
And it is because of this, because I am not afraid of these
deformities, that I can al ways take the patterns in the or-
der of the l anguage; and because of this that I can al ways
make a building whi ch is natural and free, like the wild
wi l l ow.
Yet, at the very moment when you first relax, and let
the language generate the buildings in your mind, you
will begin to see how limited your language is.
Once you realize that the only thing whi ch matters is the
reality of the situation whi ch surrounds the building, and
not your images of it, you are able to relax, and al l ow the
patterns of the l anguage to combine themselves freely in
your mi nd, wi thout trying to impose an artificial image on
their combination.
But at the same moment , you will begin to realize that
the reality of the situation is not only more important than
your images, but also more important than the l anguage
too. T h e l anguage, no matter how useful or how power-
540
T H E K E R N E L OF T H E WA Y
f ul , is fallible, and you cannot accept its patterns aut o-
mat i cal l y, or hope that they wi l l ever generat e a l i vi ng
t hi ng mec hani c al l ybec aus e, once agai n, it is onl y the ex-
tent to whi ch you yoursel f become ordi nary and nat ural ,
that in the end det ermi nes how nat ural , and f ree, and
whol e the bui l di ng can become.
One flace can have "good" patterns in ity and yet be
dead.
For i nstance, it so happens that there is a little square in
San Fr anci sco whi ch has f our pat t erns in it. T h e square is
smal l ; there is a BUILDING THOROUGHFARE; there are
ACTIVITY POCKETS in i t; and there are STAIR SEATS. But
each of these patterns is subtl y wr o ng . T h e space is smal l
y e s . But it is pl aced in such a wa y that it wi l l never be
used, so the density of pedestri ans wi l l still be too l o wt h e
point of the SMALL PUBLIC SQUARE, to make the pl ace f eel
f ul l , is mi ssi ng. T h e r e is a BUILDING THOROUGHFARE
t hrough the pl ace, l eadi ng into a bui l di ngyes . But there
are no places f or i nvol vement al ong this path, so the path
becomes i rrel evant , and helps not hi ng come to l i fe. T h e r e
are ACTIVITY POCKETSy e s t he r e are smal l corners
pl aced around the edge of the square. But t hey are so
pl aced that no acti vi ty coul d ever gat her in t h e mt h e y
have the wr o n g relationship to the paths of access. An d the
pl aces wher e people would, nat ural l y gat her , if t hey coul d,
are obstructed by stairs, and barriers.
T h e square has these patterns in i t b u t it still f ai l s be -
cause in each case the point of the pat t ern, the spirit of the
5 41
T HE KERNEL OF T HE WAY
pattern, is missing. For the man who made this square,
these patterns were empty formal tools, whi ch didn' t help
the place to come alive at all.
Another place can be without the patterns which apply
to it, and yet still be alive.
For the same square could also not have these patterns in
it, and yet be whol e.
Suppose the square is l ar gepar t of it could be treated
like a par kand the corner of it where people are most
likely to gather, made into a small, partly enclosed space.
Thi s woul d be in the spirit of SMALL PUBLIC SQUARE,
but without f ol l owi ng the letter.
Suppose there is no way of maki ng a path go through the
square for BUILDING THOROUGHFAREthen it woul d be
possible to put a children' s pl ayground, or a grove of trees,
at the back, and t wo activity pockets, right on the street.
Thi s woul d be in the spirit of BUILDING THOROUGHFARE,
by using the street itself as a tangent t horoughf arebut
again wi thout f ol l owi ng the letter of the pattern.
So l ong as you are using patterns slavishly, mechanically,
they will interfere with your sense of reality, as much as
any other images. Yo u will be able to use them properly,
only at that moment when you have the proper disregard
for them.
So paradoxically you learn that you can only make a
building live when you are ]ree enough to reject even
the very patterns which are helping you.
542
T HE KERNEL OF T HE WAY
And, in the end, the buildings will become alive only when
the person who uses the l anguage is himself egoless and
free. Onl y then will he be able to recognize the forces as
they really are, instead of being overawed by images.
But at that moment he no l onger needs the l anguage.
Onc e a person has freed himself to such an extent, that he
can see the forces as they really are, and make a building
whi ch is shaped by them alone, and not affected or dis-
torted by his i mageshe is then free enough to make the
building wfthout patterns at al l because the knowl edge
whi ch the patterns contain, the knowl edge of the way the
forces really act, is his.
It may seem to you that pattern languages are useless
then.
If it is true that you cannot make a building live, even with
the help of a pattern l anguage, unless you are first egoless
and f ree; and if it is also true that once you have reached
this state of freedom, you will be able to make a living
thing, no matter how you do i t; it seems to follow then
that the pattern l anguage is useless.
But it is just your -pattern language which helps you
become egoless.
T h e patterns in a living l anguage are based on funda-
mental realities, whi ch everyone already knows, in his in-
nermost self. Yo u know that small alcoves, arcades, l ow
ceilings, opening wi ndows, sheltering roofs, make funda-
543
T HE KERNEL OF T HE WAY
mental senseand you forget it only because our society
has filled your mind with other distorting images. Th e
l anguage only shows you what you yourself already know.
Th e l anguage is able to awaken you to your own inner-
most feelings, and to what is true. Gradual l y, by f ol l owi ng
the l anguage you feel free to escape from the artificial
images which society has imposed upon you. And, as you
escape from these images, and the need to manufacture
things according to these images, you are able to come
more into touch wi th the simple reality of things, and
thereby become egoless and free.
The language frees you to be yourself, because it gives
you permission to do what is natural, and shows you
your innermost feelings about building while the
world is trying to suppress them.
One student of architecture, here in Berkel ey, his mind
filled with the images of steel frames, flat roofs, and mod-
ern buildings, read the ALCOVE pattern, then came to
his teacher and said, in wonderment : " I didn' t know we
were al l owed to do things like this. " Allowed!
T h e more I watch our pattern l anguage being used, the
more I realize that the l anguage does not teach people
new facts about their envi ronment. It awakens old feel-
ings. It gives people permission to do what they have al-
ways known they wanted to do, but have shunned, in
recent years, because they have been frightened and
ashamed by architects who tell them that it is not " mod-
ern. " People are afraid of being laughed at, for their ig-
norance about " a r t " ; and it is this fear which makes them
544
T HE KERNEL OF T HE WAY
abandon their own stable knowl edge of what is simple
and right.
A language gives you back your confidence in what
seemed once like trivial things.
Th e first t hi ngst he innermost secret likes and dislikes
we havear e fundamental .
We give them up, and try to be important, and c l e v e r
because we are afraid that people will laugh at us.
T h e SHELTERING ROOF, for instance, is so full of feel-
ing, that many people daren' t admit it.
A l anguage will allow this inner thing, whi ch carries
feelings, to guide your acts.
At this final stage, the patterns are no longer impor-
tant: the patterns have taught you to be receptive to
what is real.
It is no l onger the pattern ALCOVE which tells you to
create alcoves. It is because you see reality in a particular
case. And the reality you see shows you that a particular
alcove is the right thing to do.
The pattern ALCOVEwhi ch first functioned as an in-
tellectual crutchis no longer necessary to you. You
see reality directly, like an animal. You make the al-
cove as an animal might make an alcovenot because
of the conceptbut directly, simply because it is ap-
propriate.
545
T HE KERNEL OF T HE WAY
At this stage, you work directly with what is real.
But do not think, casually, that you can do this now, and
that you therefore do not need a l anguage to enable you to
do it. At this moment, you cannot see the reality: because
your mind is full of images and concepts. And while you
are at the stage where you rely on images and concepts
(style, flat roofs, sheets of glass, white painted steel, thick
redwood trim, shingles, rounded corners, diagonal lines)
you cannot face reality di rect l yyou cannot tell the dif-
ference between what is real, and what unreal. And in this
state, the only way you can escape these images, is to re-
place them with more accurate i magest hat is what pat-
terns are. But finally, you can release yourself from images
entirely.
And in this sense, the language is the instrument
which brings about the state of mind} which I call
egoless.
It is the use of pattern l anguages which after patient use,
will allow you to come back to that part of yourself which
has al ways been there, and is there now, but is obscured by
images and ideas and theories, which make it impossible for
you to be yourself, or act as nature does.
Th e impulse to make wi ndows overl ooking life, to make
ceilings vary in height, columns thick enough to lean
against, small wi ndow panes, sheltering steeply sloping
roofs, arcades, seats by the front door, bay wi ndows, al-
coves, is already part of you. But you have been told so
much, that you no l onger value these inner-impulses. Yo u
546
T HE KERNEL OF T HE WAY
curb them, because you think that someone else knows
better. Yo u are perhaps afraid that people may laugh at
you for being so ordinary.
A pattern l anguage does nothi ng really, except to wake
these feelings once again.
It is the gate which leads you to the state of mind, in
which you live so close to your own heart that you
no longer need a language.
I t is utterly ordinary. It is what is in you already. Yo ur
first, most primitive impulses are right, and will lead you
to do the right thing, if you will only let yourself.
The r e is no skill required. It is only a question of whether
you will al l ow yourself to be ordinary, and to do what
comes naturally to you, and what seems most sensible, to
your heart, al ways to your heart, not to the images which
false l earni ng has coated on your mind.
This is the final lesson of the timeless way.
Imagi ne yourself building a simple porch outside your
room: a col umn to lean against; a gusset to strengthen its
connection to the beam; decorated with f ret work, so that
the light falls softly, without glare f rom the sky; a rail to
lean on easily, so that you can wal k out and lean and smell
the summer ai r; and the yel l ow sunlight, lit by the yel l ow
grass, war mi ng the unpainted wooden planks.
Imagi ne that you have reached the point, in your life,
547
T HE KERNEL OF T HE WAY
where you are maki ng such a porch. Yo u are a di fferent
person now. T h e fact that you have understood the im-
portance of these details, in your life, that you have under-
stood how much they influence your life, means that you
are now alive in a more simple sense.
So finally you learn that you already know how to create
this ageless species whi ch is the physical embodiment, in
buildings, of the quality without a name, because it is a part
of y oubut that you cannot come to it until you first
master a pattern l anguage, and then pass beyond the l an-
guage, once it has taught you to al l ow yourself to act as
nature does.
To act as nature does is the most ordinary thing in the
world. It is as ordinary as a simple act of slicing straw-
berries.
One of the most movi ng moments in my life, was also one
of the most ordinary. I was with a friend in Denmar k.
We were having strawberries for tea, and I noticed that
she sliced the strawberries very very fine, almost like
paper. Of course, it took longer than usual, and I asked
her why she did it. Wh e n you eat a strawberry, she said,
the taste of it comes from the open surfaces you touch.
Th e more surfaces there are, the more it tastes. T h e finer
I slice the strawberries, the more surfaces there are.
Her whol e life was like that. It is so ordinary, that it is
hard to explain what is so deep about it. Ani mal almost,
nothing superfluous, each thing that is done, done totally.
T o live like that, it is the easiest thing in the wor l d; but
548
T HE KERNEL OF T HE WAY
foi a man whose head is full of images, it is the hardest. I
learned more about building in that one moment, than in
ten years of building.
When we are as ordinary as that, with nothing left
in any of our actions, except what is requiredthen
we can make towns and buildings which are as in-
finitely various, and as peaceful, and as wild and liv-
ing, as the fields of windblown grass.
Almost everybody feels at peace with nature: listen-
ing to the ocean waves against the shore, by a still
lake, in a feld of grass, on a windblown heath. One
day, when we have learned the timeless way again, we
shall feel the same about our towns, and we shall feel
as much at peace in them, as we do today walking by
the ocean, or stretched out in the long grass of a
meadow.
549
ACKNOWLEDGMENTS
During the fourteen long years that it has taken me to write
this book, I have been helped by three people, above all. First,
by my beloved Ingrid, who inspired me always, who under-
stood so clearly what I was trying to do, who was always ready
to talk, to look at it again, to help me see exactly what feelings
were created by a passage in it, and who found many of the
most beautiful photographs. Second, by my dear friend Sara,
who came, year after year, at a moment's notice, to discuss a
chapter, or a page, or a sentence . . . but most often, during
our work together, helped me in our discussions of the theory
as a whole. And third, by Peter Mailloux, who spent almost a
year, the most difficult year of all, helping to see me through
the final editing.
As for the photographs, many of them come from books and
magazines long out of print, and it has not always been pos-
sible to find the names of the photographers. However, both in
the few cases which follow where I have been able to identify
the photographer, and in all those other cases where I have not,
I am immensely grateful to the photographers for their won-
derful pictures, which do so much to explain the central mean-
ing of the book.
55
A C K N O W L E D G M E N T S
6 Erik Lundberg 190 Ed Allen (drawing)
20 E. O. Hoppe 194 Werner Bischof
2 +
Augustin Myska 196 David Vestal
+ 2
Marvin Bolotksy 197 John Durniak
44
Henri Cartier-Bresson 2 12 Erik Lundberg
+6 Ernst Haas 213 Eugene Atget
56
Bernard Wolf 21 4 Luc Joubert
57
Bernard Wolf 278 Henri Cartier-Bresson
O
O
Henri Cartier-Bresson 280 Bruce Davidson
59
Andre Kertesz
281 Henri Cartier-Bresson
60 Andre Kertesz
366 Lennart Nilsson
61 Henri Cartier-Bresson 388 Ursula Pfistcrmcister
76 Henri Cartier-Bresson
389
Eugene Atget
102 Roderick Cameron 406 Erik Lundberg
0
0
K. Nakamura 408 Erik Lundberg
142 David Sellin 480 Eugene Atget
0
0
Kocjanic
518 Prip Moiler
I 7 0 Bernard Rudofsky
535
Henri Cartier-Bresson
I 8 9 Ed Allen (drawing) 536 Werner Bischof