Small House Projects in Japan:
Housing Experiments for Sustainability and Open Building Concept
Nobuhisa MOTOOKA, Associate Professor, Sugiyama Jogakuen University
Shin MURAKAMI, Associate Professor, Sugiyama Jogakuen University
Eiichi TOBE, Professor, Sugiyama Jogakuen University
Chikusaku Hoshigaoka Motomachi 17-3,
Nagoya, 464-8662, JAPAN
[email protected]Abstract
Focusing on minimum or small houses has become one of the recent trends in housing
design in Japan, as has been observed in many house design works. Periodical
coverage can tell that the number of such works has clearly been increasing since the
90s, as compared with the 70s and 80s. The trend of small houses was also observed
in the 50s. In those postwar years of economic growth, it was driven by the conditions
of the time, such as supply and housing shortages and urban centralization. Today’s
social conditions are significantly different from those in the 50s, and naturally, the
whole concept of small houses has greatly changed from the past.
In this research, we evaluate the experiments of small houses, from the view of the
idea of sustainability and open building concept. Specifically, the study compares the
small houses of the 50s and those after 1990 to examine their differences or
similarities in terms of size, structure and building systems. And thus clarify how
industrialization and standardization reflect on these experiments.
The former period, most were constructed on wood, with traditional construction
method. As Japanese traditional structural system contains sustainability and
flexibility originally, the purpose of design was rather how to adapt the
industrialization to the traditional construction and how to realize the modern way of
living in the smallest space, than fulfillment of flexibility. Moreover, low cost was
also included in the design purpose.
In latest examples, not only wooden but various structural systems are seen, such as
steal, reinforced concrete or mixed structure. Also multi-storey houses are major. It
cannot be said that those houses are cheaper than “non-small” houses. The total floor
areas are larger than those of the 50s, and also the floor area ratios are. The “small”
means “small building area” rather than “small space for life and minimal cost for
construction”. The experimental projects were conducted by the intention exploring
new possibilities and diversities of space design, with various highly industrialized
materials. The small houses after 1990 can be regarded as experimental efforts to
explore new approaches to skeletons within the context of urban tissue.
Key words: small house, experimental house, floor area, building area, Japan
1
1. Purpose
Focusing on minimum or small houses has become one of the recent trends in housing
design in Japan, as has been observed in many house design works. Periodical
coverage can tell that the number of such works has clearly been increasing since the
90s, as compared with the 70s and 80s. The trend of small houses was also observed
in the 50s. In those postwar years of economic growth, it was driven by the conditions
of the time, such as supply and housing shortages and urban centralization. Today’s
social conditions are significantly different from those in the 50s, and naturally, the
whole concept of small houses has greatly changed from the past.
Given this background, this study looks into what approaches are adopted for small
houses, evaluating the design efforts made for this housing type, particularly in terms
of building systems such as prefabrication and open building concepts. Specifically,
the study compares the small houses of the 50s and those after 1990 to examine their
differences or similarities in terms of size, structure and building systems. In so doing,
it attempts to shed light on the changes in the notion of architects towards the
relationship between small house design efforts and prefabrication or open building
systems.
Figure 1 Number of small houses
2. Study objects and method
The house works studied here were cited from representative architectural periodicals, namely
Shinkenchiku (Japan Architect, hereinafter “JA”), Shinkenchiku Jutaku-Tokushu
(hereinafter, “JT”) and Kenchiku Bunka (hereinafter, “KB”)1.
Among the works covered by these periodicals, this study dealt with those that were
described in such expressions as “a small house” or “a small lot house” (hereinafter
collectively called “small houses”). The “small houses” studied here were those that
had been designed to represent the architects’ specific notion of “smallness”. The
criteria to define small houses are not based on square measure or cost. Thus, the
issue here is how the concept of architects of the time towards small houses has been
represented in their works. Likewise, the criteria to determine “low-cost houses” or
“small lot houses” are based on whether the architects use expressions that convey
these characters when describing their works.
First, the number of houses described as “small houses” (those actually constructed)
between 1946 and 2003 was counted. Figure 1 shows the number of small houses
covered in each periodical between 1946 and 2003, organized by the year of
construction2. After peaking in the mid 50s, the number of small houses covered in
2
the periodicals was rather small throughout the 70s and 80s, which has been on an
increase in recent years. In other words, the graph indicates that small houses have
experienced a boom twice, namely during the 50s and the recent 10 years.
Addressing these two phases, a comparative study is carried out here. The small
houses studied here are those covered by JA between 1950 and 1959, JT since 1990,
and KB between 1950 and 1959 as well as after 1990. The number of small houses
studied was 163 for the 50s and 197 after 1990.
3. Square measure of small houses
In the 50s, which is sometimes termed as the “age of small house architects”, small
houses were one of the key issues for the architects of the time. During the postwar
reconstruction period, though there was a surging demand for housing construction,
there was also a serious shortage in building materials. Given this, the government
issued a temporary building restriction ordinance in 1946, which restricted the size of
houses to 12 tsubo (1 tsubo =3.954 sq.yds or 3.31 sq.m) for families of five. This
restriction was relaxed to 15 tsubo (49.5m2) in the following 1948 and then was lifted
all together in 1950.
In the same year, the Government Housing Loan Corporation Law was issued to
support the nation to own houses. However, according to the lending contract term,
the square measure had to be between 9 and 30 tsubo to be entitled for loan, of which
up to 18 tsubo only was covered. It is easy to imagine that such detailed loan
conditions served as a guideline for postwar housing design.
Under such circumstances, it can be said that houses of approximately 15 tsubo were
generally regarded as “small houses” in those days. This is also implied in Figure 2,
which shows the building/total floor areas of small houses in the 50s. Also, as
characterized by the fact that the houses having the total floor area of 60m2 or smaller
account for 62 % of all small houses, most small houses had the floor area of between
40 and 60 m2 (Figure 3). Single stories account for 78% of all small houses.
The periodicals were full of small house discussions, not only about floor area but
about planning methods for materializing modern life within such a small space. The
common theme was to what extent a living space could be made small, which was in
other words the question of how to put into practice houses with small total areas.
Figure 2 Relation between total floor area and building area
Figure 3 Number of examples, classified by sizes of total floor area
3
On the other hand, the small house boom in the 90s could be characterized by the
consideration for the relationship between houses and the urban tissue. While site
conditions were not a big issue in the 50s, the size of the site and its conditions have
become a major issue among the small houses in the 90s cited in this study. Of all
small houses built after 1990, 73 % have been designed with the focus on the
smallness of sites (Figure 5). According to the graph of site/total areas (Figure 4),
many of the 50s small houses were characterized by a total floor area of
approximately 50m2 built on relatively large sites, while after 1990, the majority has
been those with a total area of approximately 100m2 built on approximately 70m2
sites. Also, as shown in the graph of building/total floor areas (Figure 2), while many
of the 50s small houses were single stories, recently there have been many high-floor-
ratio houses constructed high up on the small sites. In the periodicals, architects often
describe their works in relation to the urban tissue, elaborating on the problem of how
to dwell on small sites under severe urban conditions.
Figure 4 Relation between total floor area and site area
Figure 5 Ratio of low-cost houses / Ratio of small site area houses
4. Structural materials used for small houses
Figure 6 shows the percentage of each structural system. Approximately 83% of small
houses were made of wood in the 50s. At the same time, 55 % of all small houses
were low-cost houses (Figure 5). Cheaper traditional wooden houses were preferred in
the 50s due to the situations concerning material supply and industrialization. Since
1990 on the other hand, purely wooden houses has accounted for as small as 32%,
while various structural systems have been adopted, such as reinforced concrete, steel
and mixed structures, which account for 21%, 19% and 27%, respectively. It is natural
that more diverse materials and structural systems are available today than the 50s.
What should be noted about the 50s is the fact that two notable works namely “Steel
House Op.1” by Kenji Hirose (1953) and “Seike House” by Kiyoshi Seike (1954),
which are regarded as having great impacts on posterity, were steel and reinforced
concrete structures, respectively. Though rare, experiments on small houses through
application of new structural materials in the 50s indeed proved fruitful.
On the other hand, what is notable in recent years is the fact that cost reduction efforts
are not that prioritized as compared in the 50s when they were the major objective for
the rationalization and standardization of construction work (Figure 5). With regard to
the structural type, wooden structures have been decreased due to the trend towards
4
high-rises and increased fireproof performance. Instead, reinforced concrete and steel
structures have been increasing, though they are more costly than wooden structures.
It seems that utilization of small sites has replaced the construction cost reduction
efforts in trying to reduce the overall cost because of the steep rise in land prices.
Figure 6. Ratio of structural materials
5. Building system experiments on small houses
5.1. Building system experiments in the 50s
As studied above, architects of the 50s attempted to create houses that materialize
modern living out of the minimum budgets, materials and space. In specific terms,
most of the discussions about such attempts focused on layout and cost. At the same
time, there seem to have been various experimental efforts on building systems. Case
studies on building system experiments in the 50s yielded certain insights, which are
categorized into the following 4 subjects:
(1) Planning by the core system
The core system is the method of materializing a small house by integrating specific-
purpose services in one place and leaving other space open. Combining this concept
with structural ideas gave rise to the notion of categorizing not only the living style
but also the structural system itself into fixed and flexible parts3. This concept was
clearly put forward by Kiyoshi Seike and Makoto Abe in their works in 1951. An
integrated flexible space can adapt to short- and long-term changes through use of
subsystems such as furniture and partitions. A house called “Residence No. 20
(1954)” by Yoh Ikebe (Figure 7) is a good example of materializing such a concept in
an extremely small space. In those days, not only Ikebe but also Bunzo Yamaguchi of
the Research Institute of Architecture made the same experiments. The small houses
designed through this approach4 that aimed to modernize services and rationalize
domestic work contained the seeds of open building concepts.
5
(2) Proposal of dimensional systems and modules as part of prefabrication efforts
Most houses of this period were based on the Japanese traditional module, namely the
Ken module. As long as this module was followed, it was possible to deal with any
changes in doors/windows or furniture easily, share craftsmanship and take flexible
measures in terms of both materials and construction. That is to say the traditional
construction method used to ensure open-system like flexibility without having to go
through prefabrication. In the 50s however, revision of the standard dimensional
system gradually attracted attention, driven by the influence of modern architecture
and the demand for prefabrication.
Studies on the dimensional proposals often made by architects when describing their
works reveal that there were two attitudes towards this issue: to propose a new
dimensional system suitable for modern living and prefabrication; and to propose new
ways of applying dimensional system for the purpose of making the most of available
materials in the marketplace. The houses that represented the former attitude were
minority, while most of the houses were based on the latter position.
A house called “Minimum-Volume House (1950)” by Yoh Ikebe was one of the
examples of the former, which proposed the 0.75-m, 1.5-m and 3-m modules.
Although various proposals were made, no unified dimensional system was
established. There was another aspect that was supposed to be as important as
dimensional systems in contributing to prefabrication and the open-building concept,
namely the specifications of joints. In his series of steel houses typified by “Steel
House Op.1 (1954)”, Kenji Hirose made various experimental efforts to explore the
possibility of prefabricated houses, combined with joint detail studies. However, as in
the case of dimensional systems, the majority of experimental works did not go
beyond the traditional building systems.
The most pressing issue for the architects of the time was how to prefabricate the
materials used in conventional building systems. There were very few works that tried
to go beyond the traditional construction regime, aiming to develop new dimensional
systems or propose prefabrication of new materials.
(3) Prefabrication of materials
The dry construction method and panel wall systems were observed in “House of
Prof. Dr. Otto Mori (1951)” and “House & Atelier of Mr. T (1952)”5. In “House of
Mr. K (1953)” by Han Konishi, exterior curtain walls were applied to a reinforced
concrete structure, while Kenji Hirose experimented with wall building systems in a
series of his steel house works. Experimental use of panels and curtain walls, whose
purposes were the prefabrication of materials and simplification of construction
process by separating walls from the structural system, had already been observed in
the early 50s.
As cited above, there were many works that rendered new ideas to promote the
prefabrication of materials. “Residence for Mr. Kimoto (1958)” by Isao Shibaoka was
designed with many prefabricated materials. According to the architect, it actually
cost more than the traditional building system because of the lack of expertise by the
6
workers. In those days, the experimental use of new materials and prefabrication of
materials were concurrently in progress, and even if a new material was found
suitable for prefabrication, it did not mean to be low cost due to the immature
prefabrication procedures.
(4) Notion of extension
Small houses gave rise to the notion of extension, i.e. to build a small house for a start
and then deal with future changes by extension work. Some of the examples that
entailed the notion of extension include “Minimum House in Musashino (1952)”,
“Architect’s House (1953)” and “House in Kamakura (1955)”6. The notion of
extending a house to deal with future lifestyle change was developed into an idea
related to the core system and structural considerations, which was to divide a house
into fixed and flexible parts and accommodate future changes with the flexible part.
The notion of extension was born out of the intention of seeking the minimum space,
which later develop the concept of small houses into sustainability and adaptability.
Throughout the 50s, there was a lack of urban-oriented viewpoint or forward-looking
ideas about sites. The common concerns were the search for the minimum layout to
live modern life and the reduction of construction cost. The aforementioned
experiments on building systems were derived from these concerns too.
5.2 Building system experiments after 1990
For the past 10 years, the term “small houses” has been used to express “houses built
on small sites” in many cases. Diverse building system experiments have been carried
out, which can be summarized as attempts to propose housing design that represents
the way it should be to live in the current urban environment.
In specific terms, building system experiments have shifted their focus from the
prefabrication of materials in the 50s to the development of completely new ways of
applying existing prefabricated materials in recent years. Below are some of the
examples of diverse building system experiments.
(1) Prefabricated units for unusual houses
Houses on complicatedly shaped small urban sites under severe legislative restrictions
often have site-specific forms. In addition, because it is impossible to expand the
space area-wise, these houses are often built high up above the ground. Considering
these, it seems natural that there are many small houses that display architects’
various spatial ingenuities.
Under these circumstances, there are very few houses that were intended to seek a
prototype or mass production for universal applications. To stretch a point, there have
been a few works that are considered as prefabrication efforts to some extent. A house
called “Y Studio (1995)” by Yasutaka Yamasaki was one such example, in which the
“framework unit method” was experimentally adopted (Figure 8). This house is quite
similar to a steel rigid frame prefabricated house.
7
However, it should be noted that the method was developed as a solution to the
question of how to deal with unusual urban conditions, such as the highly dense urban
residential environment (the units can be stuck to make up to 4 stories) and the
aftermath of the Hanshin-Awaji Earthquake (the units can be built as a temporary
building to be transformed into a permanent home).
Another example is “Natural Shelter (1999)” by Masaki Endo and Masahiro Ikeda,
which is covered with unitized curved panel shelters. These panel shelters are
prefabricated units, though they were not intended for mass production. It was an
attempt to create an unusual house using prefabrication-oriented units to meet the
requirements imposed by the specific site conditions and client’s lifestyle.
(2) New applications of existing prefabricated materials and new structural ideas
In his “Mo House (2003)” and “Ni House (2003)”, Hiroaki Kimura experimentally
adopted 9-mm thick steel sheets to cover exterior walls. Mitsuhiko Satoh used steel
pipe piles for part of a spiral staircase and structural members in one of his works,
“House in Oshima”. As cited here, there are cases where existing prefabricated
materials, such as steel plates, corrugated pipes and steel pipe piles, are applied in
unconventional manners.
Furthermore, in terms of structural systems, there are many cases where attempts have
been made to depart from the usual wall or rigid systems, trying to seek new ways of
applying prefabricated materials. “House in Koutou (2003)” by Mitsuhiko Satoh has a
steel skeleton system combined with ribbed steel plates (a steel version of 2 x 4
system, so to speak).
Because finished panels that serve as the structural units were installed from within, it
was possible to build the structure to the absolute edge of the site boundary. “House of
Shelves (2000)” by Koichi Maeda and “Wood Box” by Kiyoshi Kasai are wooden
frame structures, which proposed the utilization of structural system, as shelves for
example, in everyday life. Making walls thin, allowing the structure to stretch to the
absolute edge of the site boundary and utilizing walls for unusual purposes are all
considered to be aimed at making the most of the small site.
(3) Low cost house experiments
Though not so popular as in the 50s, there are some examples of seeking solutions to
smallness and low cost. “Minimum House (2002)” by Yasuhiro Yamashita and "Wood
Box – Tokura” by Kiyoshi Kasai are some of such low cost examples. All of them
were intended to seek the way to build a house at low cost, which is similar to seeking
a particular solution in other words, like seeking the way to eliminate unnecessary
planning factors under specific family and site conditions. The low cost efforts here
are made, not only in terms of the prefabrication of materials but also in terms of the
client’s lifestyle, contract style and building system.
Since 1990, mass production and prefabrication have not really been shared concerns
in dealing with small houses, though there are few but some examples. The common
concept is to propose a new lifestyle in cities. The diverse structural experiments
described above can be regarded as architects’ endeavor on the subject of living in the
8
current urban environment.
Figure 7. “Residence No. 20 (1954)”
by Yoh Ikebe (ext. JA , 1954.11, p. 47.)
Figure 8. “Framework unit method of
Y Studio” by Yasutaka Yamasaki (ext.
JT, 1995.9, p. 119.)
6. Relationship between architect-designed small houses and prefabricated
houses
In this section, the relationship between architect-designed small houses and
prefabricated houses is reviewed. The important issues involved in introducing the
mass production and open building concept are the standardization of dimensional
system and joints, proven construction performance, maturity of prefabrication
procedures including material supply, and corporate strategies that help establish an
industry. It is no doubt that dimensional systems, joint specifications and construction
performance were all fully explored by the small houses of the 50s, and the ideas
generated through such efforts were successfully realized in prefabricated houses later
years.
With regard to lifestyle, the design themes pursued by the 50s architects, such as the
division of dining and bed rooms, reduction of housewives’ domestic labor and
entrenchment of the notion of sitting room, were fully expressed in the prefabricated
houses of the 60s7. Small houses could not be unrelated to prefabrication in these
days, and all architects had prefabrication in their mind. This awareness was the drive
for the revision of dimensional systems, development of the panel and core systems
and notion of extension, which were all reflected in the design of prefabricated
houses.8. The small houses in the 50s advanced together with prefabrication; they
complemented each other.
On the other hand, most architects were not interested in the aforementioned
9
corporate strategies. As Shuichi Matsumura pointed out, there were huge differences
in the standpoint taken by architects and prefabricated housing manufacturers, though
they might produce houses with similar forms. The differences lie in whether there
were entrepreneur spirits and ability to lead or not9.
As a result, architects and prefabricated housing manufacturers took separate
directions. While prefabricated housing manufacturers, who supply a large number of
reasonable and high-quality “ordinary houses”, have established themselves as an
industry, architects are studying how individual small houses are related to the urban
tissue. Today’s small houses are diversified. Though there is the common awareness
of the problem of living in the urban environment, small houses continue to be
individual and unique as it is not possible to yield a single common solution.
7. Concluding remarks
This study brought the insights described below.
First, the small houses of the 50s were found to have the following characteristics:
- Most of them were wooden structures and many were single stories.
- The aim was to reduce cost, as well as the realization of richness in lifestyle in a
small space.
- The idea of integrated space was generated in trying to put everything in a small
area. The major issue in cost reduction was the development of rational building
systems.
With the shortage of housing supply as well as material supply, the architects of the
time aimed to create richness in lifestyle with the minimum budget, materials and
space10. The specific objectives were integrated layouts, modernization of services,
reduction of construction cost and rationalization of production. Many experimental
efforts were made for prefabrication and mass production, which led to the
development of prefabricated houses in the 60s and onwards. However, there was
little awareness of the site issue.
On the other hand, the following insights were gained with regard to the small houses
built after 1990:
- In addition to wooden structures, various structural types appeared, including
steel and reinforced concrete structures.
- Occasionally, there are efforts to seek new ways of applying existing
prefabricated materials or put in practice new structural ideas. Cost reduction is
not necessarily at issue.
- As compared with the 50s, the total floor area has become larger and
building/site areas smaller. There is a heightened awareness of building houses
on small sites, namely living in the crowded urban environment.
While prefabrication of traditional building systems was the main issue in the 50s,
recently there have been many examples where existing prefabricated materials were
used in completely new manners, exercising more creativity in design at the same
time. The search for the minimum possible space, exploring the question of to what
extent a dwelling space can be made small in the 50s, has not been observed since the
90s. At issue instead has been the creation of as richer space as possible under
10
restrictions of small urban sites.
While small houses in the 50s took the form of cost reduction experiments through
prefabrication of the traditional open system based on the Ken module, those in the
90s can be regarded as experimental efforts to explore new approaches to skeletons
within the context of urban tissue.
1
Shinkenchiku (JA) and Kenchiku Bunka (KB) have been published since the end of
World War II to this day, and are representative architectural periodicals of Japan. JA
was first published in 1925. Though it suspended publication during the war, it
restarted in 1946 and has been issued since. Shinkenchiku Jutaku-Tokushu (JT) was
first published in 1985 as the quarterly special issues of Shinkenchiku specializing in
residential buildings, and has been published monthly since 1986. The publisher of JA
and JT is Tokyo-based Shinkenchiku-sha. KB was first published in 1946. Though
still in publication, it has been bimonthly since February 2001. The publisher is
Tokyo-based Shokokusha Publishing.)
The houses studied here were all cited from architectural periodicals, which might
receive criticism of subject validity for dealing with fashionable aspects rather than
the real-world situations. However, it is a fact that these periodicals picked up many
houses that were considered suitable as the object of small house studies, and the
significance of studying periodicals lies here.
On the other hand, there is a problem of using periodicals as research source because
the works were originally selected according to editors’ subjective judgments. This
problem should be solved through research on many periodicals. This study is
intended to take the first step by analyzing JA and KB; the authors are planning to
study other periodicals to elaborate the research.
2
With regard to JA, it should be noted that JT came out as a periodical specializing in
dwellings in 1985. That is why housing coverage showed a notable increase after
1985, which accordingly increased the number of small houses. Still, it is observed
that small houses increased their presence after 1994.
3
Kazumasa Yamashita. (1980). “Wavering housing design” in “Thirty-five postwar
history of architecture” (in Japanese), Shinkenchiku (Japan Architect), June 1980
Extra edition, p. 98.
4
Some other good examples of the core plan include “Residence No.15” and
“Residence No. 21) by Yoh Ikebe, “Residence for Mr. Okino” by Bunzo Yamaguchi
and House with Horse Chestnut (1954)”.
“House of prof. Dr. Otto MORI (1951)” by Kiyoshi Seike and “House & Atelier of
Mr. T (1952)” by the Research Institute of Architecture.
6
Other examples are: “Minimum House Suits Surroundings (1952)” by Shunichi,
Katsuyama, “Architects House (1953)” by Keiichi Kondo and “House in Kamakura
(1955)” by the Research Institute of Architecture.
11
7
Shoichi Ikegami. (1961) “Small houses today – Ready-made and custom-made” (in
Japanese), Shinkenchiku (Japan Architect), Vol. 36, No. 1, p. 116.
8
With regard to one of the masterpiece small houses of the 50s, namely “Experimental
Minimum House”, the architect Makoto Masuzawa clearly stated that he had aimed to
materialize mass production in his descriptions of the work in the periodical. He
claimed that the house “would allow materialization without being affected too much
by the conditions of a given site”. It is quite telling that a contemporary version of this
house was actually put on the market as “9-tsubo House” in 2002. The differences in
the attitude towards small houses between the 50s and today are most clearly
demonstrated by the fact that the original name “Minimum House” was changed to
“9-tsubo House”, which stresses the building area.
9
Shuichi Matsumura. (1999). “A concept called ‘House’ – genealogy of houses in the
20th century” (in Japanese), University of Tokyo Press, p. 124.
10
Kazuko Kinoshita. (2001). “Postwar small houses of Japan” (in Japanese),
Kenchiku Bunka, December 2001, pp. 121 – 123.
12
