Content uploaded by Tatjana Schneider
Author content
All content in this area was uploaded by Tatjana Schneider on Oct 14, 2016
Content may be subject to copyright.
Introduction1
The first paper (arq 9/2, pp. 157–16 6) has already
made the arguments as to why flexible housing is a
relevant, even essential, part of future housing
provision on grounds of social, economic and
environmental sustainability. This paper examines
ways in which flexible housing may be achieved,
using examples from twentieth-century housing.
It is first necessary to define what we mean by
flexible housing. Our definition determines flexible
housing as housing that can adapt to the changing
needs of users. This definition is deliberately broad.
It includes the possibility of choosing different
housing layouts prior to occupation as well as the
ability to adjust one’s housing over time. It also
includes the potential to incorporate new
technologies over time, to adjust to changing
demographics, or even to completely change the
use of the building from housing to something else.
So flexible housing in our definition is a wider
category than that of adaptable housing, which is
the term generally used to denote housing that can
adapt to users’ changing physical needs, in
particular as they grow older or lose full mobility.2
With such a broad definition, it is maybe not
surprising that there are multiple methods of
achieving flexibility. The approach in this paper is
not prescriptive and does not attempt to organise
these different approaches into an overall
methodology. Instead it explores flexibility through
the issue of determinate and indeterminate design,
or as we identify ‘hard’ and ‘soft’ systems. This differs
from the previous approaches of categorising
flexibility, most notably that derived out of the
open building movement. Kendall and Teicher’s
Residential Open Building is probably the most
comprehensive and rigorous analysis of flexible
housing of recent years.3However, while the social
intents and economic rationale are admirable,
the method tends towards suggesting an all-or-
nothing approach that is eventually technically
determined. This is potentially off-putting to the
housing designer or provider. Our approach starts
with some basic principles and then suggests
alternative methods of achieving flexibility that
allow anything from wholesale change down to
quite discreet, but potentially extremely useful
alterations.
Built-in obsolescence
The basic principles of flexibility start with its
opposite – namely that inflexibility should be
designed out. As noted in the first paper, the UK
building industry tends to build in obsolescence, but
this can be avoided in three manifestly simple, and
non-costly, ways. First, through the consideration of
the construction; most directly through the
reduction of loadbearing or solid internal partitions
but also through the avoidance of forms of roof
construction (i.e. trussed rafters) that close down the
possibility of future expansion. Second, through
technological considerations and in particular the
reduction of non-accessible or non-adaptable
services. Third, through consideration of the use of
space, i.e. through the elimination of tight-fit
functionalism and rooms that can be used or
accessed in only one way.
While these three relatively simple principles
would go some way to avoid inflexibility, one other
aspect also needs to be addressed. To move from
avoiding inflexibility to building-in flexibility it is
useful to look to see if we can find generic principles
in two building types that are often described as
inherently flexible, the English terraced house and
the speculative office.
The terraced house
In the course of our research by far the most
commonly mentioned example of flexibility was the
London terraced house, in particular the late
eighteenth-/early nineteenth-century examples
which are typically flat-fronted and three or four
stories, plus basement. Over the course of the years,
these have been added both horizontally (at the
back) and vertically (the ubiquitous loft extension),
knocked through, divided, joined up again and used
for countless other purposes.4It is worth, therefore,
analysing the aspects of these houses that give them
their inherent flexibility. The first is the relatively
generous space provision in relation to
theory arq .vol 9 .nos 3/4 .2005 287
theory
How might flexible housing be achieved? ‘Determinate’ and
‘indeterminate’ approaches are examined using twentieth-
century examples
Flexible housing: the means to the end
Jeremy Till and Tatjana Schneider
contemporary standards. This has allowed
subdivision both horizontally and vertically;
typically the front room might be divided to provide
a kitchen next to the living room, or a bathroom
might be inserted between the first floor front and
back rooms; vertically the generous height allows
false ceilings for services where necessar y. Second,
their construction repeats a small number of simple
techniques, allowing the use of relatively unskilled
labour (the infamous and probably mythical firm of
Bodgit and Creepaway comes to mind). Finally, the
basic layout, and in particular the placing of the
staircase allows additions to be made at the rear of
the building in a flexible and infinitely variable
manner (the ubiquitous back extension).
While not all these aspects of the terraced house
can be transferred directly to the contemporary
context, they do begin to suggest some generic
principles for flexible housing. These are:
1Space: There is a correlation between amount of
space and amount of flexibility. Some recent
schemes have exploited this correlation by
providing more space but at lower specification,
arguing that flexibility in the occupation of space
is of more importance than the niceties of having a
fitted kitchen or fully decorated rooms. One such
is the Nemausus scheme in Nîmes, France (1985)
designed by Jean Nouvel. Here undivided space
with double-height areas was handed over in a
semi-finished state for the tenants to fit out,
though their ultimate choice was restricted by a
number of impositional rules that dictated things
like the colour of their curtains.5This is also
essentially the approach of the loft developer,
where an excess of undefined space is handed over
to the tenant to use as they will. The corollary of
the space-flexibility correlation is that limited
space may be seen to limit flexibility, but at the
same time there are often demands to use that
space in multiple ways; in these cases you have to
work harder to achieve flexibility through other
design methods.
2Construction: There is a relationship between
construction techniques and flexibility. The
specialist and multi-headed approach to housing
construction, particularly in the UK, limits future
flexibility in so much as one needs specialised and
multiple skills to make any adaptations. Against
this, following the example of the terraced house,
many of the most successful contemporary
flexible housing schemes rely on simple and
robust construction techniques, which allow
future intervention, or at least place the specialist
elements such as services in easily accessible and
separate zones so that only one set of specialists is
needed to make changes. This latter strategy is in
contrast to standard construction when just to
update, say, wiring one needs to get, in addition to
the electrician, a carpenter to lift floors, a plasterer
to patch the ceiling and a decorator to make good
in addition to the electrician.
3Design for adaptation: Simple, but considered
design moves such as the placing of staircases,
service cores or entrances allow future flexibility at
no extra cost. However, for this approach to be
successful the designer has to project future
scenarios and adaptations onto the plan to see
what can be accommodated. Exemplary of this
approach is the work of the architect Peter
Phippen in the UK, with the development of his
wide frontage house with a central service and
staircase core. This plan form allows free
disposition of rooms around the edges, as well the
possibility to make additions to the rear of the
property.
A few contemporary architects have specifically
followed these principles, most notably Caruso St.
John’s project for terraced housing in Berlin (2001)
which draws on the plan form of the traditional uk
terraced house, but provides still further flexibility
by making all the internal walls non-loadbearing.6
The scheme is at the same time simple and
sophisticated.
The speculative office
The second type of accommodation that is often
mentioned as inherently flexible is speculative
commercial offices. These are designed with no
specific tenant in mind and allow continual
adaptations to be made to the basic shell to suit the
occupants at any given time.7Importantly, they also
allow upgrading and easy relocation of services.
Again, we can identify generic principles that can be
transferred to flexible housing;
4Layers: The clear identification of layers of
construction from structure, skin, services,
internal partitions to finishes. This allows
increasing control (and so flexibility) as one goes
through the layers.8
5Typical plan: The speculative office is the classic
shell and core structure. The external shell is
relatively inflexible; the core provides access and
services. In between the space is indeterminate,
with large spans and open plans allowing non-
loadbearing partitions to be put in and removed at
will. The speculative office, almost by definition,
provides generic space, in contrast to the highly
specific and determined space that one finds in
most housing. It is what Rem Koolhaas denotes the
‘typical plan … zero degree architecture,
architecture stripped of all traces of uniqueness or
specificity’.9
6Services: The disposition of services in the
speculative office is carefully considered to allow
future change and upgrading. Vertical services are
marshalled into easily accessible ducts.
Horizontally, raised floors and/or dropped ceilings
allow endless permutations in the eventual
disposition of service outlets.
Some of these principles from the speculative office
have already been exploited in schemes such as
Immeubles Lods in France (1972) which explicitly
used office building technology and planning
principles to give maximum flexibility in the
arq .vol 9 .nos 3/4 .2005 theory
288
Jeremy Till and Tatjana Schneider Flexible housing: the means to the end
housing (1970). The obverse is also becoming quite
common, with 1960s and ’70s office buildings, which
no longer have the floorplates or storey height to
deal with contemporary office needs and servicing,
lending themselves to be converted into housing.
Using these six principles of inherent flexibility
from the terraced house and the speculative office
you can begin to enter the territory of designing
specifically for flexibility in housing. The next stage
of our research, and this paper, attempts to develop a
more comprehensive classification of methods by
which flexibility may be achieved in housing. Our
classification draws on some previous approaches,10
and while it attempts to cover most conditions of
flexibility, it does not intend to be prescriptive in
setting out a set of rigid rules for potential designers.
Our classification works in two directions. First,
through investigating flexibility at different scales of
housing (from the block, through the building and
unit, to the individual room), and second, by
indicating methods by which flexibility has been or
may be achieved. The rest of the paper concentrates
on the latter; we propose two broad categories, use
and technology. Use refers generally to the way that
the design affects the way that housing is occupied
over time, and generally refers to flexibility in plan.
Technology deals with issues of construction and
servicing, and the way that these affect the potential
for flexibility. We then subdivide each of these two
categories into ‘soft’ and ‘hard’ techniques.
Hard and soft
‘Soft’ refers to tactics which allow a certain
indeterminacy, whereas ‘hard’ refers to elements
that more specifically determine the way that the
design may be used. In terms of use it may appear a
contradiction that flexibility can be achieved
through being either very indeterminate in plan
form or else very determinate, but historically both
approaches have developed in parallel throughout
the twentieth century. Soft use allows the user to
adapt the plan according to their needs, the designer
effectively working in the background.11 With hard
use, the designer works in the foreground,
determining how spaces can be used over time. As we
shall see, soft use generally demands more space,
even some redundancy, and is based on a relaxed
approach to both planning and technology, whereas
hard use is generally employed where space is at a
premium and a room has to be multifunctional.
Soft use
The notion of soft use is embedded in the vernacular
house. As Paul Oliver, the key historian of vernacular
architecture notes, ‘with the growth of families,
whether nuclear or extended, the care of young
children and the infirm, and the death of the aged,
the demands on the dwelling to meet a changing
family size and structure are considerable’.12 In
vernacular housing, the range of responses to these
issues is then oriented by culture and climate,
theory arq .vol 9 .nos 3/4 .2005 289
Flexible housing: the means to the end Jeremy Till and Tatjana Schneider
1
1Britz Housing, Berlin
(1925) by Taut and
Wagner. Typical plan
from the
‘horseshoe’ block
ranging from the single space used for the whole
gamut of family rituals to the simple division of the
traditional cottage. Clearly, the intervention of the
expert designer removes such approaches from the
realm of the vernacular, but nonetheless the
principles of soft use remain. Some of the pioneering
examples of Modernist housing explicitly use the
notion of indeterminacy as a response to the
housing shortage crisis in the 1920s and ’30s, arguing
that a flexible approach could cater for a wider range
of occupants. Typical of these is the Britz Project in
Berlin (1925–31) designed by Taut and Wagner. The
design provides three similarly sized rooms (denoted
on the plan simply as ‘Zimmer’ – ‘room’) off a central
hallway, with the services (bathroom and kitchen) in
a separate zone. The occupation of the rooms is
thereby left open to the interpretation by various
possible user groups [1].
A more refined version of this strategy can be
found in one of the classic projects of Czech
Modernism, the Letná project in Prague (1935) by
Eugene Rosenberg in which each floor typically
comprises two apartments of different size.13 Within
the individual apartments, rooms are of an equal
size and can be accessed separately from a central
lobby, while the services are contained in a separate
zone [2].
This indeterminacy approach is also beautifully
exploited in the Hellmutstrasse scheme in Zurich
(199 1) by ADP Architektur und Planung, which is one
of the most sophisticated flexible housing schemes
of recent years. The project arose out of a community
led approach with all members of a housing
cooperative contributing to the design process. The
design is spilt into three distinct horizontal zones. At
the top is a line of similarly sized rooms divided by
loadbearing partitions, and with the possibility of
inserting non-loadbearing partitions to define
circulation. Below this there is a row of serviced
spaces that can be either bathrooms or kitchens.
Finally, at the lowest level, there is a zone containing
what is usually a kitchen and living space, but which
can also be used as a self-contained studio
apartment. All apartments are accessed from an
external staircase and shared balconies. The overall
arrangement allows multiple arrangements to be
achieved, from large groups of single people living
together right down to self-contained one-person
studio apartments. The zoning also allows future
changes to be made with ease, though in practice
this would require neighbouring apartments to
simultaneously agree to changes in terms of
expansion and contraction [3].
If one approach to soft use depends on the
designer providing a physically fixed, but socially
flexible, layout, a more common solution is to
provide raw space that can then be divided according
to the needs of the occupants. This is not as
straightforward as it may sound. Provision of open
space alone does not suffice, or at least it may be
inefficient in terms of space usage. The designer has
to carefully consider the best points for access
(generally in the centre of the plan), the position of
servicing (either in specific zones or else widely
arq .vol 9 .nos 3/4 .2005 theory
290
Jeremy Till and Tatjana Schneider Flexible housing: the means to the end
2
3
2Letná apartment
block, Prague (1935)
by Eugene
Rosenberg. Typical
plans
3Hellmutstrasse,
Zurich (1991), by
ADP Architektur und
Planung: possible
arrangements
theory arq .vol 9 .nos 3/4 .2005 291
Flexible housing: the means to the end Jeremy Till and Tatjana Schneider
4
4Weissenhofsiedlung,
Stuttgart (1927) by
Mies van der Rohe.
Generic layout
(bottom), typical
plans (top)
5Montereau-Surville
(1971): Les Frères
Arsène-Henry:
‘empty’ plans
prior to internal
division
distributed) and the most efficient module size (a
standard module allows repetition in structural
division and components but should not limit
options for subdivision). To ensure an efficient and
flexible use of space, designers will often use
hypothetical layouts to test their access, servicing
and module strategies.
5
This approach to indeterminate open space was
facilitated by the new constructional systems
available to the early Modernist architects, which
allowed larger span structures and light infill
partitions. Some of the earliest examples can
therefore be found in classic Modernist housing
schemes such as the model housing at the
Weissenhofsiedlung in Stuttgart (1927). In the block
designed by Mies van der Rohe, a simple framed
structure allows the residents to divide their
apartments up as they wish, though the final layout
is limited by the positioning of columns; the
provision of the servicing is also minimal by today’s
standards [4].
A more developed approach can be found in
schemes such as that at Montereau-Surville in France
(197 1) designed by Les Frères Arsène-Henry, or the
Järnbrott Experimental Housing in Sweden (1953). In
a 10-storey building at Montereau, only the central
service core is fixed with the remainder left open as
in a speculative office. The scheme is designed to a
900mm module, allowing WCs and bathrooms (900,
arq .vol 9 .nos 3/4 .2005 theory
292
Jeremy Till and Tatjana Schneider Flexible housing: the means to the end
6
7
6Maison Loucheur
(1928–9) by Le
Corbusier: day plan
(left) and night
plan (right)
7Kleinwohnung
project (1931) by
Carl Fieger: day
plan (left) and
night plan (right)
1800), bedrooms (1800, 2700 and 3600) and living
rooms (2700, 3600 and 4500) to be accommodated.
The architects designed 10 hypothetical layouts, but
in the end none of these was taken up; the
prospective tenants quickly learned to plan on a
squared grid, especially if this exercise was done in
the actual spaces. The modular system extended to
the treatment of the external facades, so that the
exterior appearance was dependent on the tenant’s
chosen internal layout and positioning of panels, a
rare example of architects passing over aesthetic
control to others [5].14
The Järnbrott Experimental Housing is one of the
few flexible housing schemes to have been properly
evaluated after occupation.15 Again it works on the
principle of an open plan space with fixed
bathrooms and kitchens, and tenants being shown
alternative layouts. A survey carried out 10 years after
completion found that the majority of changes had
been made within the smaller units. This finding
supports a common theme from our research
interviews, namely that the provision of more space
in itself provides a certain degree of flexibility, while
smaller units are exactly the ones that require the
most flexibility to be built in.
Hard use
The notion of soft space lends itself in particular to a
participative approach to design, allowing a degree
of tenant control at both design stages and over the
life of the building. In contrast, hard use is use that is
largely determined by the architect. To this extent
hard use is consistent with the typical desire of the
architect to keep control, and it is therefore maybe
not surprising that hard use is associated with some
of the twentieth century’s iconic architects. Thus in
the Maison Loucheur (1928–9) designed by Le
Corbusier, a combination of folding furniture and
sliding walls allows different configurations for day
and night. Le Corbusier, in typically polemical style,
argues that the purchaser is paying for 46m2of space
but through the cleverness of the design is actually
getting 71m2of effective space [6].
This scheme identifies some of the common
features of hard use. First, the way it operates best
where space is at a premium. Second, the use of
moving or folding components. Third, the highly
specific nature of the configurations produced.
Classic examples of hard use such as Lawn Road
apartments (1934) by Wells Coates, the seminal
Schröder Huis (1924) by Rietveld, and
Kleinwohnung project (1931 ) by Carl Fieger,
all use these strategies [7].
There are comparatively few hard use flexible
housing schemes in relation to the continuing
interest in soft use approaches; they are generally
confined to demonstration or one-off projects, and
usually accompanied by a rhetorical stance which
may be at odds with the real needs of the users, for
whom the discipline of moving walls and folding
beds on a daily basis is likely to be over-onerous. The
fate of Lawn Road is instructive here. Originally the
scheme was occupied by a group of young
intellectuals, including Marcel Breuer, László
Moholy-Nagy, and Agatha Christie, who were
sympathetic with the ideology of the project. As they
moved on, the apartments became increasingly
difficult to rent out, and eventually the empty
scheme fell into disrepair. Following its recent
renovation, Lawn Road now houses a mixture of
government ‘keyworkers’ in shared ownership
tenure (nurses, teachers and so on who would
otherwise not be able to afford London house prices)
and young professionals, apparently attracted by the
lifestyle choice that the apartments offered, though
not put off by the very high cost per square metre of
the leasehold apartments.16 The minimal space
standards and discipline associated with hard use
may thus have a future relevance for two groups of
people, one of which has no other choice but accept
small, the other which sees small as a beautiful
lifestyle accessory.
Techn olo gy
If one method of achieving flexibility in housing is
through a consideration of use through design in
plan, another is through the deployment of
technology.17 Clearly these two approaches are not
mutually exclusive. We have seen how the
development of long span structures allowed the
elimination of loadbearing partitions, which in turn
allows soft use. However, in other schemes it is the
chosen technology rather than the specifics of the
plan that is seen as the primary means of achieving
flexibility. Technology here encompasses
construction techniques, structural solutions, and
servicing strategies, or a combination of these
approaches. Again we have divided these approaches
into ‘hard’ and ‘soft’.
Hard technology
By hard technologies we mean those technologies
that are developed specifically to achieve flexibility,
and which are the determining feature of the
scheme. Of all these, the approach that has been most
systematically developed is that of the open building
movement. This grew out of John Habraken’s research
and his book Supports: an alternative to mass housing.
The book is based around a critique of mass housing,
arguing that mass housing ‘reduces the dwelling to a
consumer article and the dweller to consumer’.18
Habraken argues for an alternative in which the user
is given control over the processes of dwelling. The
bold move in the book is then to harness this social
programme to a technical solution. The basic
principles are straightforward, namely that housing
should be considered as a structure of supports and
infills. The supports provide the basic infrastructure
and are designed as a long-life permanent base. The
infills are shorter life, user determined and
adaptable. The support and infill approach also
implies different levels of involvement on the part of
the user and professional, with professionals
relinquishing complete control, particularly at the
level of the infills. So far so good. Reading the sections
on supports forty years after they were first written,
one is struck by their polemical and open-ended
nature; they should be seen as a challenge to
theory arq .vol 9 .nos 3/4 .2005 293
Flexible housing: the means to the end Jeremy Till and Tatjana Schneider
normative ways of thinking, but have instead been
taken on as a more determinist orthodoxy under the
flag of open building.
In the few examples of open building that have
actually been constructed, the emphasis has shifted
to the technical and constructional aspects and away
from the more socially grounded implications of
flexible housing. So it is not surprising that the
initial interest in open building in the early 1970s
waned because of the dearth of available technical
solutions such as suitable infill systems.19 One of the
few UK projects constructed at the time (the so-called
PSSHAK scheme at Adelaide Road, 1979) has hardly
been changed over the intervening years despite the
explicit flexible intent of the constructional
technique; the main reason is apparently that the
instructions as to how the infill kit could be adapted
were not passed on to subsequent tenants.20
The main outlet for open building has been in
Japan where the Ministry of Construction has
funded a number of experimental projects, most of
which have been driven by a technically determined
agenda.21 This is not to deny that the essential
principles of open building do address the needs of
flexible housing. However, there is a danger in these
and other open building projects of getting obsessed
by the techniques, and in this the technology
becomes an end in itself rather than a background
means to an end. This determinist aspect of hard
technology22 is not, of course, limited to the field of
flexible housing, but can be traced in the course of
twentieth-century attitudes to technology.23
Soft technology
Architects are notoriously susceptible to siren calls of
technology as a means (but quickly becoming the
ends) of denoting presumed progress. The
foregrounding of hard technology allows these
delusions to be perpetuated. This suggests that in
flexible housing, as in other architecture, one
should move from the determinism of hard
technology to the enabling background of soft
technology. Soft technology is the stuff that enables
flexible housing to unfold in a manner not
completely controlled by the foreground of
construction techniques. In flexible housing this
approach can be seen in a number of schemes, many
of which exploit the layering principles of open
building, but in a more relaxed and less determinist
manner. So in the Genter Strasse scheme in Munich
(1972) designed by Otto Steidle with Doris and Ralph
Thut, a prefabricated frame can be filled according to
users’ needs and wants.24 Over the last 30 years,
volume, interiors, and uses have changed
considerably [8].
This scheme, and others like it, exploits soft
technology in the form of a structural system that
allows changes to be made at a future date. This
system may be in the form of an expressed frame, as
in the Genter Strasse scheme, or else simply a grid
structure that does away with the need for
loadbearing internal partitions, as in the
Brandhöfchen scheme, Frankfurt (1995) designed by
Rüdiger Kramm in which the structure’s only
loadbearing elements are beams and columns; none
of the internal walls is loadbearing, meaning that
even party walls can be removed to combine two
smaller units into one large unit. Small service cores
are located on each grid line (at the north facade),
allowing for a range of connection possibilities.
The provision of services, as in the Brandhöfchen
scheme, is another aspect to be considered in the soft
technology approach. This can be achieved in three
ways. First, through the strategic placing of service
cores to allow kitchens and bathrooms to be placed
within specific zones but not to be permanently
fixed. Second, through the ability to access services
so that they can be updated at a later date. Third,
through the distribution of services across the
floorplate so that they can be accessed in any plan
arrangement. The way that most housing is serviced,
and in particular the wiring, remains an obstacle to
flexible housing. The incorporation of electrical
outlets into internal partitions is often necessary to
meet standards of provision or disability
requirements, but this also limits future change. The
distribution of services through raised floors
following office design principles gets over the first
but not the second of these problems; it is also
potentially more expensive. An alternative solution
to the servicing problem is the ‘living wall’ concept
developed by PCKO Architects, in which services are
concentrated along a linear service wall, which is
easily accessed from the outside (to allow service
arq .vol 9 .nos 3/4 .2005 theory
294
Jeremy Till and Tatjana Schneider Flexible housing: the means to the end
8
9
8Genter Strasse,
Munich (1972) by
Otto Steidle with
Doris and Ralph
Thut; frame and infill
9Diagoon Houses,
Delft (1971) by
Herman
Hertzberger.
providers to make changes with minimal disruption)
as well as internally (when internal layouts are
changed).
Conclusion
It will be seen from the above that our sympathies lie
with a ‘soft’ approach to the design of flexible
housing. As noted, soft use and soft technology are by
no means mutually exclusive, and in the best
examples the social and technical aspects of the
project support each other, as in the Diagoon
Houses, Delft (19 71) by Herman Hertzberger. The
principle behind the houses is based on the idea of
the ‘incomplete building’; meaning that a basic
frame leaves space for the personalised
interpretation of the user, i.e. number of rooms,
positioning, functional uses. The occupants
themselves will be able to decide how to divide the
space and live in it, where they will sleep and where
they will eat. If the composition of the family
changes, the house can be adjusted, and to a certain
extent enlarged. The structural skeleton is a half-
product which everyone can complete according to
his own needs. The interiors show a fine balance
between order and chaos, the architecture accepting
but not getting overwhelmed by the vicissitudes and
changes of everyday life [9].
While most of the examples given have been at the
scale of the individual apartment, it is important to
note that our approach to flexible housing can and
should be applied at all scales. At the scale of the
whole building, soft use and/or technology will allow
both a variety of sizes in housing provision, as well as
the potential for mixed use. At the scale of the room,
flexibility can be considered through both use and
technology, the former in the consideration of how a
single room may be used for a variety of purposes,
the latter through the incorporation of flexible
components. At whatever scale, it is clear that
flexible housing can be achieved through a careful
consideration of use and technology and without
significant, if any, additional cost; it does not rely on
an overt display of formal or technological
gymnastics. In this, the soft approach to flexible
housing assumes a background role which, while
not necessarily according to the tendencies of
progressive display that architecture often adopts,
weaves itself into the heart of social empowerment.
theory arq .vol 9 .nos 3/4 .2005 295
Flexible housing: the means to the end Jeremy Till and Tatjana Schneider
Notes
1. This paper is based on a research
project, ‘The Past, Present and
Future of Flexible Housing’,
funded by the Arts and
Humanities Research Council.
2. Caitriona Carroll, Julie Cowans
and David Darton, Meeting Part M
and Designing Lifetime Homes (York:
Joseph Rowntree Foundations,
1999).
Julie Brewerton and David Darton,
Designing Lifetime Homes (York:
Joseph Rowntree Foundations,
1997).
Christian Woetmann Nielsen and
Ivor Ambrose, ‘Lifetime Adaptable
Housing in Europe’, in Technology
and Disability 10, pt. 1(1999), 11–20.
3. Stephen Kendall and Jonathan
Teicher, Residential Open Building
(London and New York: E & FN
Spon, 2000).
4. Stefan Muthesius, The English
Terraced House (New Haven: Yale
University Press, 1982).
Michael Thompson, Rubbish Theory:
The Creation and Destruction of Value
(Oxford: Oxford University Press,
1979).
5. Lionel Duroy, ‘Le Quartier
Nemausus’, in Architecture
d’aujourd’hui 252 (1987) 2-10.
Jacques Lucan, ‘Nemausus 1,
Wohnüberbauung mit Lofts,
Nîmes, 1985’, in werk, bauen +
wohnen 77/44 pt. 3(1990), 56–58.
6. Caruso St. John, ‘Terraced House’,
in a+t 13 (1999), 44–47.
7. Bea Sennewald, ‘Flexibility by
Design’, in Architecture (AIA) 76, pt. 4
(1987) 89–94 (p. 90).
8. John Habraken, Supports: an
alternative to mass housing (London:
Architectural Press, 1972).
9. Rem Koolhaas and Bruce Mau, S, M,
L, XL (Rotterdam: 010 Publishers,
1995), p. 325.
10. Hugo Priemus, ‘Flexible housing:
fundamentals and background’,
in Open House International 18, pt. 4
(1993), 19–26.
Alexander Henz and Hannes Henz,
Anpassbare Wohnungen (Zürich: ETH
Wohnforum, 1997).
Stephen Kendall and Jonathan
Teicher, Residential Open Building
(London and New York: E & FN
Spon, 2000).
11. The term ‘soft’ is drawn from
Jonathan Raban’s seminal book on
urban culture, Soft City: ‘… the city
goes soft; it awaits the imprint of
an identity. For better or worse, it
invites you to remake it, to
consolidate it into a shape you can
live in’.
Jonathan Raban, Soft City (London:
Hamilton, 1974), p. 11.
12. Paul Oliver, Dwellings: The Vernacular
House Worldwide (London: Phaidon,
2003), pp. 166–7.
13. Rostislav Svácha, The Architecture of
New Prague 1895–1945 (Cambridge
MA: MIT Press, 1995), pp. 52–4.
14. Harald Deilmann, Jörg C.
Kirschenmann and Herbert
Pfeiffer, Wohnungsbau. The Dwelling.
L’habitat. (Stuttgart: Karl Krämer
Verlag, 1973), pp. 37–8.
15. Statens institut for
byggnadsforskning, Flexible Flats:
an investigation in an experimental
block of flats in Järnbrott, Gothenburg
(Stockholm: Statens institut for
byggnadsforskning, 1966).
16. Richard Carr, ‘Lawn Road Flats’,
(2004), from http://www.studio-
international.co.uk/architecture/
lawn_road_flats_7_6_04.htm
[accessed May 2005]
17. Alice Friedman makes it clear that
the divisible open plan design was
conceived in close collaboration
with the client for the house,
Truus Schröder, and so not the
whim of the architect as
controller.
Alice Friedman, Women and the
Making of the Modern House(New
York: Abrams, 1998), p. 71.
18. Habraken, p. 11 .
19. Andrew Rabeneck, David Sheppard
and Peter Town, ‘Housing
flexibility?’ Architectural Design43,
pt. 11 (1973), 698–727 (p. 701).
20. Rebecca Pike and Christopher
Powell, ‘Housing Flexibility
Revisited’ in MADE 1(2004), 64–71
(p. 66).
21. Seiji Sawada and John Habraken,
‘Experimental Apartment
Building, Osaka, Japan’, in Domus
819 (1999), 18 –25 (p. 21).
S. Kendall and J. Teicher, Residential
Open Building (London and New
York: E & FN Spon, 2000), pp. 24–5.
22. Open building is by no means
the only example of ‘hard
technology’ in the flexible
housing sphere. Others include
the Danish Flexibo system and the
‘Optima’ housing system designed
by Cartwright Pickard for the UK
market. See respectively:
Fællestegnestuen ‘Flexibo,
Amager: Arkitekter:
Fællestegnestuen, with Viggo
Møller-Jensen, Tyge Arnfred and
Jørn Ole Sørensen’, in Arkitektur
DK 23, pt. 6(1979), 232–9and D.
Littlefield, ‘DIY for Architects’, in
Building Design 1643 (2004) 20–1.
23. It should be noted that a few
schemes also consider the section
as means of allowing flexible use
over time. For example, the
unbuilt Projekt Wohnhaus (1984)
by Anton Schweighofer is based
on an unfinished double-height
space which the occupants would
‘grow’ into through the addition
of floors and walls. Anton
Schweighofer, ‘A central space
as a “kitchen studio”: changeable
flats: projects and buildings;
Architects: Anton Schweighofer’,
in werk, bauen + wohnen 71/38, pt. 3
(1984), 28-35.
24. Otto Steidle, ‘Anpassbarer
Wohnungsbau’, Baumeister 74 , pt.
12 (1977), 1163–6.
Illustration credits
arq gratefully acknowledges:
Authors, 1–2, 4–7
ADP Architektur und Planung,
Zürich, 3
Steidle Architekten, Munich, 8
Herman Hertzberger, 9
Biographies
Jeremy Till is an architect and
Professor and Director of Architecture
at the University of Sheffield where he
has established an international
reputation in educational theory and
practice. His widely published written
work includes Architecture and
Participation and the forthcoming
Architecture and Contingency. He is a
Director in Sarah Wigglesworth
Architects, Chair of RIBA Awards Group
and will represent Britain at the 2006
Venice Architecture Biennale.
Tatjana Schneider studied
architecture in Kaiserslautern,
Germany and Glasgow, Scotland. She
practised in Munich and Glasgow, and
researched a PhD at Strathclyde
University. Now a Research Associate
at Sheffield University, she is co-
writing with Professor Jeremy Till a
book on Flexible Housing for
publication by Architectural Press.
Authors’ address
Professor Jeremy Till
Dr Tatjana Schneider
School of Architecture
University of Sheffield
Western Bank
Sheffield
s10 2tn
uk
j.till@sheffield.ac.uk
t.schneider@sheffield.ac.uk
arq .vol 9 .nos 3/4 .2005 theory
296
Jeremy Till and Tatjana Schneider Flexible housing: the means to the end
