Singapore - Acknowledgments and Preface
Singapore
The editor and authors are grateful to numerous individuals in the
international community, in various agencies of the United States
government, and in private organizations who gave of their time,
research materials, and special knowledge to provide data and
perspective for this study. Especially appreciated are the helpful
suggestions and economic insights of Edward Chesky and Morris Crawford
and the generous assistance of the staff of the Embassy of Singapore,
Washington, D.C.
The editor and authors also wish to express their appreciation to
staff members of the Federal Research Division, Library of Congress,
whose high standards and dedication helped shape this volume. These
include Martha E. Hopkins, who managed editing; Marilyn L. Majeska, who
reviewed editing and managed book production; and Barbara Edgerton and
Izella Watson, who did word processing. David P. Cabitto and Sandra K.
Ferrell prepared the maps and other graphics for the book, Carolina E.
Forrester reviewed the maps, and Arvies J. Staton contributed to the
charts on military ranks and insignia. Special thanks go to Kimberly A.
Lord, who designed the illustrations for the cover of the volume and the
title pages of the chapters, and Donald R. DeGlopper, who assisted with
many of the editorial duties.
The following individuals are gratefully acknowledged as well: Shari
Villarosa of the Department of State for reviewing all the chapters;
Mimi Cantwell for editing the chapters; Cissie Coy for the final
prepublication editorial review; Joan Cook for preparing the index; and
Linda Peterson of the Printing and Processing Section, Library of
Congress, for phototypesetting, under the direction of Peggy Pixley. The
inclusion of photographs in the book was made possible by the generosity
of private individuals and public agencies, especially Ong Tien Kwan of
Kuala Lumpur and Chiang Yin-Pheng and Joyce Tan of the Singapore
Ministry of Communications and Information.
Finally, the editor and authors wish to thank Federal Research
Division staff members Andrea M. Savada, Sandra W. Meditz, and Richard
Nyrop for reviewing all or parts of the manuscript, and Robert L. Worden
for assisting in the final stages of editing the completed manuscript.
The first edition of the Area Handbook for Singapore was
published in 1977. Prior to that edition, Singapore was included in the Area
Handbook for Malaysia and Singapore, which was published in 1965
just before Singapore became a separate, independent nation. The current
volume, a complete revision of the 1977 edition, covers a period of
remarkable economic growth and political stability for a nation in
existence for only a quarter century. During the 1977-89 period,
Singapore moved assuredly into the category of newly industrializing
economy, and its reknowned port grew from being fourth in the world in
terms of volume of shipping to being the world's busiest port.
Singapore: A Country Study is an attempt to present an
objective and concise account of the dominant social, economic,
political, and national security concerns of contemporary Singapore
within an historical framework. The volume represents the combined
efforts of a multidisciplinary team, which used as its sources a variety
of scholarly monographs and journals, official reports of government and
international organizations, and foreign and domestic newspapers and
periodicals.
The authors have limited the use of foreign and technical terms,
which are defined when they first appear in the study. Spellings of
Singaporean personal names used in the study conform to standard
Singaporean usage, and contemporary place names are generally those
approved by the United States Board on Geographic Names. All
measurements are given in the metric system.
Singapore
Singapore - Introduction
Singapore
The world's busiest port, the modern nation of the Republic of
Singapore, was founded as a British trading post on the Strait of
Malacca in 1819. Singapore's location on the major sea route between
India and China, its excellent harbor, and the free trade status
conferred on it by its visionary founder, Sir Thomas Stamford Raffles,
made the port an overnight success. By 1990 the multiethnic population
attracted to the island had grown from a few thousand to 2.6 million
Singaporeans, frequently referred to by Prime Minister Lee Kuan Yew as
his nation's greatest resource. If Raffles had set the tone for the
island's early success, Lee had safeguarded the founder's vision through
the first quarter-century of Singapore's existence as an independent
nation, providing the leadership that turned it into a global city that
offered trading and financial services to the region and to the world.
Modern Singapore would be scarcely recognizable to Raffles, who
established his trading center on an island covered with tropical
forests and ringed with mangrove swamps. Towering skyscrapers replace
the colonial town he designed, and modern expressways cover the tracks
of bullock carts that once led from the harbor to the commercial
district and the countryside beyond. Hills have been leveled, swamps
filled, and the island itself expanded in size through extensive land
reclamation projects. Offshore islands are used for recreation parks,
oil refineries, and military training bases. Despite the scarcity of
land for real estate, the government has worked to maintain and expand
the island's parks, gardens, and other green spaces. By housing 88
percent of its population in mostly multistoried public housing,
Singapore has kept a rein on suburban sprawl. In Raffles's town plan,
separate areas were set aside for the various ethnic groups of the time:
Malays, Chinese, Arabs, Bugis, and Europeans. Government resettlement
programs begun in the 1960s broke up the former ethnic enclaves by
requiring that the public housing projects--called housing estates--that
replaced them reflect the ethnic composition of the country as a whole.
As a result, modern Singapore's three main ethnic groups--Chinese,
Malays, and Indians--live next door to each other and share the same
housing development facilities, shops, and transportation.
Despite efforts to maintain an ethnic balance in housing, however,
the stated goal of the nation's leaders is not that Singapore become a
mini-melting pot, but, rather, a multiethnic society. Of the country's
2.6 million inhabitants, about 76 percent are Chinese, 15 percent Malay,
6.5 percent Indian, and 2.5 percent other. There are, however, mixtures
within this mixture. The designation Chinese lumps together speakers of
more than five mutually unintelligible dialects; Singaporean Malays
trace their forebears to all of the major islands of the Indonesian
archipelago, as well as to the Malay Peninsula; and the ancestral homes
of Indians include what are the modern states of India, Pakistan,
Bangladesh, and Sri Lanka. Out of this diversity, the government
leadership has attempted to establish a what is calls "Singaporean
identity," which would include certain unifying and modernizing
elements but yet retain essential variations, based on Asian culture and
values. One of the unifying factors is the English language, selected as
the medium for educational instruction both because of its neutrality in
the eyes of the three dominant ethnic groups and because of its position
as the international language of business, science, and technology. In
order not to lose touch with their Asian heritage, however, Singaporean
school children are also required to study an appropriate "mother
tongue," designated by the government as either Malay, Tamil, or
Mandarin Chinese--a vast oversimplification of the polyglot of
Singaporean native languages.
Singaporean identity, as envisioned by the country's leadership,
calls for rugged individualism with an emphasis on excellence; the
government constantly exhorts its citizens to be the best they can be.
Education, home ownership, and upward mobility are all considered
appropriate goals. Although Singaporeans are expected to be modern in
their outlook, they also are encouraged to retain a core of traditional
Asian values and culture. In a society in which all share a common
education system, public housing, recreation facilities, and military
training, the government considers it important to highlight the
uniqueness of the three official ethnic groups--Chinese, Malays, and
Indians-- through the setting aside of national ethnic holidays and the
sponsorship of ethnic festivals. Singaporean ethnic differences are
usually maintained, however, not so much by these somewhat
self-conscious displays of ethnicity but rather by membership in
ethnically exclusive associations. Usually religious, charitable, or
business in nature, many of these associations had their origins in
colonial Singapore and represent finer distinctions of ethnicity than
those supported by the government. Chinese trade associations, for
example, are usually restricted to speakers of a particular dialect.
Hindu temples are sometimes associated with worshipers who trace their
heritage to a particular region of India.
Singapore is multireligious as well as multiethnic. Major religious
preferences reported in 1988 were Buddhism (28 percent), Christian (19
percent), no religion (17 percent), Islam (16 percent), Daoist (13
percent), and Hindu (5 percent). Singapore's nineteenth-century
immigrants valued the social as well as religious aspects of their
congregations, and their descendants are more likely to concern
themselves with social activities centered around their temples and
mosques than with elaborate ritual or ceremony. The government, although
secular, views religion as a positive force for instilling moral values
in the society. At the same time, it keeps a watchful eye out for social
or political activism within religious groups. Muslim fundamentalists
and over-zealous Christian proselytizers alike are kept under careful
scrutiny, lest they upset the religious and ethnic harmony of the
country.
Singapore closely resembles developed countries in terms of its low
birth rates, high life expectancy (73.8 years at birth), and major
causes of death--heart disease, cancer, and stroke. Although in the
early years of independence the government mounted campaigns to lower
the country's high birth rate, it became concerned in the 1980s when the
rate dropped below the replacement level. Campaigns and incentives were
instituted to encourage those who could afford it to have more than two
children. College-educated women were especially encouraged by
exhortations and incentives to marry and have children.
In terms of public health, Singapore also closely resembles developed
countries. Although some observers criticize the country's modern,
sanitized environment and mourn the loss of the old port's charm, they
probably either have forgotten or never knew the open sewers,
tuberculosis sanatoriums, and opium dens of colonial Singapore. Whereas
the manufacture and sale of opium continued to be a major source of
revenue for the colonial government up until World War II, the
government effectively combats drug use in modern Singapore through
antidrug campaigns, rehabilitation centers, and a mandatory death
penalty for trafficking. The government heavily subsidizes services in
order to make them affordable to all and sets aside 6 percent of the
monthly income of each worker into a personal Medisave account, which
can be used to pay hospitalization costs for any family member. The
Medisave account is part of the Central Provident Fund, which is
Singapore's compulsory national social security savings plan.
Contribution rates due to be phased in in the early 1990s mandate a
contribution of 40 percent of the gross wages of employees under
fifty-five, with employee and employer sharing the burden equally.
Singaporeans can use these funds to invest in approved securities, to
purchase homes in government housing projects, or to pay for
hospitalization and retirement. By 1990 some 88 percent of Singaporeans
lived in Housing and Development Board apartments, a vast public housing
and urban redevelopment project initiated in the early postwar years.
Under the program, which began in earnest after independence,
Singapore's slums and ethnic neighborhoods gradually were replaced with
modern housing estates, self-contained units providing shopping,
restaurant, and recreation facilities as well as apartments of various
sizes, scattered outward from the old central city. A network of
superhighways and a state-of-the-art mass rapid transit system connect
Singapore's housing estates with commercial and industrial areas.
Although Singapore's founder and other nineteenth-century residents
would no longer recognize the island, they would at least be able to
identify with certain aspects of its modern economy. The principle of
free trade laid down by Raffles was still largely in effect in the late
1980s, with only a few revenue tariffs levied on such things as tobacco
and liquor. Trade continued to be the island's lifeblood; in 1988 the
value of Singapore's international trade was triple the total of its
gross domestic product (GDP). Although some aspects of the trade have
changed, others remained the same. The island's initial success resulted
from its role as a conveniently located and duty-free entrep�t for the
three-way trade among China, India, and various parts of the Malay
Archipelago. This trade was an ancient commerce, and trading posts
probably had flourished intermittently at that favored location for two
millenia. In early colonial times, silks from China, manufactures from
Europe, incense from India, and spices from the Moluccas all were
shipped on the various seasonal trade winds to Singapore, where they
were bought, sold, traded, or stored for a future customer. By the late
nineteenth century, however, the British overlords of Singapore had
extended their influence or control throughout the Malay Peninsula, and
the port acquired a large hinterland rich in resources. Singapore became
the outlet for Malaya's tin and rubber, as well as the gateway through
which were funneled supplies and workers for the peninsula's mines and
plantations. Tin smelting and rubber processing were added to the list
of services that Singapore provided--a long list that already included
wholesaling, ship repair and provisioning, warehousing, and a host of
banking and financial services.
In 1990 the economy of modern Singapore was still based on the same
services that were performed by the colonial port, although most of
these services had been greatly expanded or modified and new ones added.
The major sectors of the economy were the regional entrep�t trade,
export-oriented manufacturing, petroleum refining and shipping,
production of goods and services for the domestic economy, and a vastly
expanded services industry.
When independence was suddenly thrust upon Singapore in 1965, its
economic prospects looked bleak, if not precarious. In the aftermath of
World War II, Singapore had faced staggering problems of high
unemployment, slow economic growth, inadequate housing, decaying
infrastructure, and labor and social unrest. Separation from Malaysia
meant the loss of its economic hinterland, and Indonesia's policy of
military Confrontation directed at Singapore and Malaysia had dried up
the entrep�t trade from that direction. Moreover, with the announcement
in 1968 of Britain's departure from the island's bases, Singapore faced
the loss of 20 percent of its jobs. These problems led Singapore's
leadership to take a strong role in guiding the nation's economy. The
government aggressively promoted export-oriented, labor-intensive
industrialization through a program of incentives designed to attract
foreign investment. By 1972 one-quarter of Singapore's manufacturing
firms were either foreign-owned or joint-venture companies, with the
United States and Japan both major investors. The response of foreign
investors to Singapore's favorable investment climate and the rapid
expansion of the world economy at that time were factors in the annual
double-digit growth of the country's GDP during most of the period from
1965 through 1973. By the late 1970s, however, government planners had
adopted a policy of replacing Singapore's labor-intensive manufacturing
with skill- and technology-intensive, high value-added industries.
Information technology was particularly targeted for expansion, and by
1989 Singapore was the world's largest producer of disk drives and disk
drive parts. In that year, earnings from manufacturing accounted for 30
percent of the country's GDP.
Although Singapore lost its former hinterland when it separated from
Malaysia, its northern neighbor remained the leading source of primary
imports and a major destination for Singapore's manufactured exports.
Malaysia was Singapore's third largest overall trading partner in 1988,
and Singaporean companies were major investors in Malaysia's southern
state of Johor. The entrep�t trade with Indonesia had long since
revived following the end of Confrontation in 1966. By the late 1980s,
Singapore was the world's third largest petroleum-refining center as
well as third largest oil-trading center, serving the needs of oil-rich
Indonesia and Malaysia. By 1988 Singapore had nosed out Rotterdam as the
world's busiest port in terms of tonnage. Some 700 shiplines used its
modern facilities each year, including Singapore's own merchant fleet,
which ranked fifteenth worldwide. Four major shipyards employed about
70,000 workers, about 40 percent of whom were from neighboring Asian
countries.
One of the fastest growing sectors of the economy was Singapore's
international banking and financial services sector, which accounted for
nearly 25 percent of the country's GDP in the late 1980s. Historically,
Singapore served as the financial services center for Southeast Asia,
and in the late 1980s it ranked with Hong Kong as the two most important
Asian financial centers after Tokyo. The government provided incentives
for the continuing diversification and automation of financial services,
and Singapore's political stability and top-notch infrastructure were
important attractions for international bankers and investors. Trade,
manufacturing, and international financial services were closely linked
in Singapore, which in 1990 hosted more than 650 multinational companies
and several thousand international financial institutions and trading
firms. Singapore's reliance on the international economy, over which it
had little control, provided incentive for the government to play a
strong role in regulating domestic conditions. Soon after independence,
the government brought under control the serious labor unrest of the
1950s and early 1960s in order to present a more favorable climate for
foreign investment. Discipline imposed on the labor force was
counterbalanced, however, by provisions for workers' welfare. While the
booming economy of the late 1960s and 1970s brought new jobs to the
private sector, government provision of subsidized housing, education,
health services, and public transportation created jobs in the public
sector. The Central Provident Fund, built up by compulsory contributions
by both employer and employee, provided the necessary capital for
government projects as well as for the country's comprehensive social
security scheme.
Singapore, Inc., as some observers refer to the country, spent the
first twenty-five years of its independence under the same management.
Led by Lee Kuan Yew, the country's first and only prime minister, the
People's Action Party ( PAP) won all or nearly all of the seats in
parliament in the six elections held between 1959 and 1988. Based on a
British parliamentary system, with free and open elections, the
Singapore government was recognized for its stability, honesty, and
effectiveness. Critics complained, however, that the government's
authoritarian leadership reserved for itself all power of decision
making and blocked the rise of an effective opposition. A small nucleus
of leaders centered around Lee had indeed closely guided the country
from its turbulent preindependence days and crafted the policies that
led to Singapore's economic development. During the 1980s, however, a
second generation of leaders was carefully groomed to take over, and in
early 1990, only Lee remained of the first generation leaders.
In late 1989, Lee announced that he would step down in late 1990 and
that his successor, First Deputy Prime Minister Goh Chok Tong, had
already largely taken over the day-to-day management of the government.
However, based on the prime minister's own assertions that he was not
yet ready to relinquish all control, observers speculated on just what
powers Lee would continue to hold. Goh acknowledged in late 1989 the
growing sophistication and rising expectations of younger Singaporeans,
who want a greater participation in the country's political life, and
noted that he expected the opposition to claim a larger share of seats
in parliament in the 1990s. In contrasting his leadership style with
that of Lee, Goh stated that Lee "believes in firm government from
the center . . . whereas our style is a little more consultative, more
consensus-building." Behind Goh in the Singapore leadership queue
was believed to be Lee Kuan Yew's son, Brigadier General Lee Hsien
Loong, who served in the cabinet as minister for trade and industry and
second minister for defence. His meteoric rise in the late 1980s through
the ranks of bureaucratic and political responsibility was regarded with
interest by both foreign and domestic observers.
The transition to a new generation of leaders was a phenomenon not
unique to Singapore. In neighboring Malaysia and Indonesia, the
independence generation was also rapidly dwindling, and the 1990s will
surely mark the passing from the scene of Prime Minister Mahathir
Mohamad and President Soeharto as well as Lee Kuan Yew. The close
relationship between Singapore and both its neighbors had been built to
a large extent on personal ties between Lee and his counterparts in
Malaysia and Indonesia. Nonetheless, the new leadership of these
countries will very likely continue to build on the foundation laid by
their predecessors.
In late 1989, Goh discussed the prospect of Johor State, the nearby
Indonesian island of Batam (currently being developed), and Singapore
forming a "triangle of growth" within the region in a
cooperative rather than competitive effort. There were also signs of
increased military cooperation among the three countries. Singapore, for
example, conducted bilateral land exercises for the first time with both
Malaysia and Indonesia in 1989. Bilateral air and naval exercises had
been conducted with both countries during most of the 1980s. All three
countries (along with Thailand, Brunei, and the Philippines) were
members of the Association of Southeast Asian Nations ( ASEAN), formed
in 1967 to promote closer political and economic cooperation within the
region. The invasion of Cambodia by Vietnam in 1978 brought increased
unity to the organization throughout the 1980s, as it sought to find a
peaceful solution to the Cambodian problem. Although there was
considerable bilateral military cooperation among ASEAN states, the
organization was not viewed by its members as a military alliance.
However, Singapore and Malaysia, along with Britain, Australia, and New
Zealand, were also members of the 1971 Five-Powers Defence Agreement,
which provided for consultation and support by the latter three nations
in the event of an attack on Singapore or Malaysia. Cooperation under
the agreement diminished during the 1970s, but by the late 1980s
extensive military exercises involving all five participants were again
being held.
In August 1989, Lee Kuan Yew created a stir within the region by
stating that Singapore was "prepared to host some United States
facilities to make it easier for the Philippines to host the United
States bases there." Malaysia reacted negatively to the
announcement, and other ASEAN countries expressed some dismay. In
October, however, the Singapore foreign ministry clarified the issue by
stating that an increased use of Singapore's maintenance and repair
facilities by United States ships had been agreed on by the two
countries, as had short-term visits by United States aircraft to
Singapore's Paya Lebar Air Base. The agreement followed a period of
somewhat strained relations between the two nations, during which the
United States had been critical of Singapore's use of its Internal
Security Act to detain dissidents indefinitely, and Singapore had
accused Washington of meddling in its internal affairs. The United
States, however, was Singapore's largest trading partner and foreign
investor, and the relationship was one that neither country was eager to
upset.
By the last decade of the twentieth century, the former colonial port
of Singapore had become a global financial, trading, and industrial
center that continued to live by its wits in the world of international
trade, just as it had done in the nineteenth century. Singapore's
leadership and its people have always managed to adapt to the changing
demands of the world economy, on which so much of their livelihood
depended. In the coming decade, however, a new generation of leaders
will take full control of the nation's government and economy. Before
them lies the task of reconciling the need to steer a steady course in
the nation's continuing development with the people's growing
aspirations for an increased share in political and economic decision
making.
Singapore
Singapore - History
Singapore
Precolonial Era
Located astride the sea routes between China and India, from ancient
times the Malay Archipelago served as an entrep�t, supply point, and
rendezvous for the sea traders of the kingdoms and empires of the Asian
mainland and the Indian subcontinent. The trade winds of the South China
Sea brought Chinese junks laden with silks, damasks, porcelain, pottery,
and iron to seaports that flourished on the Malay Peninsula and the
islands of Sumatra and Java. There they met with Indian and Arab ships,
brought by the monsoons of the Indian Ocean, carrying cotton textiles,
Venetian glass, incense, and metalware. Fleets of swift prahu
(interisland craft) supplied fish, fruit, and rice from Java and pepper
and spices from the Moluccas in the eastern part of the archipelago. All
who came brought not only their trade goods but also their cultures,
languages, religions, and technologies for exchange in the bazaars of
this great crossroads.
In time, the ports of the peninsula and archipelago formed the
nucleus of a succession of seabased kingdoms, empires, and sultanates.
By the late seventh century, the great maritime Srivijaya Empire, with
its capital at Palembang in eastern Sumatra, had extended its rule over
much of the peninsula and archipelago. Historians believe that the
island of Singapore was probably the site of a minor port of Srivijaya.
Temasek and Singapura
Although legendary accounts shroud Singapore's earliest history,
chroniclers as far back as the second century alluded to towns or cities
that may have been situated at that favored location. Some of the
earliest records of this region are the reports of Chinese officials who
served as envoys to the seaports and empires of the Nanyang (southern
ocean), the Chinese term for Southeast Asia. The earliest first-hand
account of Singapore appears in a geographical handbook written by the
Chinese traveler Wang Dayuan in 1349. Wang noted that Singapore Island,
which he called Tan-ma-hsi (Danmaxi), was a haven for several hundred
boatloads of pirates who preyed on passing ships. He also described a
settlement of Malay and Chinese living on a terraced hill known in Malay
legend as Bukit Larangan (Forbidden Hill), the reported burial place of
ancient kings. The fourteenth-century Javanese chronicle, the Nagarakertagama,
also noted a settlement on Singapore Island, calling it Temasek.
A Malay seventeenth-century chronicle, the Sejarah Melayu (Malay
Annals), recounts the founding of a great trading city on the
island in 1299 by a ruler from Palembang, Sri Tri Buana, who named the
city Singapura ("lion city") after sighting a strange beast
that he took to be a lion. The prosperous Singapura, according to the Annals,
in the mid-fourteenth century suffered raids by the expanding Javanese
Majapahit Empire to the south and the emerging Thai kingdom of Ayutthaya
to the north, both at various times claiming the island as a vassal
state.
The Annals, as well as contemporaneous Portuguese accounts,
note the arrival around 1388 of King Paramesvara from Palembang, who was
fleeing Majapahit control. Although granted asylum by the ruler of
Singapura, the king murdered his host and seized power. Within a few
years, however, Majapahit or Thai forces again drove out Paramesvara,
who fled northward to found eventually the great seaport and kingdom of
Malacca. In 1414 Paramesvara converted to Islam and established the
Malacca Sultanate, which in time controlled most of the Malay Peninsula,
eastern Sumatra, and the islands between, including Singapura. Fighting
ships for the sultanate were supplied by a senior Malaccan official
based at Singapura. The city of Malacca served not only as the major
seaport of the region in the fifteenth century, but also as the focal
point for the dissemination of Islam throughout insular Southeast Asia.
Johore Sultanate
When the Portuguese captured Malacca in 1511, the reigning Malaccan
sultan fled to Johore in the southern part of the Malay Peninsula, where
he established a new sultanate. Singapura became part of the new Johore
Sultanate and was the base for one of its senior officials in the latter
sixteenth century. In 1613, however, the Portuguese reported burning
down a trading outpost at the mouth of the Temasek (Singapore) River,
and Singapura passed into history.
In the following two centuries, the island of Temasek was largely
abandoned and forgotten as the fortunes of the Johore Sultanate rose and
fell. By 1722 a vigorous seafaring people from the island of Celebes
(modern Sulawesi, Indonesia) had become the power behind the throne of
the Johore Sultanate. Under Bugis influence, the sultanate built up a
lucrative entrep�t trade, centered at Riau, south of Singapore, in
present-day Sumatra. Riau also was the site of major plantations of
pepper and gambier, a medicinal plant used in tanning. The Bugis used
waste material from the gambier refining process to fertilize pepper
plants, a valuable crop, but one that quickly depletes soil nutrients.
By 1784 an estimated 10,000 Chinese laborers had been brought from
southern China to work the gambier plantations on Bintan Island in the
Riau archipelago (now part of Indonesia). In the early nineteenth
century, gambier was in great demand in Java, Siam, and elsewhere, and
cultivation of the crop had spread from Riau to the island of Singapore.
The territory controlled by the Johore Sultanate in the late
eighteenth century was somewhat reduced from that under its precursor,
the Malacca Sultanate, but still included the southern part of the Malay
Peninsula, the adjacent area of Sumatra, and the islands between,
including Singapore. The sultanate had become increasingly weakened by
division into a Malay faction, which controlled the peninsula and
Singapore, and a Bugis faction, which controlled the Riau Archipelago
and Sumatra. When the ruling sultan died without a royal heir, the Bugis
had proclaimed as sultan the younger of his two sons by a commoner wife.
The sultan's elder son, Hussein (or Tengku Long) resigned himself to
living in obscurity in Riau.
Although the sultan was the nominal ruler of his domain, senior
officials actually governed the sultanate. In control of Singapore and
the neighboring islands was Temenggong Abdu'r Rahman, Hussein's
father-in-law. In 1818 the temenggong (a high Malay official)
and some of his followers left Riau for Singapore shortly after the
Dutch signed a treaty with the Bugis-controlled sultan, allowing them to
station a garrison at Riau. The temenggong's settlement on the
Singapore River included several hundred orang laut (sea
gypsies in Malay) under Malay overlords who owed allegiance to the temenggong.
For their livelihood the inhabitants depended on fishing, fruit growing,
trading, and occasional piracy. Large pirate fleets also used the strait
between Singapore and the Riau Archipelago as a favorite rendezvous.
Also living on the island in settlements along the rivers and creeks
were several hundred indigenous tribespeople, who lived by fishing and
gathering jungle produce. Some thirty Chinese, probably brought from
Riau by the temenggong, had begun gambier and pepper production
on the island. In all, perhaps a thousand people inhabited the island of
Singapore at the dawn of the colonial era.
Singapore
Singapore - Founding and Early Years
Singapore
By the early seventeenth century, both the Dutch and the English were
sending regular expeditions to the East Indies. The English soon gave up
the trade, however, and concentrated their efforts on India. In 1641 the
Dutch captured Malacca and soon after replaced the Portuguese as the
preeminent European power in the Malay Archipelago. From their capital
at Batavia on Java, they sought to monopolize the spice trade. Their
short-sighted policies and harsh treatment of offenders, however,
impoverished their suppliers and encouraged smuggling and piracy by the
Bugis and other peoples. By 1795, the Dutch enterprise in the East was
losing money and, in Europe, the Netherlands was at war with France. The
Dutch king fled to Britain where, in desperation, he issued the Kew
Letters, by which all Dutch overseas territories were temporarily placed
under British authority in order to keep them from falling to the
French.
Anglo-Dutch Competition
In the late eighteenth century, the British began to expand their
commerce with China from their bases in India through both private
traders and the British East India Company. The company had occupied a
small settlement at Bencoolen (Bengkulu) on the western coast of Sumatra
since 1684; from there it had engaged in the pepper trade after being
forced out of Java by the Dutch. Acutely aware of the need for a base
somewhere midway between Calcutta and Guangzhou, the company leased the
island of Penang, on the western coast of the Malay Peninsula, from the
sultan of Kedah in 1791. From these posts at Penang and Bencoolen, the
British began in 1795 to occupy the Dutch possessions placed temporarily
in their care by the Kew Letters, including Malacca and Java. After war
in Europe ended in 1814, however, the British agreed to return Java and
Malacca to the Dutch. By 1818 the Dutch had returned to the East Indies
and had reimposed their restrictive trade policies. In that same year,
the Dutch negotiated a treaty with the Bugis-controlled sultan of Johore
granting them permission to station a garrison at Riau, thereby giving
them control over the main passage through the Strait of Malacca.
British trading ships were heavily taxed at Dutch ports and suffered
harassment by the Dutch navy. Meanwhile, the British government and the
British East India Company officials in London, who were concerned with
maintaining peace with the Dutch, consolidating British control in
India, and reducing their commitments in the East Indies, considered
relinquishing Bencoolen and perhaps Penang to the Dutch in exchange for
Dutch territories in India.
Raffles' Dream
Sir Thomas Stamford Raffles, the lieutenant governor of Bencoolen in
1818, vigorously opposed his government's plan to abandon control of the
China trade to the Dutch. Raffles, who had started his career as a clerk
for the British East India Company in London, was promoted at the age of
twenty-three to assistant secretary of the newly formed government in
Penang in 1805. A serious student of the history and culture of the
region and fluent in Malay, Raffles served as governor general of Java
(1811-16). In 1818 Raffles sailed from Bencoolen to India, where he
convinced Governor General Lord Hastings of the need for a British post
on the southern end of the Strait of Malacca. Lord Hastings authorized
Raffles to secure such a post for the British East India Company,
provided that it did not antagonize the Dutch. Arriving in Penang,
Raffles found Governor General James Bannerman unwilling to cooperate.
When he learned that the Dutch had occupied Riau and were claiming that
all territories of the sultan of Johore were within their sphere of
influence, Raffles dispatched Colonel William Farquhar, an old friend
and Malayan expert, to survey the Carimon Islands (modern Karimun
Islands near Riau). Disregarding Bannerman's orders to him to await
further instructions from Calcutta, Raffles slipped out of Penang the
following night aboard a private trading ship and caught up with
Farquhar. Raffles knew of Singapore Island from his study of Malay texts
and determined to go there.
On January 28, 1819, Raffles and Farquhar anchored near the mouth of
the Singapore River. The following day the two men went ashore to meet
Temenggong Abdu'r Rahman, who granted provisional permission for the
British East India Company to establish a trading post on the island,
subject to the approval of Hussein. Raffles, noting the protected
harbor, the abundance of drinking water, and the absence of the Dutch,
began immediately to unload troops, clear the land on the northeast side
of the river, set up tents, and hoist the British flag. Meanwhile, the temenggong
sent to Riau for Hussein, who arrived within a few days. Acknowledging
Hussein as the rightful sultan of Johore, on February 6 Raffles signed a
treaty with him and the temenggong confirming the right of the
British East India Company to establish a trading post in return for an
annual payment (in Spanish dollars, the common currency of the region at
the time) of Sp$5,000 to Hussein and Sp$3,000 to the temenggong.
Raffles then departed for Bencoolen, leaving Farquhar in charge, with
instructions to clear the land, construct a simple fortification, and
inform all passing ships that there were no duties on trade at the new
settlement.
The immediate reaction to Raffles' new venture was mixed. Officials
of the British East India Company in London feared that their
negotiations with the Dutch would be upset by Raffles' action. The Dutch
were furious because they considered Singapore within their sphere of
influence. Although they could easily have overcome Farquhar's tiny
force, the Dutch did not attack the small settlement because the angry
Bannerman assured them that the British officials in Calcutta would
disavow the whole scheme. In Calcutta, meanwhile, both the commercial
community and the Calcutta Journal welcomed the news and urged
full government support for the undertaking. Lord Hastings ordered the
unhappy Bannerman to provide Farquhar with troops and money. Britains
foreign minister Lord Castlereagh, reluctant to relinquish to the Dutch
"all the military and naval keys of the Strait of Malacca,"
had the question of Singapore added to the list of topics to be
negotiated with the Dutch, thus buying time for the new settlement.
The opportunity to sell supplies at high prices to the new settlement
quickly attracted many Malacca traders to the island. Word of
Singapore's free trade policy also spread southeastward through the
archipelago, and within six weeks more than 100 Indonesian interisland
craft were anchored in the harbor, as well as one Siamese and two
European ships. Raffles returned in late May to find that the population
of the settlement had grown to nearly 5,000, including Malays, Chinese,
Bugis, Arabs, Indians, and Europeans. During his four-week stay, he drew
up a plan for the town and signed another agreement with Hussein and the
temenggong establishing the boundaries of the settlement. He
wrote to a friend that Singapore "is by far the most important
station in the East; and, as far as naval superiority and commercial
interests are concerned, of much higher value than whole continents of
territory."
Early Administration and Growth
Although the India-China trade was partly responsible for the
overnight success of Singapore, even more important was the
wellestablished entrep�t trade of the East Indies that the new port
captured from Riau and other trade centers. The news of the free port
brought not only traders and merchants but also permanent settlers.
Malays came from Penang, Malacca, Riau, and Sumatra. Several hundred
boatloads of Hussein's followers came from Riau, and the new sultan had
built for himself an istana (palace in Malay), thus making
Singapore his headquarters. The growing power of the Dutch in Riau also
spurred several hundred Bugis traders and their families to migrate to
the new settlement. Singapore was also a magnet for the Nanyang Chinese
who had lived in the region for generations as merchants, miners, or
gambier farmers. They came from Penang, Malacca, Riau, Manila, Bangkok,
and Batavia to escape the tariffs and restrictions of those places and
to seek their fortunes. Many intermarried with Malay women, giving rise
to the group known as the Baba
Chinese. The small Indian population included both
soldiers and merchants. A few Armenian merchants from Brunei and Manila
were also attracted to the settlement, as were some leading Arab
families from Sumatra. Most Europeans in the early days of Singapore
were officials of the British East India Company or retired merchant sea
captains.
Not wanting the British East India Company to view Singapore as an
economic liability, Raffles left Farquhar a shoestring budget with which
to administer the new settlement. Prevented from either imposing trade
tariffs or selling land titles to raise revenue, Farquhar legalized
gambling and the sale of opium and arak, an alcoholic drink.
The government auctioned off monopoly rights to sell opium and spirits
and to run gambling dens under a system known as tax farming, and the
revenue thus raised was used for public works projects. Maintenance of
law and order in the wideopen seaport was among the most serious
problems Farquhar faced. There was constant friction among the various
immigrant groups, particularly between the more settled Malays and
Chinese from Malacca and the rough and ready followers of the temenggong
and the sultan. The settlement's merchants eventually funded night
watchmen to augment the tiny police force.
When Raffles returned to Singapore from Bencoolen in October 1822, he
immediately began drawing up plans for a new town. An area along the coast about five kilometers long and one
kilometer deep was designated the government and commercial quarter. A
hill was leveled and the dirt used to fill a nearby swamp in order to
provide a place for the heart of the commercial area, now Raffles Place.
An orderly and scientifically laid out town was the goal of Raffles, who
believed that Singapore would one day be "a place of considerable
magnitude and importance." Under Raffles' plan, commercial
buildings were to be constructed of brick with tiled roofs, each with a
two-meter covered walkway to provide shelter from sun and rain. Spaces
were set aside for shipyards, markets, churches, theaters, police
stations, and a botanical garden. Raffles had a wooden bungalow built
for himself on Government Hill.
Each immigrant group was assigned an area of the settlement under the
new plan. The Chinese, who were the fastest growing group, were given
the whole area west of the Singapore River adjoining the commercial
district; Chinatown was further divided among the various dialect
groups. The temenggong and his followers were moved several
kilometers west of the commercial district, mainly in an effort to
curtail their influence in that area. The headmen or kapitans
of the various groups were allotted larger plots, and affluent Asians
and Europeans were encouraged to live together in a residential area
adjacent to the government quarter.
In the absence of any legal code, Raffles in early 1823 promulgated a
series of administrative regulations. The first required that land be
sold on permanent lease at a public auction and that it must be
registered. The second reiterated Singapore's status as a free port, a
popular point with the merchants. In his farewell remarks, Raffles
assured them that "Singapore will long and always remain a free
port and no taxes on trade or industry will be established to check its
future rise and prosperity." The third regulation made English
common law the standard, although Muslim law was to be used in matters
of religion, marriage, and inheritance involving Malays.
Raffles was an enlightened administrator for his time. He believed in
the prevention of crime and the reform, rather than the mere punishment,
of criminals. Payment of compensation to the injured by the offender was
to be considered as important as punishment. Only murder was to be
considered a capital offense, and various work and training programs
were used to turn prisoners into useful settlers. Raffles shut down all
gambling dens and heavily taxed the sale of liquor and opium. He
abolished outright slavery in 1823, but was unable to eradicate debt
bondage, by which immigrants often were forced to work years at hard
labor to pay for their passage.
Raffles felt that under Farquhar the temenggong and the
sultan had wielded too much power, receiving one-third of the proceeds
from the opium, liquor, and gambling revenues, and demanding presents
from the captains of the Asian ships that dropped anchor there. Hussein
and the temenggong, however, viewed Singapore as a thriving
entrep�t in the mold of the great port cities of the Malay maritime
empires of Srivijaya, Malacca, and Johore. As rulers of the island, they
considered themselves entitled to a share of the power and proceeds of
the settlement. In June 1823, Raffles managed to persuade Hussein and
the temenggong to give up their rights to port duties and their
share in the other tax revenues in exchange for a pension of Sp$1,500
and Sp$800 per month, respectively. Because the Dutch still contested
the British presence in Singapore, Raffles did not dare push the issue
further. On March 17, 1824, however, the AngloDutch Treaty of London was
signed, dividing the East Indies into two spheres of influence. The
British would have hegemony north of a line drawn through the Strait of
Malacca, and the Dutch would control the area south of the line. As a
result, the Dutch recognized the British claim to Singapore and
relinquished power over Malacca in exchange for the British post at
Bencoolen. On August 3, with their claim to Singapore secure, the
British negotiated a new treaty with the sultan and the temenggong,
by which the Malay rulers were forced to cede Singapore and the
neighboring islands to the British East India Company for cash payments
and increased pensions. Under the treaty, the Malay chiefs also agreed
to help suppress piracy, but the problem was not to be solved for
several more decades.
In October 1823, Raffles left Singapore for Britain, never to return.
Before leaving, he replaced Farquhar with the Scotsman John Crawfurd, an
efficient and frugal administrator who guided the settlement through
three years of vigorous growth. Crawford continued Raffles' struggles
against slavery and piracy, but he permitted the gambling houses to
reopen, taxed them, and used the revenue for street widening, bridge
building, and other civic projects. He failed to support, however,
Raffles' dream of higher education for the settlement. As his last
public act, Raffles had contributed Sp$2,000 toward the establishment of
a Singapore Institution, which he had envisioned as a training ground
for Asian teachers and civil servants and a place where European
officials could gain an appreciation of the rich cultural heritage of
the region as Raffles himself had. He had hoped that the institution
would attract the sons of rulers and chiefs of all the region. Crawfurd,
however, advised the company officials in Calcutta that it would be
preferable to support primary education. In fact, education at all
levels was neglected until much later.
Singapore
Singapore - A Flourishing Free Port
Singapore
In the first half-century after its founding, Singapore grew from a
precarious trading post of the British East India Company populated by a
few thousand to a bustling, cosmopolitan seaport of 85,000. Although the
general trend of Singapore's economic status was upward during this
period, the settlement endured economic recessions as well as
prosperity, fires and floods as well as building booms, and bureaucratic
incompetence as well as able administration. In 1826 the British East
India Company combined Singapore with Penang and Malacca to form the
Presidency of the Straits Settlements, with its capital at Penang. The
new bureaucratic apparatus proved to be expensive and cumbersome,
however, and in 1830 the Straits Settlements were reduced to a
residency, or subdivision, of the Presidency of Bengal. Although
Singapore soon overshadowed the other settlements, Penang remained the
capital until 1832 and the judicial headquarters until 1856. The
overworked civil service that administered Singapore remained about the
same size between 1830 and 1867, although the population quadrupled
during that period. Saddled with the endless narrative and statistical
reports required by Bengal, few civil servants had time to learn the
languages or customs of the people they governed.
Although the European and Asian commercial community was reasonably
satisfied with the administration of the settlement under Bengal, an
economic depression in the 1840s caused some to consider the merits of
Singapore being administered directly as a crown colony. The advent of
the steamship had made Singapore less dependent on Calcutta and more
closely tied to the London commercial and political scene. By
mid-century, the parent firms of most of Singapore's British-owned
merchant houses were located in London rather than Calcutta. In 1851,
following a visit to Singapore, Lord Dalhousie, the governor general of
India, separated the Straits Settlements from Bengal and placed them
directly under his own charge. In the following sixteen years, a number
of issues arose that caused increased agitation to remove the Straits
Settlements completely from administration from India and place it
directly under the British Colonial Office. Among these issues were the
need for protection against piracy and Calcutta's continuing attempts to
levy port duties on Singapore. Mostly as a result of the need for a
place other than fever-ridden Hong Kong to station British troops in
Asia, London designated the Straits Settlements a crown colony on April
1, 1867.
Financial Success
Trade at Singapore had eclipsed that of Penang by 1824, when it
reached a total of Sp$11 million annually. By 1869 annual trade at
Singapore had risen to Sp$89 million. The cornerstone of the
settlement's commercial success was the entrep�t trade, which was
carried on with no taxation and a minimum of restriction. The main
trading season began each year with the arrival of ships from China,
Siam, and Cochinchina (as the southern part of Vietnam was then known).
Driven by the northeast monsoon winds and arriving in January, February,
and March, the ships brought immigrant laborers and cargoes of dried and
salted foods, medicines, silk, tea, porcelain, and pottery. They left
beginning in May with the onset of the southwest monsoons, laden with
produce, spices, tin, and gold from the Malay Archipelago, opium from
India, and English cotton goods and arms. The second major trading
season began in Septeember or October with the arrival of the Bugis
traders in their small, swift prahu, bringing rice, pepper,
spices, edible bird nests and shark fins, mother-of-pearl, gold dust,
rattan, and camphor from the archipelago. They departed carrying British
manufactures, cotton goods, iron, arms, opium, salt, silk, and
porcelain. By mid-century, there were more than twenty British merchant
houses in Singapore, as well as German, Swiss, Dutch, Portuguese, and
French firms. The merchants would receive cargoes of European or Indian
goods on consignment and sell them on commission.
Most of the trade between the European and Asian merchants was
handled by Chinese middlemen, who spoke the necessary languages and knew
the needs of their customers. Many of the middlemen had trained as
clerks in the European trading firms of Malacca. With their experience,
contacts, business acumen, and willingness to take risks, the middlemen
were indispensable to the merchants. For the Chinese middlemen, the
opportunities for substantial profit were great; but so were the risks.
Lacking capital, the middlemen bought large quantities of European goods
on credit with the hope of reselling them to the Chinese or Bugis ship
captains or themselves arranging to ship them to the markets of Siam or
the eastern Malay Peninsula. If, however, buyers could not be found or
ships were lost at sea, the middlemen faced bankruptcy or prison.
Although the merchants also stood to lose under such circumstances, the
advantages of the system and the profits to be made kept it flourishing.
The main site for mercantile activity in mid-century Singapore was
Commercial Square, renamed Raffles Place in 1858. Besides the European
merchant houses located on the square, there were in 1846 six Jewish
merchant houses, five Chinese, five Arab, two Armenian, one American,
and one Indian. Each merchant house had its own pier for loading and
unloading cargo; and ship chandlers, banks, auction houses, and other
businesses serving the shipping trade also were located on the square.
In the early years, merchants lived above their offices; but by
mid-century most had established themselves in beautiful houses and
compounds in a fashionable section on the east bank of the Singapore
River.
Construction of government buildings lagged far behind commercial
buildings in the early years because of the lack of taxgenerated
revenue. The merchants resisted any attempts by Calcutta to levy duties
on trade, and the British East India Company had little interest in
increasing the colony's budget. After 1833, however, many public works
projects were constructed by the extensive use of Indian convict labor.
Irish architect George Drumgold Coleman, who was appointed
superintendent of public works in that year, used convicts to drain
marshes, reclaim seafront, lay out roads, and build government
buildings, churches, and homes in a graceful colonial style.
Probably the most serious problem facing Singapore at midcentury was
piracy, which was being engaged in by a number of groups who found easy
pickings in the waters around the thriving port. Some of the followers
of the temenggong's son and heir, Ibrahim, were still engaging
in their "patrolling" activities in the late 1830s. Most
dangerous of the various pirate groups, however, were the Illanun
(Lanun) of Mindanao in the Philippines and northern Borneo. These fierce
sea raiders sent out annual fleets of 50 to 100 well-armed prahu,
which raided settlements, attacked ships, and carried off prisoners who
were pressed into service as oarsmen. The Illanun attacked not only
small craft from the archipelago but also Chinese and European sailing
ships. Bugis trading captains threatened to quit trading at Singapore
unless the piracy was stopped. In the 1850s, Chinese pirates, who boldly
used Singapore as a place to buy arms and sell their booty, brought the
trade between Singapore and Cochinchina to a standstill. The few patrol
boats assigned by the British East India Company to protect the Straits
Settlements were totally inadequate, and the Singapore merchants
continually petitioned Calcutta and London for aid in stamping out the
menace.
By the late 1860s, a number of factors had finally led to the demise
of piracy. In 1841, the governor of the Straits Settlements, George
Bonham, recognized Ibrahim as temenggong of Johore, with the
understanding that he would help suppress piracy. By 1850 the Royal Navy
was patrolling the area with steam-powered ships, which could navigate
upwind and outmaneuver the pirate sailing ships. The expansion of
European power in Asia also brought increased patrolling of the region
by the Dutch in Sumatra, the Spanish in the Philippines, and the British
from their newly established protectorates on the Malay Peninsula. China
also agreed to cooperate in suppressing piracy under the provisions of
treaties signed with the Western powers in 1860.
Singapore's development and prosperity at mid-century were largely
confined to the coast within a few kilometers of the port area. The
interior remained a dense jungle ringed by a coastline of mangrove
swamps. Attempts to turn the island to plantation agriculture between
1830 and 1840 had met with little success. Nutmeg, coffee, sugar,
cotton, cinnamon, cloves, and indigo all fell victim to pests, plant
diseases, or insufficient soil fertility. The only successful
agricultural enterprises were the gambier and pepper plantations,
numbering about 600 in the late 1840s and employing some 6,000 Chinese
laborers. When the firewood needed to extract the gambier became
depleted, the plantation would be moved to a new area. As a result, the
forests of much of the interior of the island had been destroyed and
replaced by coarse grasses by the 1860s, and the gambier planters had
moved their operations north to Johore. This pressure on the land also
affected the habitats of the wildlife, particularly tigers, which began
increasingly to attack villagers and plantation workers. Tigers
reportedly claimed an average of one victim per day in the late 1840s.
When the government offered rewards for killing the animals, tiger
hunting became a serious business and a favorite sport. The last year a
person was reported killed by a tiger was 1890, and the last wild tiger
was shot in 1904.
A Cosmopolitan Community
As Singapore prospered and grew, the size and diversity of its
population kept pace. By 1827, the Chinese had become the most numerous
of Singapore's various ethnic groups. Many of the Chinese came from
Malacca, Penang, Riau, and other parts of the Malay Archipelago to which
their forebears had migrated decades or even generations before. More
recent Chinese immigrants were mainly from the southeastern provinces of
Guangdong and Fujian and spoke either the Hokkien, Teochiu, Cantonese,
or Hakka dialects. In an extension of the common Chinese practice of
sojourning, in which men temporarily left their home communities to seek
work in nearby or distant cities, most migrants to Singapore saw
themselves as temporary residents intending to return to home and family
after making a fortune or at least amassing enough capital to buy land
in their home district. Many did return; more did not. Even those who
never returned usually sent remittances to families back home.
To help them face the dangers, hardships, and loneliness of the
sojourner life, most men joined or were forced to join secret societies
organized by earlier immigrants from their home districts. The secret
societies had their origin in southern China, where, in the late
seventeenth century, the Heaven, Earth, and Man (or Triad) Society was
formed to oppose the Qing (1644-1911) dynasty. By the nineteenth
century, secret societies in China acted as groups that organized urban
unskilled labor and used coercion to win control of economic niches,
such as unloading ships, transporting cotton, or gambling and
prostitution. The same pattern extended all over Southeast Asia, where
immigrants joined secret societies whose membership was restricted to
those coming from the same area and speaking the same dialect.
Membership gave the immigrants some security, in the form of guaranteed
employment and assistance in case of illness, but required loyalty to
the leaders and payment of a portion of an already meager wage. Although
the societies performed many useful social functions, they were also a
major source of crime and violence. By 1860 there were at least twelve
secret societies in Singapore, representing the various dialect and
subdialect groups. Invariably friction arose as each society sought to
control a certain area or the right to a certain tax farm. Civil war in
China in the 1850s brought a flood of new migrants from China, including
many rebels and other violent elements. Serious fighting between the
various secret societies broke out in 1854, but it remained a domestic
dispute within the Chinese community. Although not directed at the
government or the non-Chinese communities, such outbreaks disrupted
commerce and created a tense atmosphere, which led to the banning of
secret societies in 1889.
Just as the European merchant community used Chinese middlemen in
conducting their business, the Straits government relied on prominent
Chinese businessmen to act as go-betweens with the Chinese community. In
the early years, the Baba Chinese, who usually spoke English, served in
this capacity. By mid-century, however, immigrant Chinese from the
various dialect groups had begun to act as intermediaries. Some, such as
Seah Eu Chin, who was the go-between with the Teochiu community, were
well educated and from respected families. Seah, who made his fortune in
gambier and pepper plantations, was an early member of the Singapore
Chamber of Commerce, established in 1837, and a justice of the peace.
Probably the wealthiest and most prominent Chinese immigrant in the
nineteenth century was Hoo Ah Kay, nicknamed "Whampoa" after
his birthplace, who served as a go-between with the Cantonese-speaking
community. Hoo came as a penniless youth and made his fortune in
provisioning ships, merchandising, and speculating in land. He later
became the first Asian member of Singapore's Legislative Council and a
member of the Executive Council. Despite their close connections to the
European ruling class, Seah, Hoo, and other prominent Chinese carefully
retained their Chinese culture and values, as did the less prominent
immigrants.
Most Chinese immigrants fared far less well. If they survived the
rigors of the voyage, they were forced to work at hard labor for a year
or more to pay off their passage. Some were sent directly to the gambier
plantations or even to the tin mines of the Malay Peninsula. Others were
sent to toil on the docks or become construction workers. After paying
off their passage, they began earning a meager wage, which, unless
diverted for opium or gambling debts, was sent as a remittance to
families back in China. Wives were in short supply, since very few
Chinese women came to Singapore in the first few decades of the
settlement. Even by the mid-1860s, the ratio of Chinese men to women was
fifteen to one.
Until about 1860, Malays were the second largest group. The followers
of the temenggong mostly moved to Johore, where many of them
died of smallpox. The orang laut by mid-century merged with
other groups of Malay, who were drawn from Riau, Sumatra, and Malacca.
Generally peaceful and industrious, the Malays usually worked as
fishermen, boatmen, woodcutters, or carpenters.
Most of the Bugis sea traders migrated to Macassar after the Dutch
made it a free port in 1847, and by 1860 the Bugis population of
Singapore had declined to less than 1,000. Small numbers of Arabs, Jews,
and Armenians, many of them already well-to-do, were drawn to Singapore,
where they amassed even greater wealth. Another small group numbered
among Singapore's upper class were the Parsis, Indians of Iranian
descent who were adherents of Zoroastrianism.
Indians had become Singapore's second largest community by 1860,
numbering more than 11,000. Some of these people were laborers or
traders, who, like the Chinese, came with the hope of making their
fortune and returning to their homeland. Some were troops garrisoned at
Singapore by the government in Calcutta. Another group were convicts who
were first brought to Singapore from the detention center in Bencoolen
in 1825, after Bencoolen was handed over to the Dutch. Singapore then
became a major detention center for Indian prisoners. Rehabilitation
rather than punishment was emphasized, and prisoners were trained in
such skills as brick making, carpentry, rope making, printing, weaving,
and tailoring, which later would enable them to find employment.
Singapore's penal system was considered so enlightened that Dutch,
Siamese, and Japanese prison administrators came to observe it. Convict
labor was used to build roads, clear the jungle, hunt tigers, and
construct public buildings, some of which were still in use in 1989.
After completing their sentences, most convicts settled down to a useful
life in Singapore. As with Chinese and Europeans, Indian men far
outnumbered women because few Indian women came to Singapore before the
1860s. Some Indian Muslims married Malay women, however, and their
descendants were known as Jawi-Peranakan.
The highly unbalanced sex ratio in Singapore contributed to a rather
lawless, frontier atmosphere that the government seemed helpless to
combat. Little revenue was available to expand the tiny police force,
which struggled to keep order amid a continuous influx of immigrants,
often from the fringes of Asian society. This tide of immigration was
totally uncontrolled because Singapore's businessmen, desperate for
unskilled laborers, opposed restriction on free immigration as
vehemently as they resisted any restraint on free trade. Public health
services were almost nonexistent, and cholera, malnutrition, smallpox,
and opium use took a heavy toll in the severely overcrowded
working-class areas.
Singapore
Singapore - Crown Colony
Singapore
After years of campaigning by a small minority of the British
merchants, who had chafed under the rule of the Calcutta government, the
Straits Settlements became a crown colony on April 1, 1867. Under the
crown colony administration, the governor ruled with the assistance of
executive and legislative councils. The Executive Council included the
governor, the senior military official in the Straits Settlements, and
six other senior officials. The Legislative Council included the members
of the Executive Council, the chief justice, and four nonofficial
members nominated by the governor. The numbers of nonofficial members
and Asian council members gradually increased through the years.
Singapore dominated the Legislative Council, to the annoyance of Malacca
and Penang.
By the 1870s, Singapore businessmen had considerable interest in the
rubber, tin, gambier, and other products and resources of the Malay
Peninsula. Conditions in the peninsula were highly unstable, however,
marked by fighting between immigrants and traditional Malay authorities
and rivalry among various Chinese secret societies. Singapore served as
an entrep�t for the resources of the Malay Peninsula and, at the same
time, the port of debarkation for thousands of immigrant Chinese,
Indians, Indonesians, and Malays bound for the tin mines and rubber
plantations to the north. Some 250,000 Chinese alone disembarked in
Singapore in 1912, most of them on their way to the Malay states or to
the Dutch East Indies.
Although most Chinese immigrants merely passed through Singapore, the
Chinese population of the island grew rapidly, from 34,000 in 1878 to
103,000 in 1888. The colonial government established the Chinese
Protectorate in 1877 to deal with the serious abuses of the labor trade.
William Pickering, the first appointed Protector of Chinese, was the
first British official in Singapore who could speak and read Chinese.
Pickering was given power to board incoming ships and did much to
protect the newly arrived immigrants. In the early 1880s, he also
extended his protection to Chinese women entering the colony by working
to end forced prostitution. Because of his sympathetic approach and
administrative ability, the protector soon spread his influence and
protection over the whole Chinese community, providing arbitration of
labor, financial, and domestic disagreements, thereby undermining some
of the powers of the secret societies. Although no longer able to engage
in illegal immigration practices, the societies continued to cause
problems by running illegal gambling houses and supporting large-scale
riots that often paralyzed the city. In 1889, Governor Sir Cecil
Clementi-Smith sponsored a law to ban secret societies, which took
effect the following year. The result was to drive the societies
underground, where many of them degenerated into general lawlessness,
engaging in extortion, gambling operations, gang fights, and robbery.
The power of the secret societies, however, was broken.
The largest Chinese dialect group in the late nineteenth century were
the Hokkien, who were traditionally involved in trade, shipping,
banking, and industry. The next largest group, the Teochiu, engaged in
agricultural production and processing, including gambier, pepper, and
rubber production, rice and lumber milling, pineapple canning, and fish
processing. Cantonese served as artisans and laborers and a few made
their fortunes in tin. The two smallest groups, the Hakka and Hainanese,
were mostly servants, sailors, or unskilled laborers. Because wealth was
the key to leadership and social standing within the Chinese community
at that time, the Hokkien dominated organizations such as the Singapore
Chinese Chamber of Commerce and supplied most of the Chinese members of
the Legislative Council and the Chinese Advisory Board. The latter,
established in 1889 to provide a formal link between the British
government of the colony and the Chinese community, served as a place to
air grievances but had no power.
The affluent among Singapore's Chinese community increasingly saw
their prosperity and fortunes tied to those of the crown colony and the
British Empire. Western education, customs, and pastimes were adopted,
and the sons of Chinese businessmen were often sent to Britain for
university training. The Straits Chinese British Association was formed
in 1900 by Baba Chinese leaders to promote loyalty to the British Empire
as well as to advance the education and welfare of Singapore's Chinese.
Visiting British royalty were warmly received and British causes and
victories enthusiastically supported. The Straits
Chinese contributed generously to the British war
effort in World War I.
Although the Chinese upper class, particularly the Straits-born
Chinese, grew increasingly Westernized, the homeland exerted a
continuing pull on its loyalties that increased during this period.
Visits to China by Singapore Chinese became more common with the advent
of steamship travel. The relaxation by the 1870s of China's law
forbidding emigration (repealed in 1893) and the protection afforded
Singaporeans by British citizenship made it relatively safe for
prosperous businessmen to visit their homeland and return again to
Singapore. Upper-class Singapore Chinese frequently sent their sons to
school in China and encouraged them to find brides there, although they
themselves had often married local women.
In the latter part of the nineteenth century, China's ruling Qing
dynasty began to take an interest in the Nanyang Chinese and sought to
attract their loyalty and wealth to the service of the homeland. Chinese
consulates were established in Singapore, Malaya, the Dutch East Indies,
and other parts of the Nanyang. Whampoa was appointed Singapore's first
consul in 1877. He and his successors worked diligently to strengthen
the cultural ties of the Singapore Chinese to China by establishing a
cultural club, a debating society, Singapore's first Chinese--language
newspaper (Lat Pau), and various Chinese-language schools, in
which the medium of instruction was Chinese. One of the most important
functions of the consul, however, was to raise money for flood and
famine relief in China and for the general support of the Qing
government. With the upheaval in China following the Hundred Days'
Reform Movement in 1898, and its suppression by the Qing conservatives,
the Singapore Chinese and their pocketbooks were wooed by reformists,
royalists, and revolutionaries alike. Sun Yat-sen founded a Singapore
branch of the Tongmeng Hui, the forerunner of the Guomindang
(Kuomintang--Chinese Nationalist Party), in 1906. Not until the
successful Wuchang Uprising of 1911, however, did Sun receive the
enthusiastic support of Singapore Chinese.
Much smaller than the Chinese community and less organized in the
late nineteenth century was the Singapore Indian community. By 1880
there were only 12,000 Indians, including Hindus, Muslims, Sikhs, and
Christians, each group with its own temple, mosque, or church. South
Indians tended to be shopkeepers or laborers, particularly dockworkers,
riverboatmen, and drivers of the ox carts that were the major transport
for goods to and from the port area. North Indians were usually clerks,
traders, and merchants. Both groups came to Singapore expecting to
return to their homeland and were even more transient than the Chinese.
Malays continued to be drawn to Singapore from all over the
archipelago, reaching a population of 36,000 by 1901. Malay traders and
merchants lost out in the commercial competition with Chinese and
Europeans, and most Malay immigrants became small shopkeepers, religious
teachers, policemen, servants, or laborers. The leadership positions in
the Malay-Muslim community went to the Jawi-Peranakan, because of their
facility in English, and to wealthy Arabs. In 1876 the first
Malay-language newspaper of the region, Jawi Peranakan, was
published in Singapore. Several other Malay-language journals supporting
religious reform were begun in the early twentieth century, and
Singapore became a regional focal point for the Islamic revival movement
that swept the Muslim world at that time.
A number of events beginning in the late nineteenth century
strengthened Singapore's position as a major port and industrial center.
When the Suez Canal opened, the Strait of Malacca became the preferred
route to East Asia. Steamships began replacing sailing ships,
necessitating a chain of coaling stations, including Singapore. Most of
the major European steamship companies had established offices in
Singapore by the 1880s. The expansion of colonialism in Southeast Asia
and the opening of Thailand to trade under King Chulalongkorn (Rama V)
brought even more trade to Singapore. The spread of British influence in
Malaya increased the flow of rubber, tin, copra, and sugar through the
island port, and Singapore moved into processing and light
manufacturing, some of which was located on its offshore islands. To
serve the growing American canning industry a tin smelter was built in
1890 on Pulau Brani (pulau means island). Rubber processing
expanded rapidly in response to the demands of the young automobile
industry. Oil storage facilities established on Pulau Bukum made it the
supply center for the region by 1902.
In the early twentieth century, Singapore had expanded its financial
institutions, communications, and infrastructure in order to support its
booming trade and industry. British banks predominated, although by 1905
there were Indian, Australian, American, Chinese, and French-owned banks
as well. Telegraph service from India and Europe reached Singapore in
1870, and telephone service within Singapore was installed in 1879 and
extended to Johore in 1882. The more than sixty European-owned companies in the Straits
Settlements crown colony in the 1870s were largely confined to Singapore
and Penang. Far more prosperous were some of the Chinese firms in
Singapore that were beginning to expand their business links throughout
Asia.
Singapore's port facilities failed to keep up with its commercial
development until the publicly owned Tanjong Pagar Dock Board (renamed
Singapore Harbour Board in 1913) set about replacing old wharves and
warehouses and installing modern machinery and a new graving dock (dry
dock). Trucks gradually replaced ox carts for transporting goods from
the harbor to the town, and by 1909 it was possible to travel from
Singapore to Penang by train and railroad ferry. The Johore Causeway
linked road and rail transportation between Singapore and the peninsula
after 1923.
At the turn of the century, social advancement lagged far behind
economic development in Singapore. While the wealthy enjoyed their
social clubs, sports facilities, mansions, and suburban estates, the
lower classes endured a grim existence marked by poverty, overcrowding,
malnutrition, and disease. Malaria, cholera, and opium addiction were
chiefly responsible for Singapore's mortality rate, which in 1896 was
higher than that of Hong Kong, Ceylon, or India. A 1907 government
commission to investigate the opium problem found that the majority of
opium deaths were among the poor, who were reduced to smoking the dregs
of used opium. Campaigns by missionaries and European-educated Chinese
to ban opium use were successfully opposed by tax farmers and
businessmen. By 1900 the opium tax provided one-half of the revenue of
the colonial government, and both Asian and European businessmen
resisted its replacement with an income tax. As an alternative, the
government in 1910 took over all manufacture and sale of opium, setting
up a factory at Pasir Panjang. Opium sales continued to constitute half
of the government's revenue, but the most dangerous use of the drug had
been curtailed.
Education was generally in a backward state. Most primary schools in
which Malay, Chinese, or Tamil was the medium of instruction were of
poor quality. English-language primary schools were mostly run by
Christian missionaries, and the only secondary education was provided by
Raffles Institution beginning in 1884. In 1902 the government formulated
an Education Code, under which it took responsibility for providing
English-language primary schools; the following year it took over
Raffles Institution. With the support of the Chinese community, the
government opened a medical school in 1905 that had a first class of
twenty-three students. Upgraded to the King Edward Medical College in
1920, the school formed the cornerstone of the future Singapore
University. The affluent of Singapore sent their children to the
English-language schools, which had steadily improved their standards.
The brightest students vied for the Queen's Scholarships, founded in
1889, which provided for university education in Britain for Asian
students. Many prosperous Asian families themselves sent their children
to school in Britain. An English-language education at either the
secondary or university level provided many Asians with the key to
government, professional, or business employment. It also created a bond
among the upper classes of all ethnic groups.
Under the leadership of reformist Chinese, Singapore's Chinese-
language schools were also expanded and modernized at this time. A
scientific curriculum was added to the traditional education in Chinese
classics and Confucian morality. Students from Chinese- language schools
often continued their education in China, where a school for Nanyang
students had been opened in Nanjing in 1907 to prepare them for a role
in Singapore's Chinese community. At the turn of the century, schools
were even established in Singapore for Chinese women, who before that
time had led severely cloistered lives under the domination of their
husbands and mothers-in-law. By 1911 Chinese women were receiving
instruction in Malay, English, Chinese, music, sewing, and cooking.
Malay and Tamil-language primary schools continued to decline, and few
students were able to progress from them to the English-language
secondary level.
Responsibility for Singapore's defense had been a contentious issue
between London and Singapore almost since its founding. The Singapore
merchants resisted any attempts to levy taxes for fortifications and
even objected to paying the cost of maintaining a small garrison on this
island. In 1886 troubles with Russia over Afghanistan and worry over the
Russian navy in the Pacific, prompted the British to begin fortifying
the port area and building new barracks and other military facilities.
The Singapore business community resisted strenuously London's proposal
to double the colony's annual military contribution, insisting that the
island was a critical link in the imperial chain. The colony,
nonetheless, was required to pay a larger sum although slightly less
than originally demanded. The British signed a defensive treaty with
Japan in 1902. The Japanese defeat of the Russian navy in 1905 removed
that threat to Britain's seapower in Asia, thus enabling Britain to
concentrate its navy in its home waters in response to a German naval
buildup.
Singapore essentially sat out World War I. Fear that the island would
be attacked by German; East Asiatic Squadron never materialized.
Singapore's German business community, nonetheless, was rounded up and
interned comfortably at their Teutonia Club. The only incident of the
war period was the mutiny of Singapore's small garrison, the 800 troops
of the Fifth Light Infantry Regiment. The regiment, composed entirely of
Punjabi Muslims, was angered that Britain was at war with Muslim Turkey.
When the regiment was ordered to Hong Kong in February 1915, rumors
spread through the unit that they were actually being sent to fight in
France or Turkey. On the eve of its departure, the regiment mutinied,
killed the officers, and terrorized the town. Within ten days the
rebellion had been put down by a combined force of the Singapore
Volunteer Artillery (a unit of 450 volunteers formed in 1914), police,
Malay troops from Johore, the crews of British, French, Japanese, and
Russian warships in port, and several hundred civilians. After the
mutineers were rounded up, thirty-six were shot in public executions and
the others were imprisoned or sent on active duty elsewhere.
Subseqquently, hard feelings were created in Singapore's Indian
community by a requirement that its members register with the
government. A small British detachment was brought in to garrison the
post for the rest of the war, with the aid of the Singapore Volunteer
Artillery.
Singapore
Singapore - Between the World Wars
Singapore
The Singapore economy experienced much the same roller-coaster effect
that Western economies did in the period between the world wars. A
postwar boom created by rising tin and rubber prices gave way to
recession in late 1920 when prices for both dropped on the world market.
By the mid-1920s, rubber and tin prices had soared again and fortunes
were made overnight. Tan Kah Kee, who had migrated from Xiamen (Amoy) at
age seventeen, reportedly made S$8 million in 1925 in rubber, rice milling, andd
shipping; and Hakka businessman Aw Boon Haw earned the nickname
"Tiger Balm King" for the multimillion-dollar fortune he made
from the production and sale of patent medicines. Although they never
amassed the great fortunes of Singapore's leading Asian businessmen, the
prosperous European community increasingly lived in the style and
comfort afforded by modern conveniences and an abundance of servants.
The Baba Chinese leaders focused their attention on improving
educational opportunities, which meant lobbying for free Englishlanguage
primary schools and more scholarships for English-language secondary
schools. Although English-language schools expanded rapidly, most
educated Straits-born Chinese studied at Chineselanguage schools. Of the
72,000 children in Singapore schools in 1939, 38,000 were in Chinese
schools, 27,000 in English schools, 6,000 in Malay schools, and 1,000 in
Tamil schools.
The Straits-born Chinese increased their share of Singapore's Chinese
population from 25 percent in 1921 to 36 percent in 1931. Chinese
immigration was drastically cut by the Immigration Restriction Ordinance
of 1930, which limited immigration of unskilled male laborers. Put in
force to combat unemployment resulting from the Great Depression, the
ordinance dropped the number of Chinese immigrants from 242,000 in 1930
to 28,000 in 1933. Immigration was further restricted by the Aliens
Ordinance of 1933, which set quotas and charged landing fees for aliens.
Executive Council member Tan Cheng Lock and others bitterly opposed the
policy in the Legislative Council as anti-Chinese.
The administration of the colony continued to be carried out by the
governor and top-level officials of the Malayan Civil Service, posts
that could be held only by "natural-born British subjects of pure
European descent on both sides." The governor continued to consult
with the Legislative Council, which included a handful of wealthy Asian
business and professional leaders, who served as nonofficial members of
the council. The mid-level and technical civil service positions were
open to British subjects of all races. Very few Asians opposed the
system, which gave the official members the majority on the legislative
and executive councils. In the 1930s, Tan agitated unsuccessfully for
direct popular representation and a nonofficial majority for the
legislative council, but most Chinese were satisfied to devote their
attentions to commercial and professional affairs and the growing
interest in nationalism in China.
The sympathies of even the Straits-born Chinese lay with their
homeland in the period between the wars. A Singapore branch of the
Guomindang was active for a few years beginning in 1912, and
Chinaoriented businessmen led boycotts in 1915 against Japanese goods in
response to Japan's Twenty-One Demands against China. These demands were
a set of political and economic ultimatums, which if acceded to, would
have made China a protectorate of Japan. Mass support for Chinese
nationalism became more evident in 1919 when demonstrations, which
turned violent, were staged in Singapore. In the early 1920s, Sun
Yat-sen was successful in convincing Singapore's China-born businessmen
to invest heavily in Chinese industry and to donate large sums of money
for education in China. Tan Kah Kee contributed more than S$4 million
for the founding of Amoy (Xiamen) University in 1924. The Guomindang
also sent teachers and textbooks to Singapore and encouraged the use of
Mandarin (or Guoyu) in Singapore's Chinese schools.
Although Mandarin was not the language of any of Singapore's major
dialect groups, it was considered a unifying factor by the various
Chinese leadership factions of both Singapore and China. Singapore's
first Chinese secondary school, established by Tan in 1919, taught in
Mandarin, as did a growing number of Chinese primary schools. In 1927
the Guomindang increased the number of promising students brought to
China for university education and began a concerted effort to extend
its control over Chinese schools in the Nanyang by supervising their
curriculum and requiring the use of Mandarin. In the late 1920s, the
colonial authorities had become increasingly aware of growing left-wing
politics in the Chinese schools and sought to discourage the use of
Mandarin as required by the Guomindang. By 1935, however, Mandarin had
become the medium of instruction in all of Singapore's Chinese schools.
Following the breakup of the short-lived alliance between the
Guomindang and the Chinese Communist Party (CCP), the communists
established a Nanyang Communist Party in 1928. Outlawed and harassed by
the Singapore police, the party was reorganized in 1930 as the Malayan
Communist Party (
MCP), centered in Singapore. For the remainder of that
year, it had some success in infiltrating teacher and student
organizations and staging student strikes. In early 1931, however, the
seizure by the police of an address book containing information on the
newly organized party and its connections with the Far Eastern Bureau of
the Communist International (
Comintern) in Shanghai, led to arrests and the near
destruction of the CPM by the following year. The Guomindang also had
its problems during this period. The party's membership in Singapore had
expanded rapidly until 1929, when the colonial administration banned the
Singapore branch of the Guomindang and fund-raising for the party in
China. Concerned about the increase of anticolonial propaganda, the
Singapore government censored the vernacular press, severely restricted
immigration, and cut off aid to Chinese and Tamil schools. During the
1930s, attempts by the communists and the Guomindang to organize labor
and lead strikes were also suppressed by the colonial government.
Chinese nationalism and anti-Japanese sentiment in Singapore
increased throughout the 1930s. The fortunes of both the Guomindang and
the MCP rose with invasion of Manchuria by Japan in 1931 and the start
of the Sino-Japanese War in 1937. The CCP and the Guomindang formed a
united front in December 1936 to oppose Japanese aggression. The
Guomindang called upon the Nanyang Chinese for volunteer and financial
support for the Republic of China, which had promulgated a Nationality
Law in 1929, by which it claimed all persons of Chinese descent on the
paternal side as Chinese nationals. Tan Kah Kee headed both the Nanyang
Chinese National Salvation Movement and the Singapore Chinese General
Association for the Relief of Refugees, as well as the fund-raising
efforts for the homeland among the Malayan Chinese. Chinese government
agents used the Singapore Chinese Chamber of Commerce and other local
organizations to organize highly effective boycotts against Japanese
goods. Singapore's Chinese also boycotted Malay or Indian shops selling
Japanese goods, and Chinese merchants who ignored the boycott were
severely punished by extremist groups.
The British authorities struggled vainly to control the tide of
anti-Japanese feeling by forbidding anti-Japanese demonstrations and by
banning importation of anti-Japanese textbooks from China and the
teaching of anti-Japanese slogans and songs in Chinese schools. They
were alarmed at the communist infiltration of the Nanyang Chinese
National Salvation Movement and other Chinese patriotic groups. The
banned MCP claimed a membership of more than 50,000 by early 1940.
Although nominally partners in a united front in opposition to the
Japanese, the MCP and the Guomindang competed for control of such
organizations as the Nanyang Chinese Relief General Association.
Nonetheless, Singapore's Chinese contributed generously to the support
of the Chinese government.
Singapore
Singapore - World War II
Singapore
The British had begun building a naval base at Singapore in 1923,
partly in response to Japan's increasing naval power. A costly and
unpopular project, construction of the base proceeded slowly until the
early 1930s when Japan began moving into Manchuria and northern China. A
major component of the base was completed in March 1938, when the King
George VI Graving Dock was opened; more than 300 meters in length, it
was the largest dry dock in the world at the time. The base, completed
in 1941 and defended by artillery, searchlights, and the newly built
nearby Tengah Airfield, caused Singapore to be ballyhooed in the press
as the "Gibralter of the East." The floating dock, 275 meters
long, was the third largest in the world and could hold 60,000 workers.
The base also contained dry docks, giant cranes, machine shops; and
underground storage for water, fuel, and ammunition. A self-contained
town on the base was built to house 12,000 Asian workers, with cinemas,
hospitals, churches, and seventeen soccer fields. Above-ground tanks
held enough fuel for the entire British navy for six months. The only
thing the giant naval fortress lacked was ships.
The Singapore naval base was built and supplied to sustain a siege
long enough to enable Britain's European-based fleet to reach the area.
By 1940, however, it was clear that the British fleet and armed forces
were fully committed in Europe and the Middle East and could not be
spared to deal with a potential threat in Asia. In the first half of
1941, most Singaporeans were unaffected by the war on the other side of
the world, as they had been in World War I. The main pressure on the
Straits Settlements was the need to produce more rubber and tin for the
Allied war effort. Both the colonial government and British military
command were for the most part convinced of Singapore's impregnability.
Even by late autumn 1941, most Singaporeans and their leaders
remained confident that their island fortress could withstand an attack,
which they assumed would come from the south and from the sea. Heavy
fifteen-inch guns defended the port and the city, and machine-gun
bunkers lined the southern coast. The only local defense forces were the
four battalions of Straits Settlements Volunteer Corps and a small civil
defense organization with units trained as air raid wardens, fire
fighters, medical personnel, and debris removers. Singapore's Asians
were not, by and large, recruited into these organizations, mainly
because the colonial government doubted their loyalty and capability.
The government also went to great lengths to maintain public calm by
making highly optimistic pronouncements and heavily censoring the
Singapore newspapers for negative or alarming news. Journalists' reports
to the outside world were also carefully censored, and, in late 1941,
reports to the British cabinet from colonial officials were still
unrealistically optimistic. If Singaporeans were uneasy, they were
reassured by the arrival at the naval base of the battleship Prince
of Wales, the battle and four destroyers cruiser Repulse,
on December 2. The fast and modern Prince of Wales was the
pride of the British navy, and the Repulse was a veteran
cruiser. Their accompanying aircraft carrier had run aground en route,
however, leaving the warships without benefit of air cover.
The Japanese Malaya Campaign
On December 8, 1941, the Japanese troops of two large convoys, which
had sailed from bases in Hainan and southern Indochina, landed at
Singora (now Songkhla) and Patani in southern Thailand and Kota Baharu
in northern Malaya. One of Japan's top generals and some of its best
trained and most experienced troops were assigned to the Malaya
campaign. By the evening of December 8, 27,000 Japanese troops under the
command of General Yamashita Tomoyuki had established a foothold on the
peninsula and taken the British air base at Kota Baharu. Meanwhile,
Japanese airplanes had begun bombing Singapore. Hoping to intercept any
further landings by the Japanese fleet, the Prince of Wales and
the Repulse headed north, unaware that all British airbases in
northern Malaya were now in Japanese hands. Without air support, the
British ships were easy targets for the Japanese air force, which sunk
them both on December 10.
The main Japanese force moved quickly to the western side of the
peninsula and began sweeping down the single north-south road. The
Japanese divisions were equipped with about 18,000 bicycles. Whenever
the invaders encountered resistance, they detoured through the forests
on bicycles or took to the sea in collapsible boats to outflank the
British troops, encircle them, and cut their supply lines. Penang fell
on December 18, Kuala Lumpur on January 11, 1942, and Malacca on January
15. The Japanese occupied Johore Baharu on January 31, and the last of
the British troops crossed to Singapore, blowing a fifty-meter gap in
the causeway behind them.
Singapore faced Japanese air raids almost daily in the latter half of
January 1942. Fleeing refugees from the peninsula had doubled the
550,000 population of the beleaguered city. More British and
Commonwealth of nations fleets and armed foces were brought to Singapore
during January, but most were poorly trained raw recruits from Australia
and India and inexperienced British troops diverted from the war in the
Middle East. Singapore's Chinese population, which had heard rumors of
the treatment of the Malayan Chinese by the invading Japanese, flocked
to volunteer to help repel the impending invasion. Brought together by
the common enemy, Guomindang and communist groups banded together to
volunteer their services to Governor Shenton Thomas. The governor
authorized the formation of the Chung Kuo Council (China National
Council), headed by Tan Kah Kee, under which thousands volunteered to
construct defense works and to perform other essential services. The
colonial government also reluctantly agreed to the formation of a
Singapore Chinese Anti-Japanese Volunteer Battalion, known as Dalforce
for its commander, Lieutenant Colonel John Dalley of the Federated Malay
States police force. Dalley put his volunteers through a ten-day crash
training course and armed them with basic weapons, including shotguns,
knives, and grenades.
From January 1-8, 1942, the two armies faced each other across the
Johore Strait. The Japanese stepped up their air raids, bombing the
airfields, naval base, and harbor area. Bombs also fell in the
commercial and residential sections of the city, causing great
destruction and killing and wounding many civilians. With their mastery
of the skies, the Japanese could choose the time and place for invasion
and maintain an element of surprise. Yamashita, however, had only 30,000
troops and limited ammunition available to launch against a British
force of about 70,000 armed personnel. As the General Officer Commanding
Malaya, Lieutenant General Arthur E. Percival commanded the defense of
Singapore under the direction of General Archibald Wavell, the newly
appointed commander in chief Far East, who was headquartered in Java.
Percival's orders from British prime minister Winston Churchill through
Wavell called for defending the city to the death, while executing a
scorched-earth policy: "No surrender can be contemplated . . . .
every inch of ground . . . defended, every scrap of material or defences
. . . blown to pieces to prevent capture by the enemy . . . ."
Accordingly, the troops set about the task of destroying the naval base,
now useless without ships, and building defense works along the northern
coast, which lay totally unprotected.
On the night of February 8, using collapsible boats, the Japanese
landed under cover of darkness on the northwest coast of Singapore. By
dawn, despite determined fighting by Australian troops, they had two
divisions with their artillery established on the island. By the next
day the Japanese had seized Tengah Airfield and gained control of the
causeway, which they repaired in four days. The British forces were
plagued by poor communication and coordination, and, despite strong
resistance by Commonwealth troops aided by Dalforce and other Chinese
irregulars, the Japanese took Bukit Timah--the highest point on the
island--on February 11. The British forces fell back to a final
perimeter around the city, stretching from Pasir Panjang to Kallang, as
Yamashita issued an invitation to the British to surrender. On February
13, the Japanese broke through the final perimeter at Pasir Panjang,
putting the whole city within range of their artillery.
As many as 2,000 civilians were killed daily as the Japanese
continued to bomb the city by day and shell it at night. Governor Thomas
cabled London that "there are now one million people within radius
of three miles. Many dead lying in the streets and burial impossible. We
are faced with total deprivation of water, which must result in
pestilence...." On February 13, Percival cabled Wavell for
permission to surrender, hoping to avoid the destruction and carnage
that would result from a house-to-house defense of the city. Churchill
relented and on February 14 gave permission to surrender. On the evening
of February 15, at the Japanese headquarters at the Ford factory in
Bukit Timah, Yamashita accepted Percival's unconditional surrender.
Shonan: Light of the South
The Japanese occupied Singapore from 1942 until 1945. They designated
it the capital of Japan's southern region and renamed it Shonan, meaning
"Light of the South" in Japanese. All European and Australian
prisoners were interned at Changi on the eastern end of the island--the
2,300 civilians at the prison and the more than 15,000 military
personnel at nearby Selarang barracks. The 600 Malay and 45,000 Indian
troops were assembled by the Japanese and urged to transfer their
allegiance to the emperor of Japan. Many refused and were executed,
tortured, imprisoned, or sent as forced laborers to Thailand, Sumatra,
or New Guinea. Under pressure, about 20,000 Indian troops joined the
Japanese-sponsored Indian National Army to fight for India's
independence from the British.
The Asian civilian population watched with shock as their colonial
rulers and supposed protectors were marched off to prison and the
Japanese set about establishing their administration and authority. The
Chinese were to bear the brunt of the occupation, in retribution for
support given by Singapore Chinese to China in its struggle against
Japan. All Chinese males from ages eighteen to fifty were required to
report to registration camps for screening. The Japanese or military
police arrested those alleged to be antiJapanese , meaning those who
were singled out by informers or who were teachers, journalists,
intellectuals, or even former servants of the British. Some were
imprisoned, but most were executed, and estimates of their number range
from 5,000 to 25,000. Many of the leaders of Singapore's anti-Japanese
movement had already escaped, however, and the remnants of Dalforce and
other Chinese irregular units had fled to the peninsula, where they
formed the Malayan People's Anti-Japanese Army.
The harsh treatment by the Japanese in the early days of the
occupation undermined any later efforts to enlist the support of
Singaporeans for the Japanese vision of a Greater East Asia CoProsperity
Sphere, which was to comprise Japan, China, Manchuria, and Southeast
Asia. Singapore's prominent Chinese leaders and businessmen were further
disaffected when the Japanese military command bullied them into raising
a S$10 million "gift" to the Japanese as a symbol of their
cooperation and as reparation for their support for the government of
China in its war against Japan. The Chinese and English schools were
pressured to use Japanese as the medium of instruction. The Malay
schools were allowed to use Malay, which was considered the indigenous
language. The Japanesecontrolled schools concentrated on physical
training and teaching Japanese patriotic songs and propaganda. Most
parents kept their children at home, and total enrollment for all the
schools was never more than 7,000. Although free Japanese language
classes were given at night and bonuses and promotions awarded to those
who learned the language, efforts to replace English and Chinese with
Japanese were generally unsuccessful.
Serious disruption of not only the economy but the whole fabric of
society marked the occupation years in Singapore. Food and essential
materials were in short supply since the entrep�t trade that Singapore
depended on to provide most goods was severely curtailed by the war.
Chinese businessmen collaborated with corrupt Japanese officials to
establish a flourishing black market for most items, which were sold at
outrageous prices. Inflation grew even more rampant as Japanese military
scrip flooded the economy. Speculation, profiteering, bribery, and
corruption were the order of the day, and lawlessness against the
occupation government almost a point of honor.
As the war wound down and Japanese fortunes began to fade, life grew
even more difficult in Shonan. Military prisoners, who suffered
increasing hardship from reduced rations and brutal treatment, were set
to work constructing an airfield at Changi, which was completed in May
1945. Not only prisoners of war but also Singapore's unemployed
civilians were impressed into work gangs for labor on the Burma-Siam
railroad, from which many never returned. As conditions worsened and
news of Japanese defeats filtered in, Singaporeans anxiously awaited
what they feared would be a bloody and protracted fight to reoccupy the
island. Although Japan formally surrendered to the Allies on August 15,
1945, it was not announced in the Singapore press until a week later.
The Japanese military quietly retreated to an internment camp they had
prepared at Jurong. On September 5, Commonwealth troops arrived aboard
British warships, cheered by wildly enthusiastic Singaporeans, who lined
the five-kilometer parade route. A week later, on the steps of the
municipal building, the Japanese military command in Singapore
surrendered to the supreme Allied commander in Southeast Asia, Admiral
Lord Louis Mountbatten.
Singapore
Singapore - Aftermath of War
Singapore
The abrupt end of the war took the British by surprise. Although the
Colonial Office had decided on the formation of a Malayan Union, which
would include all the Malay states, Penang, and Malacca, no detailed
plans had been worked out for the administration of Singapore, which was
to be kept separate and serve as the headquarters of the British
governor general for Southeast Asia. Many former colonial officials and
businessmen opposed the separation of Singapore from peninsular Malaya,
arguing that the two were economically interdependent and to exclude
Singapore would "cut the heart out of Malaya." The Colonial
Office maintained that the separation did not preclude union at some
future date, but that union should not be forced on "communities
with such widely different interests." In September 1945, Singapore
became the headquarters for the British Military Administration (BMA)
under Mountbatten. Although Singaporeans were relieved and happy at the
arrival of the Commonwealth troops, their first-hand witnessing of the
defeat of the British by an Asian power had changed forever the
perspective from which they viewed their colonial overlords.
Economic and Social Recovery
The British returned to find their colonies in sad shape. Food and
medical supplies were dangerously low, partly because shipping was in
total disarray. Allied bombing had taken its toll on Singapore's harbor
facilities, and numerous wrecks blocked the harbor. Electricity, gas,
water, and telephone services were in serious disrepair. Severe
overcrowding had resulted in thousands of squatters living in shanties,
and the death rate was twice the prewar level. Gambling and
prostitution, both legalized under the Japanese, flourished, and for
many opium or alcohol served as an escape from a bleak existence. The
military administration was far from a panacea for all Singapore's ills.
The BMA had its share of corrupt officials who helped the collaborators
and profiteers of the Japanese occupation to continue to prosper. As a
result of the inefficiency and mismanagement of the rice distribution,
the BMA was cynically known as the "Black Market
Administration." However, by April 1946, when military rule was
ended, the BMA had managed to restore gas, water, and electric services
to above their prewar capacity. The port was returned to civilian
control, and seven private industrial, transportation, and mining
companies were given priority in importing badly needed supplies and
materials. Japanese prisoners were used to repair docks and airfields.
The schools were reopened, and by March 62,000 children were enrolled.
By late 1946, Raffles College and the King Edward Medical College both
had reopened.
Food shortages were the most persistent problem; the weekly per
capita rice ration fell to an all-time low in May 1947, and other foods
were in short supply and expensive. Malnutrition and disease spawned
outbreaks of crime and violence. Communist-led strikes caused long work
stoppages in public transport, public services, at the docks, and at
many private firms. The strikers were largely successful in gaining the
higher wages needed by the workers to meet rising food prices.
By late 1947, the economy had began to recover as a result of a
growing worldwide demand for tin and rubber. The following year,
Singapore's rubber production reached an all-time high, and abundant
harvests in neighboring rice-producing countries ended the most serious
food shortages. By 1949 trade, productivity, and social services had
been restored to their prewar levels. In that year a five-year social
welfare plan was adopted, under which benefits were paid to the aged,
unfit, blind, crippled, and to widows with dependent children. Also in
1949, a ten-year plan was launched to expand hospital facilities and
other health services. By 1951 demand for tin and rubber for the war in
Korea had brought economic boom to Singapore.
By the early postwar years, Singapore's population had become less
transitory and better balanced by age and sex. The percentage of Chinese
who were Straits-born rose from 36 percent in 1931 to 60 percent by
1947, and, of those born in China, more than half reported in 1947 that
they had never revisited and did not send remittances there. Singapore's
Indian population increased rapidly in the postwar years as a result of
increased migration from India, which was facing the upheavals of
independence and partition, and from Malaya, where the violence and
hardships of the Emergency caused many to leave. Although large numbers
of Indian men continued to come to Singapore to work and then return to
India, both Indians and Chinese increasingly saw Singapore as their
permanent home.
In 1947 the colonial government inaugurated a ten-year program to
provide all children with six years of primary education in the language
of the parents' choice, including English, Malay, Chinese, and Tamil.
Seeing an English education as offering their children the best
opportunity for advancement, parents increasingly opted to send their
children to English-language schools, which received increased
government funding while support for the vernacular schools declined. In
1949 the University of Malaya was formed through a merger of Raffles
College and the King Edward Medical College.
Political Awakening
The Colonial Office established an advisory council in November 1945
to work with the BMA on the reconstruction of Singapore. Among the
seventeen members appointed to the council was Wu Tian Wang, a former
guerrilla leader and chairman of the communist Singapore City Committee.
The MCP enjoyed great popularity in the early postwar days because of
its association with the resistance and the Malayan People's
Anti-Japanese Army, which also included many noncommunists. In January
1946, the anti-January army was formally disbanded following a final
parade at which Mountbatten presented medals to the guerrilla commander,
Chin Peng, and the other resistance leaders. All arms and ammunition,
which the guerrillas had received in airdrops from the British during
the war or captured from the Japanese, were supposed to be surrendered
at that time. The CPM, however, secretly retained large stocks of its
weapons.
The British legally recognized the MCP in late 1945, largely because
of its resistance efforts and its popularity. The party by that time
commanded about 70,000 supporters. The MCP at first concentrated its
efforts on organizing labor, establishing the General Labour Union,
which covered more than sixty trade unions. It organized numerous
strikes in 1945 and early 1946, including a two-day general strike in
January in which 173,000 workers struck and transport was brought to a
halt. In February, after the formation of a Pan-Malayan Federation of
Trade Unions claiming 450,000 members in Singapore and the peninsula,
the BMA arrested twenty-seven leading communists, banishing ten of them
without trial. Thereafter, the MCP adopted a lower profile of quietly
backing radical groups that were working for constitutional changes and
increasing its control over the labor movement.
In April 1946, the BMA ended with the formation of the Malayan Union,
at which time Singapore became a separate crown colony with a civil
administration. The two entities continued to share a common currency,
institutions of higher learning, and the administration of immigration,
civil aviation, posts and telegraphs, and income tax. Opposition to the
separation of Malaya and Singapore motivated the formation in December
1945 of Singapore's first indigenous political party, the Malayan
Democratic Union (MDU). Although most leaders of the new party were not
communist, there were several prominent communists among its founders,
including Wu Tian Wang, who saw the Malayan Union as a threat to the
vision of a communist, united Malayan republic. The MDU proposed
eventual inclusion of Singapore in an independent Malaya within the
Commonwealth of Nations. Meanwhile, on the peninsula, conservative Malay
leaders, who were concerned about provisions in the Malayan Union scheme
that conferred equal political status on immigrant communities, formed
the United Malays National Organization (UMNO) in March 1946. After
various mass rallies, movements and countermovements, proposals and
counterproposals, the British acceded to UMNO wishes. February 1948
marked the formation of the Federation of Malaya, which provided for the
gradual assimilation of immigrants into a Malay state working toward
independence under British guidance. Singapore remained a separate crown
colony.
Elections in Singapore were scheduled for March 1948, at which time a
new constitution would go into effect. That document called for an
Executive Council of colonial officials and a Legislative Council
comprising nine officials and thirteen nonofficials, four nominated by
the governor, three chosen by the chambers of commerce, and six elected
by adult British subjects who had been resident in Singapore for one
year prior to the election. The appointed governor retained power over
certain items and veto power over the proceedings of the Legislative
Council. The MDU, by then a communist front organization, boycotted the
elections and organized mass rallies opposing the new constitution. The
moderate Progressive Party was formed in August 1947 by British-educated
business and professional men who advocated gradual constitutional
reform aimed at eventual self-government. Of the six elected seats on
the Legislative Council, three were won by independents and three by
Progressives, the only party to contest the elections. In the first
municipal election in 1949, the Progressive Party won thirteen of the
eighteen seats on the twenty-seven member municipal commission. Voter
interest was very low in both elections, however, with only about 10
percent of those eligible registering to vote.
Meanwhile, the MCP had abandoned the moderate stance advocated by its
secretary general Lai Teck, who was replaced in March 1947 by Chin Peng.
Soon after, it was discovered that Lai Teck had not only disappeared
with the party's funds but also been a double agent, serving both the
Japanese and the British. Following the establishment of the Federation
of Malaya in February 1948, Singapore's communist leaders moved to the
peninsula where they reactivated the MPAJA and began fomenting acts of
violence and terrorism. This led to the declaration of a State of
Emergency in Malaya on June 18 and in Singapore a week later. Although
the twelve-year struggle was largely confined to the peninsula,
restrictions were placed on meetings and strikes, and the detention of
individuals without trial was permitted under the Emergency regulations.
The MCP was proscribed by the colonial government in Singapore, and the
MDU, fearing the same fate, voluntarily dissolved itself. Left-wing
political movements were thus stifled, and the only political party that
arose to challenge the Progressives was the Singapore Labour Party
formed in 1948. Like the Progressive Party, its positions were moderate
and its leadership mostly British educated. Nevertheless, as a result of
personal squabbles and factions, the Singapore Labour Party had largely
disintegrated by 1952.
The number of elected seats in the Legislative Council was increased
to nine in 1951, and the Progressive Party won six of the nine seats in
the election that year. The membership of the party never numbered more
than about 4,000, the majority of whom were upper or middle class and
British educated. The interests of the members of the Legislative
Council and the leadership of the Progressive Party were so closely
aligned with those of the colonial government that they were out of
touch with the masses. Participation in politics was restricted to
Straits-born or naturalized British subjects who were literate in
English. This exclusion of immigrants and those not educated in English
meant that, in the late 1940s, about one-half of Singapore's adult
population was disenfranchised.
Although the Chinese-educated took little interest in the affairs of
the Legislative Council and the colonial government, they were stirred
with pride by the success of the CCP in China. Fearful that support by
Singapore's Chinese for the CCP would translate to support for the MCP,
the colonial government attempted to curtail contacts between the
Singapore Chinese and their homeland. When Tan Kah Kee returned from a
trip to China in 1950, the colonial government refused to readmit him,
and he lived out his days in his native Fujian Province.
For graduates of Singapore's Chinese high schools, there were no
opportunities for higher education in the colony. Many went to
universities in China, despite the fact that immigration laws prohibited
them from returning to Singapore. To alleviate this problem, wealthy
rubber merchant and industrialist Tan Lark Sye proposed formation of a
Chinese-language university for the Chinese-educated students of
Singapore, Malaya, and all Southeast Asia. Singaporean Chinese, rich and
poor, donated funds to found Nanyang University, which was opened in
Singapore in 1956.
By the early 1950s, large numbers of young men whose education had
been postponed by the Japanese occupation were studying at
Chinese-language high schools. These older students were particularly
critical of the colonial government's restrictive policies toward
Chinese and of its lack of support for Chinese- language schools. The
teachers in these schools were poorly paid, the educational standards
were low, and graduates of the schools found they could not get jobs in
the civil service or gain entrance to Singapore's English-language
universities. While critical of the colonial government, the students
were becoming increasingly proud of the success of the communist
revolution in China, reading with interest the publications and
propaganda put out by the new regime.
As the Emergency on the peninsula began to go badly for the
communists, the MCP took a renewed interest in Singapore and began
organizing protest demonstrations among the disaffected students. Among
the brightest and most capable of the older Chinese high school students
were Lim Chin Siong and Fong Swee Suan, who both became involved in
organizing class boycotts that resulted in a police raid on the Chinese
High School in 1952. The two left the school, took low-paying jobs at
bus companies, and began working to build communist influence among
workers and students. In May 1954, mass student protest demonstrations
were organized to oppose a new National Service Ordinance requiring
males between the ages of eighteen and twenty to register for part-time
national service. Also in May, the Singapore Factory and Shop Workers'
Union registered with the government, with Lim as its secretary general;
Fong, who was by then general secretary of the Singapore Bus Workers'
Union, and C.V. Devan Nair, of (at that time) the Singapore Teachers'
Union, were members of the executive board. Dedicated and charismatic,
Lim led several well-organized small strikes that were successful in
gaining better conditions for the union's workers, and in attracting
thousands of recruits for the union. By late 1955, the Singapore Factory
and Shop Workers' Union included thirty industrial unions and had a
membership of about 30,000.
Singapore
Singapore - Road to Independence
Singapore
In 1953 the colonial government appointed Sir George Rendel to head a
commission to review the Singapore constitution and devise a
"complete political and constitutional structure designed to enable
Singapore to develop as a self-contained and autonomous unit in any
larger organization with which it may ultimately become
associated." The commission recommended partial internal
selfgovernment for Singapore, with Britain retaining control of internal
security, law, finance, defense, and foreign affairs. It also proposed a
single-chamber Legislative Assembly of thirty-two members, twenty-five
of whom would be elected, and a nine-member council of ministers that
would act as a cabinet. The governor retained his power to veto
legislation. The British government accepted the commission's
recommendations, and the Rendel constitution went into effect in
February 1954, with elections scheduled for the Legislative Assembly for
April 1955. Voters were to be automatically registered, which was
predicted to greatly enlarge the size of the turnout over previous
elections. Although the new constitution was a long way from offering
Singapore full independence, election fever gripped the country as new
political alliances and parties were formed.
Two former members of the Singapore Labour Party, Lim Yew Hock and
Francis Thomas, and a prominent lawyer, David Marshall, formed a new
political party, the Labour Front, in July 1954. Marshall, who was a
member of Singapore's small Jewish community, had studied law in
Britain, fought with the Singapore Volunteer Corps during the Japanese
invasion, and worked in the coal mines of Hokkaido as a prisoner of war.
Under the leadership of Marshall, a staunch anticolonialist, the party
campaigned for immediate independence within a merged Singapore and
Malaya, abolishing the Emergency regulations, Malayanization of the
civil service within four years (by which time local officials would
take over from colonial officials), multiligualism, and Singapore
citizenship for its 220,000 China-born inhabitants. Marshall, a powerful
speaker, promised "dynamic socialism" to counter "the
creeping paralysis of communism" as he denounced colonialism for
its exploitation of the masses.
People's Action Party
In November 1954, the People's Action Party ( PAP) was inaugurated at
a gathering of 1,500 people in Victoria Memorial Hall. The party was
formed by a group of British-educated, middle- class Chinese who had
returned to Singapore in the early 1950s after studying in Britain. Led
by twenty-five-year-old Lee Kuan Yew, as secretary general, Toh Chin
Chye, Goh Keng Swee, and S. Rajaratnam, the party sought to attract a
following among the mostly poor and non-English-speaking masses. Lee had
served as a legal adviser to a number of trade unions and, by 1952, had
earned a reputation for his successful defense of the rights of workers.
He also helped defend Chinese students arrested during the 1954 student
demonstrations protesting national service. Lee, a fourth- generation
Singaporean, was educated at Raffles Institution and Cambridge
University, where he took a double first (first-class honors in two
subjects) in law. Through his work with the unions and student groups,
Lee had made many contacts with anticolonialists, noncommunists and
communists alike.
Present at the inauguration of the PAP were a number of noted
communists and procommunists, including Fong Swee Suan and Devan Nair,
who both joined the new party. Also present were Malayan political
leaders Tunku Abdul Rahman, president of UMNO, and Sir Tan Cheng Lock,
president of the Malayan Chinese Association (MCA). The PAP proposed to
campaign for repeal of the Emergency regulations, union with Malaya, a
common Malayan citizenship, Malayanization of the civil service, and
free compulsory education. Ending colonialism, however, was the first
priority of Lee and the PAP leadership, although they concluded this
could be accomplished only with support from the Chinese-educated public
and the communist-controlled trade unions. The PAP, calculating that a
united front with the communists was necessary to end colonialism,
declared itself noncommunist, neither pro- nor anticommunist, preferring
to put off until after independence any showdown with the communists.
Meanwhile, two other political parties prepared to contest the
upcoming election. The Progressive Party, whose leaders had earned a
reputation as the "Queen's Chinese" for their procolonial
positions and conservative economic policies, had little appeal for the
masses of working-class Chinese who were newly enfranchised to vote in
the 1955 election. Automatic registration of voters had increased the
electorate from 76,000 in 1951 to more than 300,000. Shortly before the
elections, wealthy and influential members of the Chinese Chamber of
Commerce formed the new Democratic Party, which championed the causes of
improved Chinese education, establishment of Chinese as an official
language, and liberal citizenship terms for the China-born. Although
these issues appealed to Singapore's China-born lower classes, this same
group was disenchanted with the party's conservative economic platform,
which closely resembled that of the Progressive Party.
Election fever gripped Singapore during the month-long campaign, and
the results of the April 2 contest sent shock waves as far as Britain,
where it had been expected that the Progressive Party would win handily.
Surprising even itself, the Labour Front won ten of the twenty-five
seats and formed a coalition government with the UMNO-MCA Alliance,
which won three seats. Three ex-officio members and two nominated
members joined with the coalition, forming a group of seventeen in the
thirty-two-member assembly. The Progressives won only four seats and the
Democratic Party just two, in a clear rejection of colonial rule and
procolonial politics. The PAP won three of the four seats it had
contested, including a seat in one of Singapore's poorest sections won
by Lee Kuan Yew and one seat won by Lim Chin Siong. Lim had the backing
of organized labor and led the procommunist wing of the party while Lee
led the noncommunist wing.
The Labour Front government, with David Marshall as Singapore's first
chief minister, faced serious problems from the start. The communists
launched a campaign of strikes and student unrest in an attempt to
destabilize the government. Only about one-third of the 275 strikes
called in 1955 were for better wages and working conditions; the
remainder were sympathy strikes or strikes to protest imprisonment of
labor union officials. Riots broke out on May 12 when police attempted
to break up an illegal picket line formed by striking bus workers and
Chinese school students. Four people were killed and thirty-one injured
in that single incident, which became known as "Black
Thursday." Although the government arrested some students, Marshall
eventually backed down and agreed to the registration of the Singapore
Chinese Middle School Students' Union because he was in sympathy with
the students' grievances against the colonial education system. In
registering their union, the students agreed to the condition that the
union keep out of politics; the communist leaders of the union, however,
had no intention of keeping the agreement.
Along with problems with labor and students, Marshall faced constant
conflict with the colonial government over his determination not to be a
figurehead controlled by the governor. When the governor, Sir Robert
Black, refused to allow Marshall to appoint four assistant ministers,
Marshall threatened to resign unless Singapore was given immediate
self-government under a new constitution. The Colonial Office agreed to
hold constitutional talks, which came to be known as Merdeka
(freedom in Malay) talks, in London in April 1956. Marshall led to the
talks a thirteen-man delegation comprising members of all the
legislative parties and including Lee and Lim Chin Siong. The British
offered to grant Singapore full internal self-government but wanted to
retain control over foreign affairs and internal security. They proposed
a Defence and Internal Security Council, with three delegates each from
Britain and Singapore, to be chaired by the British high commissioner in
Singagore, who would have the casting ballot (the deciding vote in case
of a tie). Marshall had promised he would resign if he failed to obtain
internal self-government, and the talks broke down over the issue of the
casting ballot. The delegation returned to Singapore, and Marshall
resigned in June and was succeeded by the deputy chief minister, Lim Yew
Hock.
By July the Singapore Chinese Middle Schools Students' Union had
begun planning a campaign of agitation against the government. The Lim
Yew Hock government moved first, however, dissolving seven
communist-front organizations, including the student union, and closing
two Chinese middle schools. This touched off a protest sit- in at
Chinese high schools organized by Lim Chin Siong that ended in five days
of rioting in which thirteen people were killed. Troops were brought in
from Johore to end the disturbance, and more than 900 people were
arrested, including Lim Chin Siong, Fong Swee Suan, and Devan Nair. The
British approved of the Singapore government's tough action toward the
agitators, and when Lim Yew Hock led a delegation to London for a second
round of constitutional talks in March 1957, the Colonial Office
proposed a compromise on the internal security issue. The Singapore
delegation accepted a proposal whereby the Internal Security Council
would comprise three Singaporeans, three Britons, and one delegate from
what was soon to be the independent Federation of Malaya, who would hold
the casting ballot. The Singapore delegation returned to a hero's
welcome; the Legislative Assembly accepted the proposals, and a
delegation was scheduled to go to London in 1958 for a third and final
round of talks on the new constitution.
Although the moderates led by Lee Kuan Yew retained control of the
PAP Central Executive Committee, by 1956 the procommunists held sway
over the membership and many of the mass organizations and PAP branches.
At the annual general meeting in August 1957, the procommunists won six
of the twelve seats on the committee. Lee Kuan Yew and the other
moderates refused to take office in order to avoid becoming front men
for the leftists. On August 21, the Lim Yew Hock government reacted to
the situation by arresting thirty- five communists, including five of
the new members of the PAP Central Executive Committee, some PAP branch
officials, and labor and student leaders. Lee and the moderates were
able to regain control of the party and, the following November, amended
the party's constitution to consolidate moderate control by limiting
voting for the central executive committee to the full cadres (full
members), who were literate Singapore citizens over the age of
twenty-one who had been approved as cadres by the central executive
committee.
Meanwhile, the Lim Yew Hock government continued to make further
progress on issues related to Singapore's self-government. A Citizenship
Ordinance passed in 1957 provided Singapore citizenship for all born in
Singapore or the Federation of Malaya and for British citizens of two
years' residence; naturalization was offered to those who had resided in
Singapore for ten years and would swear loyalty to the government. The
Legislative Assembly voted to complete Malayanization of the civil
service within four years beginning in 1957. The Education Ordinance
passed in 1957 gave parity to the four main languages, English, Chinese,
Malay, and Tamil. By 1958 the Ministry of Education had opened nearly
100 new elementary schools, 11 new secondary schools, and a polytechnic
school and set up training courses for Malay and Tamil teachers.
Lim Yew Hock led the Singapore delegation to the third round of
constitutional talks in April 1958. The talks resulted in an agreement
on a constitution for a State of Singapore with full powers of internal
government. Britain retained control over foreign affairs and external
defense, with internal security left in the hands of the Internal
Security Council. Only in the case of dire emergency could Britain
suspend the constitution and assume power. In August 1958, the British
Parliament changed the status of Singapore from a colony to a state, and
elections for the fifty- one-member Legislative Assembly were scheduled
for May 1959. Voting was made compulsory for all adult Singapore
citizens, but the British refused to allow persons with records of
subversive activity to stand for election. Ten parties contested the
election, but none was as well organized as the PAP, which under Lee
Kuan Yew ran a vigorous campaign with huge weekly rallies. Campaigning
on a platform of honest efficient government, social and economic
reform, and union with the Federation of Malaya, the PAP scored a
stunning victory by winning forty-three of the fifty-one seats. The
badly divided and scandal-ridden Labour Front had reorganized as the
Singapore People's Alliance, which won four seats, including one for Lim
Yew Hock. The remaining seats were won by three UMNO- MCA Alliance
candidates and one independent. Marshall's Workers' Party failed to win
any seats.
Both foreign and local businesses feared that the PAP victory
signaled Singapore's slide toward communism, and many moved their
headquarters to Kuala Lumpur. Lee indeed refused to take office until
the eight procommunist PAP detainees arrested in 1956 and 1957 were
released, and he appointed several of them, including Lim Chin Siong,
Fong Swee Suan, and Devan Nair, to government posts. Lee's closest
advisors, however, were moderates Goh Keng Swee, Toh Chin Chye, and S.
Rajaratnam.
The first task of the new PAP government was to instill a sense of
unity and loyalty in Singapore's diverse ethnic populace. A new national
flag, crest, and anthem were introduced, and the new Ministry of Culture
organized open-air cultural concerts and other events designed to bring
the three main ethnic groups together. Malay, Chinese, Tamil, and
English were all made official languages, but, with its eye on a future
merger with Malaya, the government made Malay the national language.
Considered the indigenous people and yet the most disadvantaged, Malays
were provided with free primary and secondary education.
After national unity, the second most important task facing the new
government was that of transforming Singapore from an entrep�t economy
dependent on the Malayan commodity trade with no tradition of
manufacturing to an industrialized society. A four-year development
plan, launched under Minister of Finance Goh Keng Swee in 1961, provided
foreign and local investors with such incentives as low taxation rates
for export-oriented manufactures, tax holidays for pioneer industries,
and temporary protective tariffs against imports. The plan set aside a
large area of swamp wasteland as an industrial estate in the Jurong area
and emphasized labor- intensive industries, such as textiles. The
overhaul of Singapore's economy was urgently needed in order to combat
unemployment and pay for badly needed social services. One of the most
serious problems was the lack of adequate housing. In 1960 the Housing
and Development Board (HDB) was set up to deal with the problems of slum
clearance and resettlement. Under the direction of the banker and
industrialist Lim Kim San, the HDB constructed more than 20,000 housing
units in its first three years. By 1963 government expenditures on
education had risen to S$10 million from S$600,000 in 1960.
Despite the signs of economic progress, the PAP leaders believed that
Singapore's survival depended on merger with Malaysia. "Major
changes in our economy are only possible if Singapore and the Federation
are integrated as one economy," remarked Goh Keng Swee in 1960.
"Nobody in his senses believes that Singapore alone, in isolation,
can be independent," stated an official government publication that
same year. The procommunists within the party, however, opposed merger
because they saw little chance of establishing a procommunist government
in Singapore as long as Kuala Lumpur controlled internal security in the
new state. Meanwhile, the leaders of the conservative UMNO government in
Kuala Lumpur, led by Tengku Abdul Rahman, were becoming increasingly
resistant to any merger with Singapore under the PAP, which they
considered to be extremely left wing.
Moreover, Malayan leaders feared merger with Singapore because it
would result in a Chinese majority in the new state. When a fiercely
contested Singapore by-election in April 1961 threatened to bring down
the Lee Kuan Yew government, however, Tengku Abdul Rahman was forced to
consider the possibility that the PAP might be replaced with a
procommunist government, a "Cuba across the causeway."
Accordingly, on May 27, 1961, in a speech in Kuala Lumpur to the
Foreign Correspondents' Association, Tengku Abdul Rahman made a surprise
proposal of an association of states that would include the Federation
of Malaya, the British Borneo territories, and Singapore. In this
proposed Malaysia, the Malay population of Sarawak and North Borneo (now
Sabah) would offset numerically the Singapore Chinese, and the problem
of a possible "Cuba across the causeway" would be solved.
The proposal, however, led almost immediately to a split between the
moderate and procommunist forces within the PAP. In July Lee demanded
and received a vote of confidence on the issue of merger from the
Legislative Assembly. Following the vote, Lee expelled sixteen rebel PAP
assemblymen from the party along with more than twenty local officials
of PAP. In August the rebel PAP assemblymen formed a new opposition
party, the Barisan
Sosialis (The Socialist Front) with Lim Chin Siong as
secretary general. The new party had considerable support among PAP
local officials as well as at the grass-roots level. Of the fifty- one
branch committees, thirty-five defected to Barisan, which also
controlled two-thirds of organized labor.
The battle lines were clearly drawn when Lee Kuan Yew announced a
referendum on the question of merger to be held in September 1962. Lee
launched a campaign of thirty-six radio broadcasts in three languages to
gain support for the merger, which was opposed by the Barisan Sosialis
as a "sell-out." Of the three merger plans offered on the
referendum, the PAP plan received 70 percent of the votes, the two other
plans less than 2 percent each, and 26 percent of the ballots were left
blank.
Having failed to stop the merger at home, the Barisan Sosialis turned
its efforts abroad, joining with left-wing opposition parties in Malaya,
Sarawak, Brunei, and Indonesia. These parties were opposed to the
concept of Malaysia as a "neocolonialist plot," whereby the
British would retain power in the region. President Sukarno of
Indonesia, who had entertained dreams of the eventual establishment of
an Indonesia Raya (Greater Indonesia) comprising Indonesia, Borneo, and
Malaya, also opposed the merger; and in January 1963 he announced a
policy of Confrontation (Konfrontasi) against the proposed new state.
The Philippines, having revived an old claim to Sabah, also opposed the
formation of Malaysia. The foreign ministers of Malaya, Indonesia, and
the Philippines met in June 1963 in an attempt to work out some
solution. Malaya agreed to allow the United Nations (UN) to survey the
people of Sabah and Sarawak on the issue, although it refused to be
bound by the outcome. Brunei opted not to join Malaysia because it was
unable to reach agreement with Kuala Lumpur on the questions of federal
taxation of Brunei's oil revenue and of the sultan of Brunei's relation
to the other Malay sultans.
Singapore as Part of Malaysia
The leaders of Singapore, Malaya, Sabah, and Sarawak signed the
Malaysia Agreement on July 9, 1963, under which the Federation of
Malaysia was scheduled to come into being on August 31. Tengku Abdul
Rahman changed the date to September 16, however, to allow the UN time
to complete its survey. On August 31, Lee declared Singapore to be
independent with the PAP government to act as trustees for fifteen days
until the formation of Malaysia on September 16. On September 3, Lee
dissolved the Legislative Assembly and called for a new election on
September 21, to obtain a new mandate for the PAP government. In a
bitterly contested campaign, the Barisan Sosialis denounced the merger
as a "sell-out" and pledged increased support for Chinese
education and culture. About half of Barisan's Central Executive
Committee, including Lim Chin Siong, were in jail, however, following
mass arrests the previous February by the Internal Security Council of
political, labor, and student leaders who had supported a rebellion in
Brunei. The mass arrests, although undertaken by the British and
Malayans, benefited the PAP because there was less opposition. The party
campaigned on its economic and social achievements and the achievement
of merger. Lee visited every corner of the island in search of votes,
and the PAP won thirty-seven of the fifty-one seats while the Barisan
Sosialis won only thirteen.
On September 14, the UN mission had reported that the majority of the
peoples of Sabah and Sarawak were in favor of joining Malaysia. Sukarno
immediately broke off diplomatic and trade relations between Indonesia
and Malaysia, and Indonesia intensified its Confrontation operations.
Singapore was particularly hard hit by the loss of its Indonesian barter
trade. Indonesian commandos conducted armed raids into Sabah and
Sarawak, and Singaporean fishing boats were seized by Indonesian
gunboats. Indonesian terrorists bombed the Ambassador Hotel on September
24, beginning a year of terrorism and propaganda aimed at creating
communal unrest in Singapore. The propaganda campaign was effective
among Singapore Malays who had hoped that merger with Malaysia would
bring them the same preferences in employment and obtaining business
licenses that were given Malays in the Federation. When the PAP
government refused to grant any economic advantages other than financial
aid for education, extremist UMNO leaders from Kuala Lumpur and the
Malay press whipped up antigovernment sentiment and racial and religious
tension. On July 21, 1964, fighting between Malay and Chinese youths
during a Muslim procession celebrating the Prophet Muhammad's birthday
erupted into racial riots, in which twenty-three people were killed and
hundreds injured. In September Indonesian agents provoked communal
violence in which 12 people were killed and 100 were injured. In
Singapore, which normally prided itself on the peace and harmony among
its various ethnic groups, shock and disbelief followed in the wake of
the violence. Both Lee Kuan Yew and Tengku Abdul Rahman toured the
island in an effort to restore calm, and they agreed to avoid wrangling
over sensitive issues for two years.
The first year of merger was also disappointing for Singapore in the
financial arena. No progress was made toward establishing a common
market, which the four parties had agreed would take place over a
twelve-year period in return for Singapore's making a substantial
development loan to Sabah and Sarawak. Each side accused the other of
delaying on carrying out the terms of the agreement. In December 1964,
Kuala Lumpur demanded a higher percentage of Singapore's revenue in
order to meet defense expenditures incurred fighting Confrontation and
also threatened to close the Singapore branch of the Bank of China,
which handled the financial arrangements for trade between Singapore and
China as well as remittances.
Political tensions between Singapore and Kuala Lumpur also escalated
as each began getting involved in the politics of the other. UMNO ran
candidates in Singapore's September 1963 elections, and PAP challenged
MCA Alliance candidates in the Malaysian general election in April 1964.
UMNO was unable to win any seats in the Singapore election, and PAP won
only one seat on the peninsula. The main result was increased suspicion
and animosity between UMNO and PAP and their respective leaders. In
April 1965, the four Alliance parties of Malaya, Singapore, Sabah, and
Sarawak merged to form a Malaysian National Alliance Party. The
following month, the PAP and four opposition parties from Malaya and
Sarawak formed the Malaysian Solidarity Convention, most of whose
members were ethnic Chinese. Although the Malaysian Solidarity
Convention claimed to be noncommunal, right-wing UMNO leaders saw it as
a Chinese plot to take over control of Malaysia. In the following
months, the situation worsened increasingly, with abusive speeches and
writings on both sides. Faced with demands for the arrest of Lee Kuan
Yew and other PAP leaders by UMNO extremists, and fearing further
outbreaks of communal violence, Tengku Abdul Rahman decided to separate
Singapore from Malaysia. Informed of his decision on August 6, Lee tried
to work out some sort of compromise, without success. On August 9, with
the Singapore delegates not attending, the Malaysian parliament passed a
bill favoring separation 126 to 0. That afternoon, in a televised press
conference, Lee declared Singapore a sovereign, democratic, and
independent state. In tears he told his audience, "For me, it is a
moment of anguish. All my life, my whole adult life, I have believed in
merger and unity of the two territories."
Singapore
Singapore - Two Decades of Independence
Singapore
Reaction to the sudden turn of events was mixed. Singapore's
political leaders, most of whom were Malayan-born and still had ties
there, had devoted their careers to winning independence for a united
Singapore and Malaya. Although apprehensive about the future, most
Singaporeans, however, were relieved that independence would probably
bring an end to the communal strife and riots of the previous two years.
Moreover, many Singaporean businessmen looked forward to being free of
Kuala Lumpur's economic restrictions. Nonetheless, most continued to
worry about the viability as a nation of a tiny island with no natural
resources or adequate water supply, a population of nearly 2 million,
and no defense capability of its own in the face of a military
confrontation with a powerful neighboring country. Singaporeans and
their leaders, however, rose to the occasion.
Under Lee Kuan Yew
The Lee Kuan Yew government announced two days after separation that
Singapore would be a republic, with Malay as its national language and
Malay, Chinese, English, and Tamil retained as official languages. The
Legislative Assembly was renamed the Parliament, and the prominent Malay
leader, Yusof bin Ishak, was made president of the republic. The new
nation, immediately recognized by Britain, Australia, New Zealand, and
the United States, was admitted to the UN in September and the
Commonwealth the following month. In the early months following
separation, Singapore's leaders continued to talk of eventual reunion
with Malaysia. Wrangling between Singapore and Kuala Lumpur over
conflicting economic, defense, and foreign policies, however, soon put
an end to this discussion, and Singapore's leaders turned their
attention to building an independent nation.
The government sought to build a multiracial and multilingual society
that would be unified by a sense of a unique "Singaporean
identity." The government established a Constitutional Commission
on Minority Rights in late 1965, and official policy encouraged ethnic
and cultural diversity. Foreign Minister Rajaratnam told the UN General
Assembly that year, "If we of the present generation can
steadfastly stick to this policy for the next thirty years, then we
would have succeeded in creating a Singaporean of a unique kind. He
would be a man rooted in the cultures of four great civilizations but
not belonging exclusively to any of them." Integrated schools and
public housing were the principle means used by the government to ensure
a mixing of the various ethnic groups. The government constructed modern
highrise housing estates and new towns, in which the residents of the
city's crowded Chinatown slums and the rural Malay kampongs (villages in
Malay were thoroughly intermingled. An English-language education
continued to be the preferred preparation for careers in business,
industry, and government; English-language pupils outnumbered
Chinese-language pupils 300,000 to 130,000 by 1968. Malay-language
primary school enrollment declined from 5,000 in 1966 to about 2,000 in
1969. All students, however, were required to study their mother tongue
at least as a second language. Many of the country's British-educated
leaders, including Lee Kuan Yew, sent their children to Chinese-language
schools because they believed that they provided better character
training. The government stressed discipline and the necessity of
building a "rugged society" in order to face the challenges of
nationhood. A government anticorruption campaign was highly effective in
combating that problem at all levels of administration.
At the same time, the government addressed the problem of
establishing a national identity, it also tackled the serious economic
problems facing the new nation. The hopes pinned on establishing a
common market with Malaysia were dead, and it was clear that Singapore
would not only have to go it alone but also would face rising tariffs
and other barriers to trade with Malaysia. Under Goh Keng Swee and other
able finance ministers, the government worked hard to woo local and
foreign capital. New financial inducements were provided to attract
export industries, promote trade, and end the country's dependence on
Britain as the major source of investment capital. The generally
prosperous world economic situation in the mid-1960s favored Singapore's
growth and development. Confrontation with Indonesia had ended by 1966
after Soeharto came to power, and trade between the two countries
resumed. Trade with Japan and the United States increased substantially,
especially with the latter as Singapore became a supply center for the
United States in its increasing involvement in Indochina.
A serious problem the government had to deal with in order to attract
large-scale investment was Singapore's reputation for labor disputes and
strikes. "The excesses of irresponsible trade unions...are luxuries
which we can no longer afford," stated President Yusof bin Ishak in
December 1965, speaking for the government. Two events in 1968 enabled
the government to pass stricter labor legislation. In January Britain
announced its intention to withdraw from its bases in Singapore within
three years. Aside from the defense implications, the news was sobering
because British spending in Singapore accounted for about 25 percent of
Singapore's gross national product (
GNP) for a total of about S$450 million a year, and
the bases employed some 21,000 Singapore citizens. The government called
an election for April in order to gain a new mandate for facing the
crisis. Unopposed in all but seven constituencies, the PAP made a clean
sweep, winning all fifty-eight parliamentary seats. With the new
mandate, the government passed in August new labor laws that were tough
on workers and employers alike. The new legislation permitted longer
working hours, reduced holidays, and gave employers more power over
hiring, firing, and promoting workers. Workers could appeal actions they
considered unjust to the Ministry of Labour, and employers were
obligated to increase their contributions to the Central Provident Fund
(CPF). Workers also were given for the first time sick leave and
unemployment compensation. As a result of the new legislation,
productivity increased, and there were no strikes in 1969.
With labor relations under control, the government set up the Jurong
Town Corporation to develop Jurong and the other industrial estates. By late 1970, 271 factories in
Jurong employed 32,000 workers, and there were more than 100 factories
under construction. Foreign investors were attracted by the improved
labor situation and by such incentives as tax relief for up to five
years and unrestricted repatriation of profits and capital in certain
government-favored industries. United States firms flocked to invest in
Singapore, accounting for 46 percent of new foreign capital invested in
1972. Companies from Western Europe, Japan, Hong Kong, Taiwan, Malaysia,
and Australia also invested capital, and by 1972 one quarter of
Singapore's manufacturing firms were either foreign-owned or
joint-venture companies. Another attraction of Singapore for foreign
capital was the region's petroleum resources. Singapore was the natural
base for dozens of exploration, engineering, diving, and other support
companies for the petroleum industry in nearby Indonesia, as well as
being the oil storage center for the region. By the mid-1970s, Singapore
was the third largest oil-refining center in the world.
The government turned to advantage the British pullout by converting
some of the military facilities to commercial and industrial purposes
and retraining laid-off workers for new jobs. The former King George VI
Graving Dock was converted to the Sembawang Shipyard, employing 3,000
former naval base workers in ship building and ship repair. Singapore
also moved into shipping in 1968 with its own Neptune Orient Line. A
container complex built in 1972 made the country the container
transshipment center of Southeast Asia. By 1975 Singapore was the
world's third busiest port behind Rotterdam and New York.
By the early 1970s, Singapore not only had nearly full employment but
also faced labor shortages in some areas. As a result, immigration laws
and work permit requirements were relaxed somewhat, and by 1972
immigrant workers made up 12 percent of the labor force. In order to develop a more highly skilled work
force that could command higher wages, the government successfully
courted high-technology industries, which provided training in the
advanced skills required. Concerned that the country's economic success
not be diluted by overpopulation, the government launched a family
planning program in 1966.
The country's economic success and domestic tranquility, which
contrasted so starkly with the impoverished strife-torn Singapore of the
late 1940s, was not purchased without cost, however. Although not a
one-party state, the government was virtually under the total control of
the PAP, and the Lee Kuan Yew administration did not hesitate to block
the rise of an effective opposition. Holding a monopoly on power and
opportunity in a small state, the party could easily co-opt the willing
and suppress dissenters. The traditional bases--student and labor
organizations--used by opposition groups in the past were tightly
circumscribed. Control of the broadcast media was in the hands of the
government, and economic pressures were applied to any newspapers that
became too critical. The government leadership had adopted a
paternalistic viewpoint that only those who had brought the nation
through the perilous years could be trusted to make the decisions that
would keep Singapore on the narrow path of stability and prosperity. The
majority of Singaporeans scarcely dissented from this view and left the
planning and decision making to the political leadership. Although five
opposition parties contested the 1972 elections and won nearly one-third
of the popular vote, the PAP again won all of the seats.
Although admired for its success, Lee's government increasingly
attracted criticism from the international press for its less than
democratic style. Singapore's neighbors also resented the survival-
oriented nature of the country's foreign and economic policies. The
aggressive defense policy recommended by Singapore's Israeli military
advisers irritated and alarmed Muslim Indonesia and Malaysia. Resentful
of the profits made by Singapore in handling their commodities, Malaysia
and Indonesia began setting up their own rubber-milling and
petroleum-servicing industries. In the early 1970s, Malaysia and
Singapore separated their joint currency, stock exchange, and airlines.
A regional political grouping, the Association of Southeast Asian
Nations ( ASEAN), founded in 1967 by Singapore, Malaysia, Indonesia,
Thailand, and the Philippines, had little impact by the early 1970s on
the foreign and economic policies of the member nations. However,
regional and world developments in the 1970s, including the fall of
Indochina to communism and the Vietnamese invasion of Cambodia, steered
Singapore and its neighbors toward a new spirit of cooperation.
Toward New Leadership
Singapore successfully pursued its foreign policy goal of improved
relations with Malaysia and Indonesia in the early 1980s as Lee Kuan Yew
established cordial and productive personal relations with both Soeharto
and Malaysian prime minister Mahathir Mohamad. Cooperation agreements
were reached between Singapore and Malaysia on joint civil service and
military training programs. The economic interdependence of the two
countries was reaffirmed as Singapore continued its role as the reexport
center for the tin, rubber, lumber and other resources of the Malaysian
hinterland, as well as becoming a major investor in that country's
economy.
Throughout the early 1980s, Singapore headed the ASEAN drive to find
a solution to the Cambodia problem. Beginning in 1979, the ASEAN
countries sponsored an annual resolution in the UN calling for a
withdrawal of Vietnamese troops and a political settlement on Cambodia.
In 1981 Singapore hosted a successful meeting of the leaders of the
three Khmer liberation factions, which led to the formation of the
Coalition Government of Democratic Kampuchea the following year.
During the first half of the 1980s, the Singapore economy continued
to grow steadily, despite a worldwide recession. The economic growth
rate of about 10 percent in 1980 and 1981 dipped to 6.3 percent in 1982
but rebounded to 8.5 percent with only 2.7 percent inflation in 1984. In
his 1984 New Year's message to the nation, Lee Kuan Yew attributed
Singapore's high economic growth rate, low inflation, and full
employment during the period to its hardworking work force, political
stability and efficient administration, regional peace, and solidarity
in ASEAN. Singapore's successful economic strategy included phasing out
labor-intensive industries in favor of high-technology industries, which
would enhance the skills of its labor force and thereby attract more
international investment.
Although Lee Kuan Yew retained a firm grip on the reins of government
during the second decade of the country's independence,the shift in
leadership had been irrevocably set in motion. By the early 1980s, a
second generation of leaders were beginning to occupy the important
decision-making posts. The stars of the new team included Goh Chok Tong,
Tony Tan, S. Dhanabalan, and Ong Teng Cheong, who were all full
ministers in the government by 1980. In that year, the PAP won its
fourth consecutive general election, capturing all the seats. Its 75.6
percent vote margin was five points higher than that of the 1976
election. The PAP leadership was shaken out of its complacency the
following year, however, when Workers' Party candidate J. B. Jeyaretnam
won with 52 percent of the votes the by-election to fill a vacancy in
Anson District. In the general election held in December 1984,
Jeyaretnam retained his seat and was joined on the opposition benches by
Chiam See Tong, the leader of the Singapore Democratic Party, which was
founded in 1980.
In September 1984, power in the PAP Central Executive Committee was
transferred to the second-generation leaders, with only Lee Kuan Yew, as
secretary general, remaining of the original committee members. When Lee
hinted in 1985 that he was considering retirement, his most likely
successor appeared to be Goh Chok Tong, serving then as first deputy
prime minister and defence minister. Speculation also centered on the
prime minister's son, Lee Hsien Loong, who had resigned his military
career to win a seat in Parliament in the 1984 election. After two
decades of the highly successful, but tightly controlled, administration
of Lee Kuan Yew, it was difficult to say whether the future would bring
a more open and participatory government, yet one with the same knack
for success exhibited by the old guard. The answer to that question
would only come with the final passing of Lee Kuan Yew from the
political scene.
Singapore
Singapore - The Society and Its Environment
Singapore
MIRROR GLASS BANK TOWERS overshadowing Victorian-era government
buildings symbolized Singapore's transformation from a colonial port to
an independent city-state with the highest standard of living in
Southeast Asia in 1989. Singapore's status as a newly industrializing
economy (
NIE) was signaled by its landscaped complexes of
owner-occupied apartments and streets blocked by the private cars of
affluent citizens. The citizens increasingly considered themselves
Singaporeans rather than Chinese or Indians or Malays, and the
multiethnic population increasingly used English as the common speech in
schools, offices, and the armed forces. Singapore in the 1980s had
become a byword for orderliness and effective administration, a place
where stiff fines discouraged littering and citizens of all ethnic
groups were subject to common, impartial standards of merit and
achievement. Government efforts at social engineering extended beyond
slum clearance and the creation of housing estates to such matters as
men's hair length, the language families spoke at breakfast, and the
number of children born to women with university degrees.
Singapore's leaders reacted to the unanticipated 1965 separation from
Malaysia, which left a city without a hinterland, by deciding to
"go cosmopolitan." This meant seeking a place in the world
rather than in the regional economy; it also meant maintaining a certain
social and cultural distance from neighboring countries while
deliberately fostering a new and distinctively Singaporean culture and
social identity. By late 1989 Singapore was cosmopolitan, prosperous,
modernized, and orderly. Its population was educated in English, worked
for multinational corporations, and consumed a worldwide popular culture
of film, music, and leisure activities. English was, however, a second
language for most, and many distinctively Chinese, Indian, and Malay
customs, practices, and attitudes continued. In contrast to many
countries of the region, Singapore's avowed social values were secular,
democratic (within certain limits), and nondiscriminatory.
The content of the distinctive "Singaporean identity" and
the proper balance between cosmopolitan and traditional values were
issues that both preoccupied the leadership and would continue to shape
the society in the 1990s. There was much public discussion of social
identity, ethnicity, and the proper relation of Singaporeans to
worldwide popular culture. Such discussion, often initiated by political
leaders, tended to dichotomize habits and behavior into mutually
exclusive "Asian" or "Western" categories. The
initial premise was that Singapore should be a modernized but not a
Westernized society, and that it would be a mistake for Singaporeans to
become so thoroughly Westernized and cosmopolitan as to lose touch with
their Asian roots and values. Such concepts as tradition and modernity,
local and cosmopolitan, had a distinctively Singaporean meaning as was
indicated by the widespread use of such terms as "Asian
traditions" and "cultural ballast." The meaning of these
concepts, however, remained to be defined more precisely by the
discussions and day-to-day decisions of Singapore's citizens.
Singapore
Singapore - Geography
Singapore
Singapore is located at the tip of the Malay Peninsula at the
narrowest point of the Strait of Malacca, which is the shortest sea
route between India and China. Its major natural resources are its
location and its deep-water harbor. Singapore Island, though small, has
a varied topography. The center of the island contains a number of
rounded granitic hills that include the highest point, the 165- meter
Bukit Timah Peak. The western and southwestern regions are composed of a
series of northwest to southeast tending ridges, which are low but quite
steep. To the east is a large region of generally flat alluvial soils
where streams have cut steep-sided valleys and gullies. The island is
drained by a large number of short streams, some of which flow into the
sea through mangrove swamps, lagoons, or broad estuaries.
The island originally was covered with tropical rain forest and
fringed with mangrove swamps. Since the founding of the city in 1819,
the natural landscape has been altered by human hands, a process that
was accelerated in the 1970s and 1980s. By 1988, Singapore's land area
was 49 percent built up, and forest covered only 2.5 percent. Three
water reservoirs and their reserve catchment area, which preserves a
fragment of the original tropical forest, occupy the center of the
island. Extensive land reclamation between 1965 and 1987 increased the
size of Singapore Island from 586 square kilometers to 636 square
kilometers; further reclamation was planned for the 1990s. Hills have
been leveled, swamps drained and filled, and many of the fifty-odd small
islets and reefs have been enlarged or joined to form new larger islands
suitable for industrial uses. In 1989 three of Singapore's five oil
refineries were on offshore islands, and other small islands were used
for military gunnery or as bombing ranges. Some of the larger streams
were dammed at their mouths to form fresh-water reservoirs, and the
major stream courses through built-up areas were lined with concrete to
promote rapid drainage. Throughout the 1970s and 1980s, the municipal
authorities made great efforts to establish parks and gardens as land
became available and to plant tens of thousands of ornamental trees and
shrubs, thus completing the transformation of the natural landscape.
Singapore is two degrees north of the equator and has a tropical
climate, with high temperatures moderated by the influence of the sea.
Average daily temperature and humidity are high, with a mean maximum of
31�C and a relative humidity of 70 to 80 percent in the afternoon. Rain
falls throughout the year, but is heaviest during the early northeast
monsoon from November through January. The driest month is July in the
middle of the southeast monsoon. The intermonsoon months of April-May
and October are marked by thunderstorms and violent line squalls locally
known as Sumatras. The average annual rainfall is 237 centimeters, and
much of the rain falls in sudden showers. Singapore is free from
earthquakes and typhoons, and the greatest natural hazard is local flash
flooding, the threat of which has increased as buildings and paved roads
have replaced natural vegetation.
In spite of the high rainfall, Singapore's small size and dense
population make it necessary to import water from Malaysia. The water,
from reservoirs in upland Johor, comes through an aqueduct under the
causeway linking Singapore with the Malaysian city of Johor Baharu.
Singapore also supplies treated water to Johor Baharu, which in 1987
took about 14 percent of the 1 million cubic meters treated by Singapore
each day. Singapore has responded to this dependence on a foreign
country for water by expanding its reservoir capacity and constantly
urging household and industrial users to conserve water.
Singapore's rapid economic growth in the 1970s and 1980s was
accompanied both by increased air and water pollution and by
increasingly effective government efforts to limit environmental damage.
The government established an Anti-Pollution Unit under the Prime
Minister's Office in 1970, set up the Ministry of the Environment in
1972, and merged the Anti-Pollution Unit with that ministry in 1983 to
ensure unified direction of environmental protection. The new unit,
subsequently renamed the Pollution Control Department, had
responsibility for air and water pollution, hazardous materials, and
toxic wastes. Singapore first moved to limit air pollution,
closely monitoring oil refineries and petrochemical complexes and
limiting the sulfur content of fuel oil for power plants, factories, and
diesel motor vehicles. Because motor vehicles were the main source of
air pollution, the government required emissions controls on engines and
reduced (but not eliminated) the lead content of gasoline. The
government also acted, partly for environmental reasons, to restrict
private ownership of automobiles through very high (175 percent) import
duties, high annual registration fees, and high charges for the entry of
private automobiles to the central business district.
Between 1977 and 1987, the Ministry of the Environment carried out a
major program to clean up rivers and streams by extending the sewer
system, controlling discharges from small industries and workshops, and
moving pig and duck farms to resettlement areas with facilities to
handle animal wastes. The success of the program was demonstrated by the
return of fish and aquatic life to the lower Singapore and Kallang
rivers. Singapore, the world's third largest oil refiner, also acted to
prevent the pollution of coastal waters by oil spills or discharges from
the many large oil tankers that traversed the Strait of Malacca. The
Port of Singapore Authority maintained oil skimmers and other equipment
to clean up oil spills, and a comprehensive plan assigned both the oil
companies and Singapore's armed forces responsibilities for dealing with
major oil spills.
Singapore's environmental management program was intended primarily
to ensure public health and to eliminate immediate hazards to citizens
from toxins. Protection of the environment for its own sake was a low
priority, and the government did not respond to local conservation
societies' calls to preserve tropical forests or mangrove swamps. The
pollution control laws gave the authorities wide discretion in dealing
with offenders, and throughout the 1970s and 1980s penalties usually
were light. Enforcement of the laws often reflected an appreciation of
the economic benefits of polluting industries and provided time for
industrial polluters to find ways to limit or eliminate their
discharges.
Singapore
Singapore - Population
Singapore
Population, Vital Statistics, and Migration
Singapore had a population of 2,674,362 in July 1989 and the low
birth and death rates common to developed economies with high per capita
incomes. In 1987 the crude birth rate (births in proportion to the total
population) was 17 per 1,000 and the death rate was 5 per 1,000 for an
annual increase of 12 per 1,000. The infant mortality rate of 9.1 per
1,000 in 1986 was quite low by international standards and contributed
to a 1987 life expectancy at birth of 71.4 years for males and 76.3
years for females. As in most developed countries, the major causes of
death were heart disease, cancer, and strokes. As of 1986, 74 percent of
married women of childbearing age practiced contraception, and the total
fertility rate (a measure of the number of children born to a woman over
her entire reproductive career) was 1.6, which was below the replacement
level but comparable to that of many countries in Western Europe.
Since the city's founding in 1819, the size and composition of
Singapore's population has been determined by the interaction of
migration and natural increase. Throughout the nineteenth century,
migration was the primary factor in population growth. Natural increase
became more important after the 1920s, and by the 1980s immigration and
emigration were of minor significance. In the nineteenth and early
twentieth centuries, Singapore's population was composed largely of
immigrant adult males, and grew primarily through immigration. By the
1920s, the proportion of women, the percentage of the population that
was Singapore-born, and consequently the relative contribution of
natural increase to the population, all were increasing. By the 1947
census, 56 percent of the population had been born in Singapore, and
there were 1,217 males for every 1,000 females. The 1980 census showed
that 78 percent of the population had been born in Singapore and that
the sex ratio had reached 1,042 males for every 1,000 females.
Migration to Singapore dwindled during the Great Depression of the
1930s, ceased during the war years of 1941 to 1945, and resumed on a
minor scale in the decade between 1945 and 1955. Most nineteenth-century
and early twentieth-century immigrants came from China, India, or
Sumatra and the Malay Peninsula. Between 1945 and 1965 immigrants came
primarily from peninsular Malaya, which shared British colonial status
with Singapore and so permitted the free movement of people between
Singapore and the rural areas and small cities of the peninsula. After
independence in 1965, Singapore's government imposed strict controls on
immigration, granting temporary residence permits only to those whose
labor or skills were considered essential to the economy. Most such
workers were expected to return to their homelands when their contracts
expired or economic downturns made their labor redundant. Illegal
immigrants and Singaporeans who employed them were subject to fines or
imprisonment. The immigrants of the 1980s fell into two distinct
categories. The first category, unskilled labor for factories and
service positions, was composed largely of young unmarried people from
Malaysia, Thailand, the Philippines, Sri Lanka, and India. Regulations
prohibited their marrying without prior official permission and required
women to be tested for pregnancy every six months--measures intended to
make it difficult for them to attain Singaporean residence or
citizenship by becoming the spouse or parent of a citizen. The second
category comprised skilled workers, professionals, and managers, often
working for multinational corporations. They came from Japan, Western
Europe, North America, and Australia. Predominately middle-aged and
often accompanied by their families, they were immigrants only in the
strict sense of the government's population registration and had no
intention of settling permanently in Singapore.
The 1980 census reported that 9 percent of Singapore's population
were not citizens. The aliens were divided into permanent residents (3.6
percent of the population) and nonresidents (5.5 percent). The
acquisition of Singapore citizenship was a complex and often protracted
process that began with application to the Immigration Department for
permanent resident status. After residing in Singapore for two to ten
years, depending on skills and professional qualifications, those with
permanent resident status could apply to the Registry of Citizens for
citizenship. In 1987 citizenship was granted to 4,607 applicants and
denied to 1,603 applicants. The 1980 census showed that 85.5 percent of
citizens had been born in Singapore, 7.8 percent in China (including
Hong Kong, Macao, and Taiwan), 4.7 percent in Malaysia, and 1 percent in
the Indian subcontinent (including Pakistan, Bangladesh, and Sri Lanka).
Singapore's government, keenly aware of the country's small size and the
need to survive by selling the skills of its citizens in a competitive
international marketplace, was determined not to permit the citystate to
be overwhelmed by large numbers of unskilled rural migrants. In 1989
Singapore mounted a campaign to attract skilled professionals from Hong
Kong, offering a Chinese cultural environment with much lower living
costs than Hong Kong's. At the same time, however, that the government
was attempting to attract skilled professionals, Singaporeans themselves
were emigrating. From July 1987 to June 1988, records show that 2,700
Singaporeans emigrated to Australia, 1,000 to Canada, 400 to the United
States, and 97 to New Zealand. A large number of the emigrants were
university-educated professionals, precisely the category that Singapore
wished to keep and attract. In 1989 a special government committee was
reported to be devising policies to discourage emigration by
professionals and managers.
Population Control Policies
Since the mid-1960s, Singapore's government has attempted to control
the country's rate of population growth with a mixture of publicity,
exhortation, and material incentives and disincentives. Falling death
rates, continued high birth rates, and immigration from peninsular
Malaya during the decade from 1947 to 1957 produced an annual growth
rate of 4.4 percent, of which 3.4 percent represented natural increase
and 1.0 percent immigration. The crude birth rate peaked in 1957 at 42.7
per thousand. Beginning in 1949, family planning services were offered
by the private Singapore Family Planning Association, which by 1960 was
receiving some government funds and assistance. By 1965 the crude birth
rate was 29.5 per 1,000 and the annual rate of natural increase had been
reduced to 2.5 percent. Singapore's government saw rapid population
growth as a threat to living standards and political stability, as large
numbers of children and young people threatened to overwhelm the
schools, the medical services, and the ability of the economy to
generate employment for them all. In the atmosphere of crisis after the
1965 separation from Malaysia, the government in 1966 established the
Family Planning and Population Board, which was responsible for
providing clinical services and public education on family planning.
Birth rates fell from 1957 to 1970, but then began to rise as women
of the postwar baby boom reached child-bearing years. The government
responded with policies intended to further reduce the birth rate.
Abortion and voluntary sterilization were legalized in 1970. Between
1969 and 1972, a set of policies known as "population
disincentives" were instituted to raise the costs of bearing third,
fourth, and subsequent children. Civil servants received no paid
maternity leave for third and subsequent children; maternity hospitals
charged progressively higher fees for each additional birth; and income
tax deductions for all but the first two children were eliminated. Large
families received no extra consideration in public housing assignments,
and top priority in the competition for enrollment in the most desirable
primary schools was given to only children and to children whose parents
had been sterilized before the age of forty. Voluntary sterilization was
rewarded by seven days of paid sick leave and by priority in the
allocation of such public goods as housing and education. The policies
were accompanied by publicity campaigns urging parents to "Stop at
Two" and arguing that large families threatened parents' present
livelihood and future security. The penalties weighed more heavily on
the poor, and were justified by the authorities as a means of
encouraging the poor to concentrate their limited resources on
adequately nurturing a few children who would be equipped to rise from
poverty and become productive citizens.
Fertility declined throughout the 1970s, reaching the replacement
level of 1.006 in 1975, and thereafter declining below that level. With
fertility below the replacement level, the population would after some
fifty years begin to decline unless supplemented by immigration. In a
manner familiar to demographers, Singapore's demographic transition to
low levels of population growth accompanied increases in income,
education, women's participation in paid employment, and control of
infectious diseases. It was impossible to separate the effects of
government policies from the broader socioeconomic forces promoting
later marriage and smaller families, but it was clear that in Singapore
all the factors affecting population growth worked in the same
direction. The government's policies and publicity campaigns thus
probably hastened or reinforced fertility trends that stemmed from
changes in economic and educational structures. By the 1980s,
Singapore's vital statistics resembled those of other countries with
comparable income levels but without Singapore's publicity campaigns and
elaborate array of administrative incentives.
By the 1980s, the government had become concerned with the low rate
of population growth and with the relative failure of the most highly
educated citizens to have children. The failure of female university
graduates to marry and bear children, attributed in part to the apparent
preference of male university graduates for less highly educated wives,
was singled out by Prime Minister Lee Kuan Yew in 1983 as a serious
social problem. In 1984 the government acted to give preferential school
admission to children whose mothers were university graduates, while
offering grants of S$10,000 to less educated women who agreed to be
sterilized after the birth of their second child. The government also
established a Social Development Unit to act as matchmaker for unmarried
university graduates. The policies, especially those affecting placement
of children in the highly competitive Singapore schools, proved
controversial and generally unpopular. In 1985 they were abandoned or
modified on the grounds that they had not been effective at increasing
the fecundity of educated women.
In 1986 the government also decided to revamp its family planning
program to reflect its identification of the low birth rate as one of
the country's most serious problems. The old family planning slogan of
"Stop at Two" was replaced by "Have Three or More, if You
Can Afford It." A new package of incentives for large families
reversed the earlier incentives for small families. It included tax
rebates for third children, subsidies for daycare, priority in school
enrollment for children from large families, priority in assignment of
large families to Housing and Development Board apartments, extended
sick leave for civil servants to look after sick children and up to four
years' unpaid maternity leave for civil servants. Pregnant women were to
be offered increased counseling to discourage "abortions of
convenience" or sterilization after the birth of one or two
children. Despite these measures, the mid-1986 to mid-1987 total
fertility rate reached a historic low of 1.44 children per woman, far
short of the replacement level of 2.1. The government reacted in October
1987 by urging Singaporeans not to "passively watch ourselves going
extinct." The low birth rates reflected late marriages, and the
Social Development Unit extended its matchmaking activities to those
holding Advanced level (A-level) secondary educational qualifications as
well as to university graduates. The government announced a public
relations campaign to promote the joys of marriage and parenthood. In
March 1989, the government announced a S$20,000 tax rebate for fourth
children born after January 1, 1988. The population policies
demonstrated the government's assumption that its citizens were
responsive to monetary incentives and to administrative allocation of
the government's medical, educational, and housing services.
Population Distribution and Housing Policies
In the early 1950s, some 75 percent of the population lived in very
crowded tenements and neighborhoods; these were usually occupied by a
single ethnic group in the built-up municipality on the island's
southern shore. The remaining 25 percent lived in the northern
"rural" areas in settlements strung along the roads or in
compact villages, known by the Malay term kampong, and usually
inhabited by members of a single ethnic group. Many kampongs were
squatter settlements housing wage laborers and urban peddlers. Low-cost
public housing was a major goal of the ruling People's Action Party (
PAP). Vigorous efforts at slum clearance and resettlement of squatters
had begun with the establishment in 1960 of the Housing and Development
Board, which was granted wide powers of compulsory purchase and forced
resettlement. By 1988, Housing and Development Board apartments were
occupied by 88 percent of the population and 455,000 of these apartments
(74 percent of all built) had been sold to tenants, who could use their
pension savings from the compulsory Central Provident Fund for the
downpayment. The balance was paid over twenty years with variable rate
mortgage loans, the interest rate in 1987 being 3.4 percent. The
government envisaged a society of homeowners and throughout the 1980s
introduced various measures such as reduced downpayments and extended
loan periods to permit low-income families to purchase apartments.
The massive rehousing program had many social effects. In almost
every case, families regarded the move to a Housing and Development
Board apartment as an improvement in their standard of living. Although
high-rise apartment complexes usually are regarded as examples of
crowded, high-density housing, in Singapore the apartments were much
less crowded than the subdivided shophouses (combined business and
residence) or squatter shacks they had replaced. Between 1954 and 1970
the average number of rooms per household increased from 0.76 to 2.15,
and the average number of persons per room decreased from 4.84 to 2.52.
Movement to a public housing apartment was associated with (although not
the cause of) a family structure in which husband and wife jointly made
important decisions, as well as with a family's perception of itself as
middle class rather than working class. The government used the
resettlement program to break up the ethnically exclusive communities
and sought to ensure that the ethnic composition of every apartment
block mirrored that of the country as a whole. Malays, Indians, and
Chinese of various speech groups lived next door to each other, shared
stairwells, community centers, and swimming pools, patronized the same
shops, and waited for buses together.
Although the earliest public housing complexes built in the 1960s
were intended to shelter low-income families as quickly and cheaply as
possible, the emphasis soon shifted to creating new communities with a
range of income levels and public services. The new complexes included
schools, shops, and recreation centers, along with sites on which
residents could use their own resources to construct mosques, temples,
or churches. The revised master plan for land use called for the
creation of housing estates at the junctions of the expressways and the
mass transit railroad that were to channel urban expansion out from the
old city center. New towns of up to 200,000 inhabitants were to be
largely self-contained and thoroughly planned communities, subdivided
into neighborhoods of 4,000 to 6,000 dwelling units. In theory, the new
towns would be complete communities providing employment for most
residents and containing a mixture of income levels. In practice, they
did not provide sufficient employment, and many residents commuted to
work either in the central business district or in the heavy industrial
area of Jurong in the southwestern quadrant of the island. Public
transportation made the journey to the central business district short
enough that many residents preferred to shop and dine there rather than
at the more limited establishments in their housing estates. Thus, as in
other countries that have attempted to build new towns, Singapore's new
towns and housing estates have served largely as suburban residences and
commuter settlements, the center of life only for the very young and the
very old.
Throughout the 1980s, the government and the Housing and Development
Board made great efforts to foster a sense of community in the housing
estate complexes by sponsoring education and recreation programs at
community centers and setting up a range of residents' committees and
town councils. The apartment complexes generally were peaceful and
orderly, and the relations between residents were marked by civility and
mutual tolerance. But social surveys found that few tenants regarded
their apartment blocks as communities in any very meaningful sense.
Residents' primary social ties were with relatives, old classmates,
fellow-workers, and others of the same ethnic group, who often lived in
housing complexes some distance away. In the late 1980s, families who
had paid off their mortgages were free to sell their apartments, and a
housing market began to develop. There were also administrative
mechanisms for exchanging apartments of equivalent size and value.
Residents used sales, purchases, and apartment exchanges to move closer
to kin and friends who belonged to the same ethnic group. The result was
a tendency toward the recreation of the ethnic communities that had been
deliberately broken up in the initial resettlement.
The government criticized the tendency toward ethnic clustering as
contrary to its policy of multiracialism and in March 1989 announced
measures to halt it. Although no family would be forced to move from its
apartment, new rules prohibited the sale or exchange of apartments to
members of other ethnic groups. Although the tendency toward ethnic
resegregation apparently stemmed more from personal and pragmatic
motivations than from conscious antagonism toward other ethnic groups,
the government effort to halt it and to enforce ethnic quotas for
apartment blocks demonstrated the continued significance of ethnicity in
Singapore's society.
<"15.htm">Ethnic Categories
Updated population figures for Singapore.
Singapore
Singapore - Ethnic Categories
Singapore
Since the city's foundation in 1819, Singapore's population has been
polyglot and multiethnic. Chinese have been in the majority since 1830
but have themselves been divided into sometimes antagonistic segments
speaking mutually unintelligible Chinese languages. The colonial society
was compartmented into ethnic and linguistic groups, which were in turn
associated with distinct political and economic functions. Singapore has
never had a dominant culture to which immigrants could assimilate nor a
common language. This was the foundation upon which the efforts of the
government and ruling party to create a common Singaporean identity in
the 1970s and 1980s rested.
On July 1989 Singapore's 2,674,362 residents were divided into
2,043,213 Chinese (76.4 percent), 398,480 Malays (14.9 percent), 171,160
Indians (6.4 percent), and 61,511 others (2.3 percent). The proportions
of the ethnic components had remained substantially unchanged since the
1920s. Although the ethnic categories were meaningful in the Singaporean
context, each subsumed much more internal variation than was suggested
by the term "race." Chinese included people from mainland
China, Taiwan, and Hong Kong, as well as Chinese from all the countries
of Southeast Asia, including some who spoke Malay or English as their
first language. The Malays included not only those from peninsular
Malaya, but also immigrants or their descendants from various parts of
the Indonesian archipelago, such as Sumatra, the Riau Islands south of
Singapore, Java, and Sulawesi. Those people who in Indonesia were
members of such distinct ethnic groups as Acehnese, Minangkabau,
Buginese, Javanese, or Sundanese were in Singapore all considered
"Malays." Indians comprised people stemming from anywhere in
pre-1947 British India, the present states of India, Pakistan, and
Bangladesh, and from Sri Lanka and Burma. Singapore's Indian
"race" thus contained Tamils, Malayalis, Sikhs, Gujaratis,
Punjabis, and others from the subcontinent who shared neither physical
appearance, language, nor religion.
<"16.htm">The Chinese
<"17.htm">The Malays
<"18.htm">The Indians
More about the <"14.htm">Population
of Singapore.
Singapore
Singapore - The Chinese
Singapore
Singapore's Chinese residents were the descendants of immigrants from
coastal southeastern China, an area of much linguistic and subcultural
variation. The migrants spoke at least five mutually unintelligible
Chinese languages, each of which contained numerous regional dialects.
Singaporean usage, however, following the common Chinese tendency to
assert cultural unity, referred to mutually unintelligible speech
systems as "dialects." All the Chinese languages and dialects
shared common origins and grammatical structures and could be written
with the same Chinese idiograms, which represent meaning rather than
sound. The primary divisions in the immigrant Chinese population
therefore followed linguistic lines, dividing the populace into segments
that were called dialect communities, speech groups, or even
"tribes". In the nineteenth century, each speech group had its
own set of associations, ranging from secret societies to commercial
bodies to schools and temples. The groups communicated through leaders
conversant with other Chinese languages or through a third language such
as Malay or English.
The nomenclature for Chinese speech groups common in Singapore and
Southeast Asia is confusing, partly because each group can be referred
to by several alternate names. Most of the names refer to places in
China with characteristic regional speech or dialects, and include the
names of provinces, counties, and major cities.
The distribution of Singapore's Chinese speech groups has remained
fairly stable since 1900. The largest group were the Hokkien, who came
from the area around the trading port of Xiamen (Amoy) in southern
Fujian Province. Hokkien traders and merchants had been active in
Southeast Asia for centuries before the foundation of Singapore. In 1980
they made up 43 percent of Singapore's Chinese population. The second
largest group were the Teochiu (sometimes written Teochew), comprising
22 percent of the Chinese population. Their home area is Chaozhou, in
Chao'an County in northeastern Guangdong Province, which has as its
major port the city of Shantou (Swatow). Chaozhou is immediately south
of the Hokkien-speaking area of Fujian, and both Teochiu and Hokkien are
closely related languages of the Minnan group, mutually intelligible to
native speakers after sufficient practice. Hainanese, from the island of
Hainan south of Guangdong, made up 8 percent of the population. Hainan
was settled by people from southern Fujian who arrived by sea, and
Hainanese is a Minnan language whose native speakers can understand
Hokkien or Teochiu with relatively little difficulty after practice.
Speakers of Minnan languages thus made up 72 percent of the Chinese
population, for whom Hokkien served as a lingua franca, the language of
the marketplace.
The third most numerous group were Cantonese, from the lowlands of
central Guangdong Province around the port city of Guangzhou (Canton).
They made up 16 percent of the Chinese population. Hakka, a group
scattered through the interior hills of southern China and generally
considered migrants from northern China, were 7 percent. Other Chinese
call them "guest people", and the term Hakka (Kejia in pihyih
romanization) is Cantonese for "guest families." There also
were small numbers of people from the coastal counties of northern
Fujian, called Hokchia, Hokchiu, and Henghua, whose northern Fujian
(Minbei) languages are quite distinct from those of southern Fujian and
seldom spoken outside of Fujian. A final, residual category of Chinese
were the "Three Rivers People," who came from the provinces
north of Guangdong and Fujian. This group included people from northern
and central China, and more specifically those provinces sharing the
word "river" (jiang) in their names--Jiangxi,
Jiangsu, and Zhejiang. They would have spoken southern Mandarin dialects
or the Wu languages of Shanghai, Ningbo, and Hangzhou. In 1980 they were
1.7 percent of the Chinese population.
A significant category of Chinese, although one not listed in the
census reports, were the Baba
Chinese or Straits
Chinese. They were Chinese who after long residence
in Southeast Asia spoke Malay or English as their first language, and
whose culture contained elements from China, Southeast Asia, and
sometimes Europe as well. An indication of the size of the Baba Chinese
community was provided by the 1980 census report that 9 percent of
Chinese families spoke English at home. Stereotypically the Baba were
the offspring of Chinese migrants and local women. In the nineteenth
century, they tended to be wealthier and better educated than the mass
of immigrants and to identify more with Singapore and Southeast Asia
than with China. In spite of their language, the Baba considered
themselves Chinese, retained Chinese kinship patterns and religion, and
even when speaking Malay used a distinct Baba dialect of Malay with many
loan words from Hokkien. Never a large proportion of Singapore's Chinese
population, in the late nineteenth century they took advantage of
opportunities for education in English and promoted themselves as loyal
to Britain. In Singapore, many Baba families spoke English as a first
language and produced many of the leaders of Singapore's independent
political movements, including Lee Kuan Yew. Although the Baba, in a
sense, provided the model for the current Singaporean who is fluent in
English and considers Singapore as home, the community fragmented in the
early twentieth century as Chinese nationalism spread. After the 1920s
its members gained no advantage, economic or political, from
distinguishing themselves from the rest of the Chinese population and
tended increasingly to become Chinese again, often learning to speak
Chinese as adults. In the 1980s, Baba culture survived largely in the
form of a well-known cuisine that mixed Chinese and Malay ingredients
and in some families who continued to use English as the language of the
home.
As the majority of the population and the ethnic group that dominated
the political system and state administrative structure, Singapore's
Chinese exhibited the widest range of occupational, educational, and
class status. Those with little or no formal education occupied the
bottom rungs of the occupational hierarchy and led social lives
restricted to fellow members of the same dialect group. The level of
formal education and language of education--Chinese or English--divided
the Chinese into broad categories. Status for those working in the
internationally oriented private sector or in government service
depended on command of English and educational qualifications. In the
still substantial Chinese private sector, status and security rested on
a position in a bounded dialect community and a network of personal
relations established over a lifetime. Although the latter exclusively
Chinese category was shrinking, by the late 1980s it still contained
some quite wealthy men who helped set the international price of rubber,
controlled businesses with branches in Malaysia, Indonesia, Hong Kong,
and other countries of the region, and supported Singapore's array of
Chinese charities, hospitals, and education trusts. Singapore's Chinese
society was one with a high degree of social mobility and one in which
status increasingly was determined by educational qualifications and
command of English and Mandarin.
More about the <"15.htm">Ethnic
Categories of Singapore.
Singapore
Singapore - The Malays
Singapore
The Malay made up 15 percent of Singapore's population and were, like
the Chinese and the Indians, descendants of immigrants. They or their
ancestors came from peninsular Malaya, Sumatra, Java, and the other
islands of the Indonesian archipelago. Throughout the nineteenth and
twentieth centuries, Java was much more densely populated than
peninsular Malaya, and its people had a significantly lower standard of
living. From the mid-nineteenth century to the period just after World
War II, many Javanese migrated to Singapore, attracted both by urban
wages offering a higher living standard and by freedom from the
constraints of their native villages, where they often occupied the
lower reaches of the economic and social order. Singapore Malay
community leaders estimated that some 50 to 60 percent of the community
traced their origins to Java and an additional 15 to 20 percent to
Bawean Island, in the Java Sea north of the city of Surabaya. The 1931
census recorded the occupations of 18 percent of the Malays as fishermen
and 12 percent as farmers; the remaining 70 percent held jobs in the
urban cash economy, either in public service or as gardeners, drivers,
or small-scale artisans and retailers. The British colonialists had
considered the Malays as simple farmers and fishermen with strong
religious faith and a "racial" tendency toward loyalty and
deference; they preferentially recruited the Malays to the police, the
armed forces, and unskilled positions in the public service. In 1961
more than half of Singapore's Malays depended on employment in the
public sector. Although the colonial stereotype of the Malays as rural
people with rural attitudes persisted, Singapore's Malay residents were
for the most part no more rural than any other residents. Malay identity
was couched in religious terms, with Malay being taken almost as a
synonym for Muslim, and most Malay organizations taking a religious
form.
After independence, the government regarded the Malay preponderance
in the police and armed forces as disproportionate and a potential
threat to security and acted to make the security forces more
representative of the society as a whole, which meant in practice
replacing Malays by Chinese. The government's drive to
break up ethnic enclaves and resettle kampong dwellers in Housing and
Development Board apartment complexes had a great effect on the Malays.
Evidence of the convergence of Malay patterns of living with those of
the rest of the population was provided by population statistics, which
showed the Malay birth and death rates, originally quite high, to be
declining. In the 1940s, Malay women had married early, had many
children, and were divorced and remarried with great frequency. By the
1980s, Malays were marrying later, bearing fewer children (2.05 per
woman for mid-1986 to mid- 1987), and divorcing less frequently. By the
1980s, a large proportion of Malay women were working outside the home,
which was a major social change. Many young women in their late teens
and early to mid-twenties were employed in factories operated by
multinational corporations, which, unlike the small-scale Chinese shops
and workshops that had dominated the economy into the 1960s, paid no
attention to ethnicity in hiring. Even Malay fishing communities on the
offshore islands, which appeared to preserve the traditional way of
life, were in the 1980s losing population as young people moved to
Singapore Island, attracted by urban life and unskilled jobs that
offered higher and more reliable incomes than fishing.
Although very much a part of Singapore's modernizing society, the
Malays conspicuously occupied the bottom rungs of that society; their
position illustrated a correlation between ethnicity and class that
presented a major potential threat to social stability. With the lowest
level of educational attainment of any ethnic group, the Malays were
concentrated at the low end of the occupational hierarchy and had
average earnings that were 70 percent of those of Chinese. Malays had a
higher crime rate than other groups and in 1987 accounted for 47 percent
of the heroin addicts arrested. The 1980 census showed that 86 percent
of the Malay work force was in the clerical, service, and production
sector; 45 percent of all employed Malays worked on assembly lines,
largely in foreign-owned electronics factories. Only 8 percent of all
professional and technical workers (including schoolteachers), and 2
percent of all administrative and managerial personnel were Malays.
Malays dropped out of the competitive school system in large numbers,
and those who continued past primary school were concentrated in
vocational education programs. In 1980 they made up only 1.5 percent of
all university graduates and 2.5 percent of students enrolled in higher
education.
In sharp contrast to neighboring Malaysia with its policies of
affirmative action for the Malay majority, Singapore's government
insisted that no ethnic group would receive special treatment and that
all citizens had equal rights and equal opportunities. The potential
threat, however, posed by the overlap between Malay ethnicity and low
educational achievement and occupational status, was clear.
Demonstrating the Singaporean propensity for discussing social affairs
in terms of "race," both government spokesmen and Malay
intellectuals tended to attribute the Malays' economic position and
educational performance to something inherent in the Malay personality
or culture, or to their supposed "rural" attitudes. The ways
in which lower income and ill-educated Malays resembled or differed from
the very many lower income and ill- educated Chinese, who had very
different cultural backgrounds, were not addressed.
In 1982 the prime minister defined Malays' educational difficulties
as a national problem and so justified government action to improve
their educational performance. The colonial government had provided free
but minimal education, in the Malay language, to Malays but not to
Chinese or Indians, on the grounds that the Chinese and Indian residents
of Singapore, even those born there, were sojourners. In the colonial
period most English- language schools were run by churches or
missionaries, and many Malays avoided them for fear of Christian
proselytization. Although after independence schooling in Singapore was
not free (fees were generally low, but the government felt that people
would not value education if they did not pay something for it), Malays
continued to receive free primary education. In 1960 that benefit was
extended to secondary and higher education, although the free schooling
was offered only to those the government defined as Malay, which
excluded immigrant Indonesians whom the Malays regarded as part of their
community. Throughout the 1960s and most of the 1970s, most Malay
children continued to attend schools that taught only in Malay, or, if
they taught English at all, did so quite poorly. Opportunities for
secondary and higher education in the Malay language were very limited.
Although many Malays were employed in the public service or as drivers
or servants for foreign employers, in almost all cases the language used
at work was the grammatically and lexically simplified tongue called
Bazaar Malay.
Throughout the 1970s, relatively few Malays knew English, a language
that became progressively more necessary for high-paying professional
and technical jobs. Substantial numbers of the Chinese knew no more
English than the Malays, but they found employment in the extensive
sector of Chinese commerce and small-scale industry where hiring
demanded command of a Chinese regional language and personal
recommendation. The former Malay economic niche in the military and
police forces was eliminated in the late 1960s and 1970s, and the large
number of Malays who had been employed by the British armed forces at
British naval and other military facilities lost those secure and
well-paying positions when the British withdrew from Singapore from 1970
to 1975. Such factors as poor command of English, limited availability
of secondary and postsecondary education in Malay, and the loss of
public-sector jobs accounted for much of the low economic position of
the Malay community in 1980.
In 1981 Malay community leaders, alarmed by the results of the 1980
census that demonstrated the concentration of Malays in the lower
reaches of the occupational hierarchy, formed a foundation called
Mendaki, an acronym for Majlis Pendidikan Anak-anak Islam (Council for
the Education of Muslim Children). Mendaki (ascent in Malay), devoted
itself to providing remedial tuition classes for Malay children in
primary and secondary school, offering scholarships for living expenses
and loans for higher education, attempting to encourage parents to take
a more active role in their children's education, and holding public
ceremonies to honor Malay students who excelled in examinations or
graduated from academic secondary schools or universities. Government
support for Mendaki took the form of financing the organization through
a special voluntary checkoff on the monthly contribution of Muslim
workers to the Central Provident Fund, and through unspecified other
public donations.
Throughout the 1980s, both the number of Malay students in selective
secondary schools and institutions of higher education and the
proportion of Malays passing and scoring well on standardized
examinations slowly increased. As with the changes in birth rates, it
was difficult to separate the effects of such government-sponsored
programs as those of Mendaki from other factors, including increased
female participation in the work force, residence in apartment complexes
rather than kampong housing, exposure to television and radio, smaller
family size, and better teaching in the schools.
The use of a voluntary checkoff on the monthly Central Provident Fund
contribution as a means of raising Malay educational funds was
characteristic of Singapore in the 1980s. Malays, like other
Singaporeans, were assumed to have regular employment and salaries, and
their distinctive Malay and Muslim concerns were efficiently and
equitably addressed through a computerized government program.
More about the <"15.htm">Ethnic
Categories of Singapore.
Singapore
Singapore - The Indians
Singapore
The Indians, although a component of Singapore's society since its
founding, were in the 1980s its most immigrant-like community. In the
nineteenth and early twentieth centuries, Indian men had worked in
Singapore, sending money home to families and wives in India, whom they
would visit every few years. Indian women and complete Indian families
were rare before World War II, and the Indian sex ratio in 1931 was
5,189 men for every 1,000 women. The 1980 census showed 1,323 Indian men
for every 1,000 women; most of the surplus males were over age 60. In
the 1980s, the "Little India" off Serangoon Road contained
many dormitories where elderly single men lived, as well as some shops
and workshops whose owners, in the traditional pattern, housed and fed a
workforce of middleaged and elderly men who might or might not have
wives and children in India or Sri Lanka. Significant issues for the
Indian community included securing residence status, citizenship, or
entrance for the Indian families of men who had worked in Singapore for
decades and for the Brahman priests who were necessary for Hindu
religious life.
Almost two-thirds (64 percent) of the Indian population were Tamils
from southeastern India's Tamil Nadu state; some Tamils also came from
Jaffna in northern Sri Lanka. The great diversity of the Indian populace
was indicated by the census category "other Indians," who made
up a substantial 19 percent of the group, followed by Malayalis (8
percent); Punjabis, mostly Sikh (8 percent); and Gujaratis (1 percent).
Like the Straits Chinese, some of Singapore's Indians adopted English as
a first language, a change facilitated by the widespread use of English
in India, where it had become another Indian language. Indians were the
most religiously diverse of Singapore's ethnic categories; an estimated
50 to 60 percent were Hindu, 20 to 30 percent Muslim, l2 percent
Christian, 7 percent Sikh, and 1 percent Buddhist. Indian immigrants, like those of other nationalities, had
been primarily recruited from among poor farmers and laborers, which
meant that they included a large proportion (perhaps onethird ) of
untouchables. In Singapore untouchables were usually referred to by the
more polite Tamil term Adi-Dravidas, meaning pre-Dravidians. Although
Tamils made up nearly two-thirds of the Indian population and Tamil was
one of the country's four official languages (along with English, Malay,
and Mandarin Chinese), by 1978 more Indians claimed to understand Malay
(97 percent) than Tamil (79 percent). The 20 to 30 percent of the Indian
population who were Muslims tended to intermarry with Malays at a fairly
high rate and to be absorbed into the Malay community, continuing a
centuries-old process of assimilation of Indian males to Malay society.
The linguistic and religious diversity of the Indian population was
matched by their high degree of occupational differentiation. Indians
were represented at all levels of the occupational hierarchy in numbers
roughly proportional to their share of the total population. Within the
Indian category, occupational and education attainment was far from
equitably distributed. The untouchables for the most part did unskilled
or semiskilled labor, while the Jaffna Tamils and the Chettia caste, who
were traditionally moneylenders and merchants, were often professionals
and wealthy businessmen. After World War II, caste received no public
recognition in Singapore. Untouchables were free to enter Hindu temples,
and food was distributed at temple festivals without regard for relative
degrees of purity and pollution. Members of the Indian community were
reluctant to discuss caste in public, but it continued to play a
decisive role in marriage arrangements. The Indians were the most likely
of all ethnic groups to attempt to arrange marriages for their children,
or at least to restrict the choice of marriage partners to acceptable
caste categories. Although the relatively small size of the Indian
population and the disparate mixture of local caste groups from large
areas of southern India made it difficult for most families to insist on
strict caste endogamy (marrying only within the caste), Hindu marriages
were made within a tripartite hierarchy. The highest level was occupied
by Brahmans and Chettias, who attempted to maintain caste endogamy or at
least to marry only members of other high castes. Mid-level caste Hindus
intermarried with little difficulty, but the marriages of low-caste or
outcaste category of former hereditary washermen, barbers, and
untouchables were restricted to their own circle
More about the <"15.htm">Ethnic
Categories of Singapore.
Singapore
Singapore - Singaporean Identity
Singapore
The period after Singapore's withdrawal from Malaysia in 1965 saw
much public discussion of Singaporean identity. The discussion tended to
use terms, categories, and basic assumptions provided by the government
and ruling party. One basic assumption was that there was not, at least
in the late 1960s and 1970s, a common Singaporean identity, but that
there should be. A corollary was that Singaporean identity would not
spontaneously emerge from the country's ongoing social, political, and
cultural life. Rather, it would have to be consciously created and
"built" by policies, directives, and educational campaigns.
The content of the identity remained somewhat ill-defined, and it often
appeared easier to say what Singaporean identity was not than what it
was. The ideal seemed to combine, somewhat uneasily, a self-consciously
toughminded meritocratic individualism, in which individual Singaporeans
cultivated their talents and successfully competed in the international
economy, with an equally self-conscious identification with "Asian
roots" and "traditional values," which referred to
precolonial India, China, and the Malay world. Singaporeans were to be
modern and cosmopolitan while retaining their distinctively Asian
traditions.
Singapore's leaders explicitly rejected the ideology of the melting
pot, offering rather the vision of a confidently multiethnic society
whose component ethnic groups shared participation in such common
institutions as electoral politics, public education, military service,
public housing, and ceremonies of citizenship; at the same time they
were to retain distinct languages, religions, and customs. Singaporeans
were defined as composed of three fundamental types--Chinese, Malays,
and Indians. These ethnic categories, locally referred to as
"races," were assumed to represent self-evident,
"natural" groups that would continue to exist into the
indefinite future. Singaporean identity thus implied being a Chinese, a
Malay, or an Indian, but selfconsciously so in relation to the other two
groups. The Singaporean model of ethnicity thus required both the denial
of significant internal variation for each ethnic category and the
highlighting of contrasts between the categories.
Being Singaporean also meant being fluent in English, a language
which served both as a neutral medium for all ethnic groups and as the
medium of international business and of science and technology. The
schools, the government, and the offices of international corporations
for the most part used English as their working language. The typical
Singaporean was bilingual, speaking English as well as the language of
one of the three component ethnic groups. Hence the former
English-speaking Baba, Chinese or Indian, would seem to serve as the
model of Singaporean identity. The resulting culture would be the type
social scientists call "creolized," in which a foreign
language such as English or French is adapted to local circumstances and
the dominant culture reflects a unique blending of local and
"metropolitan" or international elements. In the 1980s, there
were signs of the emergence of such a culture in Singapore, with the
growth among youth (of all "races") of a distinctive
English-based patois called "Singlish" and the attraction of
all ethnic groups to international fashions and fads in leisure
activities.
Singapore's leaders resisted such trends toward cosmopolitan or
creole culture, however, reiterating that Singaporeans were Asians
rather than Westerners and that abandoning their own traditions and
values for the tinsel of international popular culture would result in
being neither truly Western nor properly Asian. The consequence would be
loss of identity, which in turn would lead to the dissolution of the
society. The recommended policy for the retention of Asian identity
involved an ideal division of labor by language. English was to function
as a language of utility. The Asian "mother tongues"--Mandarin
Chinese, Malay, and Tamil--would be the languages of values, providing
Singaporeans with what political leaders and local academics commonly
called "cultural ballast" or "moral compasses."
Stabilized and oriented by traditional Asian values, the Singaporean
would be able to select what was useful from the offerings of
"Western" culture and to reject that which was harmful. This
theory of culture and identity resulted in the effort to teach the
"mother tongues" in the schools and to use them as the vehicle
for moral education.
In an extension of the effort to create a suitable national identity,
in 1989 Singapore's leaders called for a "national ideology"
to prevent the harmful drift toward superficial Westernization. The
national ideology, which remained to be worked out in detail, would help
Singaporeans develop a national identity and bond them together by
finding and encouraging core values common to all the country's diverse
cultural traditions. Suggested core values included emphasizing
community over self, valuing the family, resolving issues through the
search for consensus rather than contention, and promoting racial and
religious tolerance.
Singapore
Singapore - Language
Singapore
In colonial Singapore, the nearest thing to a common language had
been Bazaar Malay, a form of Malay with simplified grammar and a very
restricted vocabulary that members of many ethnic groups used to
communicate in the marketplace. The government used English, with
translators employed when necessary, as in the courts. Among the Chinese
a simplified form of Hokkien served as the language of the marketplace.
The Chinese schools, which were founded in large numbers in the early
years of the twentieth century and associated with the rise of Chinese
nationalism, attempted to teach in Mandarin Guoyu, the use of which on
such formal occasions as weddings and Chinese national holiday
celebrations came to carry some prestige. In the terminology of
sociolinguistics, Singapore's language system was multilingual and
diglossiac, that is, characterized by two languages or dialects, high
and low, or classical and vernacular, each used in different social
contexts and carrying differential prestige. Bazaar Malay and market
Hokkien were the low languages, employed in the streets and market
places, and English and Mandarin were the high languages, used in
education, government offices, and public celebrations. In addition,
such native tongues as pure Malay, Teochiu, Tamil, or Punjabi were used
in the home and in gatherings of members of the same speech group. In a
1972 survey asking which language people understood, Hokkien came first,
at 73 percent, followed by Malay, with 57 percent. Malay was the most
important language for intergroup communication, with almost all the
Indians and 45 percent of the Chinese claiming to understand it. English
came second, understood by 47 percent of the total population. A
follow-up survey in 1978 showed that 67 percent claimed to understand
Malay and 62 percent to comprehend English. As the 1990s approached
English was replacing Malay as the common language. It was used not only
as the high language but also, in its Singlish variant, as a low
language of the streets. Bazaar Malay was declining, and Malay in its
full native complexity was increasingly used only by Malays. Even though
it was one of the four official languages and the putative "mother
tongue" of the Indian community, Tamil was used less often and
literacy in Tamil was reported to be declining.
The most ambitious aspect of Singapore's language planning and
attempted social engineering was the campaign to replace the Chinese
"dialects" with Mandarin, called the "mother
tongue." The Speak Mandarin campaign began in 1979 as a PAP project
and was subsequently institutionalized in the Mandarin Campaign
Secretariat in the Ministry of Communications and Information. The
promotion of Mandarin as a common Chinese language dates back to the
early years of the century, when it was associated with the rise of
Chinese nationalism and the foundation of Chinese schools. Learning
Mandarin would, it was argued, permit all Chinese to communicate in
their "mother tongue," be useful for doing business with
China, and, perhaps most important, promote traditional Chinese values.
All ethnic Chinese were required to study Mandarin through secondary
school and to pass examinations in it for university admission. Chinese
civil servants took a required 162-hour conversational Mandarin course,
and the Mandarin Campaign Secretariat coordinated the annual Speak
Mandarin campaigns. Mandarin classes were offered by the Singapore
Chinese Chamber of Commerce and Industry and by some native-place and
clan associations. All Chinese television broadcasting was in Mandarin,
as was most radio broadcasting. Radio programs in Chinese dialects were
limited to 9:00 P.M. to midnight on the same station that broadcast
Tamil from 5:00 A.M. to 9:00 P.M. In 1989 members of Parliament
complained that some residents were tuning in to Cantonese opera
broadcast by television stations in neighboring Malaysia. By late 1988,
some 87 percent of the Chinese population claimed to be able to speak
Mandarin. People did not agree, however, on the appropriate social
contexts for use of what was for everyone a school language. As a
result, people tended to use English or their native tongue on most
everyday occasions. During the late 1980s, the Speak Mandarin campaign
attempted to persuade people to use Mandarin when shopping and targeted
taxi drivers, bus conductors, and operators of food stalls as workers
who were to use Mandarin.
The goals of the Speak Mandarin campaign included improving
communication between Chinese speech groups, teaching people to read
Chinese, and promoting Confucianism. Some critics argued that children
were expected to learn two foreign languages in school (English and
Mandarin) and that for some students the result was fluency in neither.
The official response was that the problem would be avoided if people
would speak Mandarin at home. Some educators questioned whether a
sufficient level of Chinese literacy could be achieved with the amount
of time the schools devoted to Chinese, a point that was indirectly
supported in August 1988 when Brigadier General Lee Hsien Loong, the
minister for trade and industry and son of Prime Minister Lee Kuan Yew,
urged Chinese newspapers to use simpler language to attract younger
readers. Some academics questioned the restriction of Chinese values to
Confucianism and recalled that in the 1950s and early 1960s Chinese was
the language of radicalism and revolt rather than of loyalty and
conservatism. The necessity of learning Mandarin to conserve traditional
Chinese culture was not obvious to those Chinese who felt that Chinese
culture had been transmitted for centuries through Hokkien, Teochiu, and
Cantonese. They pointed out that the colloquial speech of modern Beijing
(upon which Mandarin is based) was as distant from the classical Chinese
of the Confucian texts as was colloquial Cantonese. Giving up the
dialects implied a major transformation of the social structure of the
Chinese community, because the associational and commercial structure of
Singapore's Chinese-oriented society rested on (and reinforced) dialect
distinctions.
Singapore
Singapore - Ethnicity and Associations
Singapore
Because Singapore was a small society open to influence from the West
through the English language and subject to the homogenizing effects of
modernization and industrialization, the persistence of ethnicity as a
fundamental element of its social structure was by no means assured. By
the late 1980s ethnic affiliations were in many ways less significant
than they had been in 1970 or 1940, and the lives of members of distinct
ethnic groups had more and more common elements. In Singapore, as
elsewhere, the forces of standardized education, impartial application
of laws and regulations, common subordination to the impersonal
discipline of the factory and the office, common pursuit of leisure
activities, and exposure to international mass media resulted in many
shared attitudes among ethnic groups. Studies of factory workers in
Malaysia and Singapore, for example, found no marked differences in the
attitudes and performance of Chinese and Malays. Psychological profiles
of a cohort of poorly educated young Chinese who had held a succession
of unskilled jobs before induction into the armed forces resembled those
of equally poorly educated and unskilled Malays. Foreign popular culture
seemed equally tempting or equally threatening to young Singaporeans of
all ethnic groups. Ethnic boundaries persisted, especially where they
corresponded with religious distinctions, and were evident in the
continuing low rate of ethnic intermarriage. In daily life, however, the
significance of ethnic affiliation had apparently diminished from the
levels of previous generations.
Government policies were a major factor in the continuation of
ethnicity as an organizing principle of Singapore's society. On the one
hand, the government and the ruling party acted to break up ethnic
enclaves, to provide public services to members of all ethnic groups,
and to reshape society with the network of People's Association
Community Centers, Residents' Committees, and Members of Parliament
Constituent Advisory Groups. On the other hand, the government's
ideology defined Singaporeans as members of component ethnic groups, and
its various ministries listed everyone's "race" on their
identity card and all official records, and remained very concerned with
such matters as the ethnic mix in apartment complexes. Official
statistics usually included breakdowns by "race," indicating
an assumption that such categorization was significant. National
holidays featured displays of the distinctive traditional cultures of
the major ethnic groups, represented by costumes, songs, and dances.
Pupils in secondary schools took required courses in the ethics and
religion of their designated traditional culture--Confucian ethics for
the Chinese, Islamic studies for the Malays, Hindu or Sikh studies for
the Indians, and Buddhism or Bible study as options open to all.
Although state policies reinforced ethnic boundaries and the habit of
ethnic categorization, they had little effect on the content of the
ethnic categories. Ethnic identity was acted out on a daily basis
through an extensive network of ethnically exclusive associations. Many
Malay and Indian associations took a religious form, such as mosque and
endowment management committees, sharia
(Muslim law) courts, Hindu temple committees and the
high-level Hindu Advisory Board, which represented Hindus to the
government. An example of the reinforcement of ethnic identity was
provided by the groups of Indian employees in one government department
who distinguished themselves from their Malay and Chinese coworkers by
jointly sponsoring festivals at a major Hindu temple. All ethnic groups
had their own education and charitable associations as well as
higher-order federations of such associations whose officers were the
recognized community leaders. Singapore law required all associations of
ten or more persons to be registered with the government, which
supervised and could dissolve them. Trade unions, financial, education,
and religious bodies were supervised by the appropriate government
departments, and the catch-all Registry of Societies listed all
associations that did not come under the authority of a specialized
department. In 1987 3,750 associations were under the Registry of
Societies.
The most elaborate set of ethnic associations was found among the
Chinese, who in 1976 supported over 1,000 clan, locality, occupational,
religious, and recreational associations. The membership of each
association usually was restricted to those speaking the same dialect or
tracing ancestry to the same small region of China. The lowest level
associations were clan or district associations, which were in turn
grouped into federations based on progressively larger administrative or
linguistic regions of China. The Singapore Chinese Chamber of Commerce
and Industry, founded in 1906, was the overarching association that
represented the entire Chinese community. A federation, its constituent
units were not individuals or individual businesses but associations.
Its basic structure consisted of representatives of seven regional
associations (Fujian, Teochiu, Cantonese, Hakka 1, Hakka 2, Hainan, and
"Three Rivers") and ninety-three trade associations, each one
usually restricted to speakers of one dialect.
The functions and activities of the associations were multiple,
reflecting the concerns of members and leaders. Common activities
included mutual aid; insurance benefits; foundation and maintenance of
schools, hospitals, or cemeteries; contributions to the same sorts of
public projects in the ancestral districts of China; settling disputes
between members; acting as spokesman for the community to the
government; and promoting good fellowship and continuing identification
with the clan or region. Associations were run by committees and met at
least once a year for a formal banquet. Association leaders were
prosperous businessmen who had played a major part in fundraising and
the management of activities. Success in business gave them both the
free time to devote to association activities and the funds to
contribute to the association and its charities. The associations
conferred prestige and public recognition on those who took the burdens
of office and community service, but the community so served was
restricted to those from the same region and speaking the same dialect.
The leadership of the lowest level associations was usually provided by
those of moderate means, while the more wealthy belonged to several or
many associations and worked for the higher level, more inclusive
associations, which conferred more public recognition and prestige. The
mechanisms of leadership and prestige and the channeling of much charity
and assistance (schools, scholarship fund, hospitals, recommendations
for employment or loans from Chinese banks, death benefits) through the
associations thus reinforced ethnic and subethnic identification for
both poor and rich.
In a pattern common to Chinese urban society in China and in
Southeast Asia, groups defined by common place of origin or dialect also
tended to specialize in certain trades or monopolies. Exactly which
regional group dominated which trade varied from place to place and
represented historical accidents and contingencies, but the principle of
a regional group also acting as an occupational group was common. As
late as the 1980s, the Singapore Hokkien were dominant in banking,
insurance, shipping, hardware, real estate, and other lucrative fields.
Within the Hokkien community, smaller subgroups controlled particular
trades. For example, 96 percent of the merchants dealing in China tea in
the 1980s traced their ancestry to Anxi County in southern Fujian.
Teochiu dominated the fresh produce trade and the jewelry and antiques
business; Cantonese predominated in furniture making, watch and clock
repair, and operating drug stores and restaurants; and the Hakka were
pawnbrokers, tailors, and dealers in Chinese herbs and medicines. The
Henghua people from northern Fujian, a small component of the Chinese
population, controlled the very important bicycle, motorcycle, and taxi
businesses. Over the years the speech groups competed for the control of
trades, and the pattern of dialect- specific occupations was a dynamic
one, with, for example, strong competition for shares of the textile
trade. In the 1980s, four textile trade associations represented
Teochiu, Hokkien, Hakka, and Cantonese traders. The competition between
speech groups reinforced both their internal solidarity and the social
boundaries between them. Regional associations were, to a certain
extent, also trade associations. For the large proportion of the Chinese
population employed in regional commerce, service trades, or small-scale
manufacturing, there remained a close relation between ethnicity and
occupation, each aspect reinforcing the other.
For the proprietors and employees of many small and medium Chinese
businesses, continued identification with dialect and subethnic
communities provided many benefits and indeed was a precondition for
engaging in many lines of trade. Although the dialect communities were
not primarily occupational groups, the social solidarities created
within the communities were economically useful. Much of the business
activity in the extensive Chinese "traditional" sector of the
economy depended on credit, personal relations, and the reputation of
individuals for trustworthiness. In the final analysis, individuals met
their obligations because failure to do so would result in immediate
loss of reputation and creditworthiness with their fellows in restricted
subethnic communities.
For many members of the Chinese community, economic self- interest
reinforced the identification with an ethnic or subethnic community and
the continued use of a regional dialect. Such individuals tended to be
both more intensely and self-consciously "Chinese" and
"Teochiu" or "Anxi Hokkien" than their fellows, who
might well be their own brothers, sons, or daughters, who worked for the
government or large multinational corporations. For the latter, formal
educational certification, command of English, and perhaps skill at golf
rather than Chinese finger games and etiquette were associated with
economic success.
Singapore
Singapore - Social Stratification and Mobility
Singapore
During the 1970s and 1980s, economic development and industrial
growth reduced poverty and income inequity and accelerated upward social
mobility. Those with educational qualifications, command of English, and
high-level technical or professional skills profited the most from the
process.
In the late 1980s, the major indices of social stratification were
education level, citizenship status, sector of the economy where
employed, and number of employed persons in the household. Residents
were sharply differentiated by the amount of education they had
completed. In 1980 about 44 percent of the population aged 25 and above
had no educational qualifications, 38 percent had completed primary
school, 15 percent secondary school, and only 3.4 percent higher
education. Those people born after 1970 were on average much better
educated than previous generations, but throughout the 1990s the work
force will contain many individuals with limited education. Wages
correlated fairly closely with educational attainment, although
education in English brought higher salaries than Chinese education.
Many benefits, such as access to a Housing and Development Board
apartment, were available only to Singapore citizens, and only citizens
and permanent residents were enrolled in the Central Provident Fund. In
1985, a recession year when many foreign factory workers lost their jobs
and residence permits, citizens made up 91 percent of the work force.
Noncitizens were concentrated in the lower and in the highest wage
levels, either as factory or service workers on short- term work
permits, or as well-paid expatriate managers and professionals. Wages
were relatively higher in government service and government-owned
corporations and in the capital intensive and largely foreign-owned
export-oriented manufacturing sector. They were lower in the service,
retail, and less highly capitalized light industrial, craft, and
commercial sector, which was dominated by small Chinese firms. Wages for
unskilled and semiskilled factory work and for unskilled service jobs
were relatively low. Those who held such jobs, often young women in
their teens and early twenties, were not entirely self- supporting but
parts of households in which several members worked at low-paying jobs.
Families of the poorly educated and unskilled improved their standard of
living between 1970 and 1990 in part because full employment made it
possible to pool the wages of several family members.
Economic growth and the associated increase in the demand for labor
from 1960 to 1989 raised living standards and sharply reduced the
incidence of poverty. A survey of living costs and household incomes in
1953-54 found 19 percent of all households to be in absolute poverty,
meaning that their members did not have enough to eat. Application of
the same standard in 1982-83 found 0.3 percent of households in absolute
poverty. A measure of moderate poverty, defined as adequate nutrition
and shelter but little discretionary income and no savings, was devised
by the Amalgamated Union of Public Employees in 1973. By that measure,
31 percent of households in 1972-73 were in moderate poverty, 15 percent
in 1977-78, and 7 percent in 1982-83. Compared with other countries in
the region, household incomes in Singapore were equitably distributed,
with most households falling in the middle or lower middle ranges of the
distribution.
The lowest income levels were those of single-person households,
representing the elderly, the disabled, and those without kin in
Singapore. Apart from the childless elderly and the disabled, those in
moderate poverty in the 1980s were overwhelmingly working poor, holding
unskilled jobs with no prospects for advancement. Such households
typically had only one wage-earner with either primary education or no
education and lived in rented housing and often a one-room or two-room
Housing and Development Board apartment. Households with two or more
members working, even at relatively low-paying jobs, were able to
contemplate purchasing a Housing and Development Board apartment, save
money for emergencies, and devote more resources to the education of
children.
Much of the alleviation of poverty and decrease in income inequality
that took place in the 1970s and 1980s resulted from the increased
participation of women in the work force. In 1985, 46 percent of all
women above the age of fifteen held paid employment; 68 percent of
single women and 33 percent of married women worked outside the home.
This trend was associated with women marrying later and having fewer
children. One reason that more households attained an adequate standard
of living in the 1980s was that there were more wives and unmarried
daughters at work and fewer young children to be supported and looked
after.
Surveys in the 1980s showed that most Singaporeans described
themselves as middle class, justifying that status by their ownership of
a Housing and Development Board apartment and the substantial and secure
savings guaranteed by their Central Provident Fund Account. Families in
the middle-income ranges usually occupied two- or three-bedroom
apartments that they were buying from the Housing and Development Board,
participated in one or more formal associations, took an active part in
planning and supervising their children's education, stocked their
apartments with a range of consumer appliances, and had money to spend
on hobbies, sports, or vacations. Automobile ownership was not common,
and most middle-income Singaporeans used public transportation. Their
mode of life rested on occupational skills and educational
qualifications, secure employment in large, bureaucratic government or
private organizations, or ownership of their own small business.
The upper levels of the society were occupied by a tripartite elite
of high-level civil servants, local managers and professionals employed
by foreign-owned multinational corporations, and wealthy Chinese
businessmen who served as leaders in the associational world of the
Chinese-speaking communities. The first two categories were marked by
fluency in English, university-level education, often in Britain or the
United States, and a cosmopolitan outlook reinforced by foreign
residence and travel. Many of the Chinese businessmen were entrepreneurs
who operated in an exclusively Chinese setting and often had minimal
educational qualifications. Their sons, however, often were graduates of
the best secondary schools and of local or foreign universities and
worked either as English-speaking representatives of their fathers'
businesses, as civil servants, or as professionals. Few of the elite had
inherited their status, and all were aware that they could not directly
pass it along to their children. Having themselves been upwardly mobile
in a society more open to individual effort than most in the region,
they valued that society's stress on competition, individual mobility,
and success through hard work. In the domestic sphere, they expressed
those values by devoting much effort to the education of their children.
Increased family incomes made possible by full employment and by such
government programs as the construction and sale of apartments and the
enrollment of nearly everyone in the Central Provident Fund are to be
distinguished from upward mobility, in which individuals moved into more
highly skilled and highly paid jobs and hence into higher social
classes. The expansion of industry, banking, and of the ranks of civil
servants created many high and mid-level positions that Singaporeans
could aspire to and compete for. Residents from every ethnic community
regarded social mobility as a common and accepted goal. Education was
regarded as the best channel for upward mobility, and most families
tried to encourage their children to do well in school and to acquire
educational qualifications and certification. This put severe pressure
on the school system and the children in it, although, as elsewhere,
middle- and upper-income families had an advantage in maneuvering their
offspring through the education system.
Individuals approached jobs with a keen appreciation for their
potential for further mobility. Most large organizations, whether
government or private, provided some training. Some foreign-owned
enterprises, such as those in the oil industry, employed large numbers
of skilled workers and ran extensive in-house training programs. The
electronics assembly factories, in contrast, offered no prospects for
advancement to their large numbers of unskilled or semiskilled assembly
line workers. Small scale enterprises, which in the late 1980s often
recruited along ethnic and subethnic lines, were associated with long
working hours and low wages, but sometimes offered the workers
opportunity to learn a skill, such as automotive repair. Workers in such
establishments commonly advanced by quitting and opening their own small
firms, often after years of saving.
In a system that reflected both the great differences in educational
attainment in the work force and the great significance attached to
educational qualifications, most large organizations, public and
private, made a sharp distinction between mental and manual labor, and
movement from the lower to the higher was very difficult and rare. Lower
level white-collar workers and skilled blue-collar workers often took
advantage of opportunities to upgrade their occupational skills, either
through training offered by the organization or through night school and
short-term courses offered by educational or other government bodies.
Unskilled workers in industry and service trades and employees in small
Chinese firms saw few prospects for advancement and considered
self-employment as their only hope for upward mobility. Vending food and
consumer goods on the streets or operating a cooked-food stall,
traditional entry points for entrepreneurs, had been practically
eliminated by government action to tidy up the environment and to limit
the numbers of mobile hawkers who obstructed traffic. Many Singapore
economists felt that the successful modernization of the economy and the
increases both in government regulation and in rents for shops and small
premises had made it more difficult for the ambitious poor to get a
start. By the late 1980s, Singapore's academics and political leaders
were discussing the perceived shortage of entrepreneurs and suggesting
solutions to the problem, although most discussion focused on industrial
innovation and growth rather than the commercial fields in which most
Singapore entrepreneurs had succeeded.
Singapore
Singapore - Family, Marriage, and Divorce
Singapore
Almost all Singaporeans lived in small nuclear families. Although
both Chinese and Indian traditions favored large extended families, such
families were always rare in immigrant Singapore where neither the
occupational structure, based on wage labor, or the housing pattern,
characterized by small, rented quarters, favored such family forms. In
the 1980s, families were important in that most individuals as a matter
of course lived with their parents until marriage and after marriage
maintained a high level of interaction with parents, brothers, and
sisters. Probably the most common leisure activity in Singapore was the
Sunday visit to the grandparents for a meal and relaxed conversation
with brothers, sisters, in-laws, uncles and aunts, cousins, and other
assorted kin. Although the age of marriage increased in the 1970s and
1980s, reaching a mean 28.5 years for grooms and 25.8 years for brides
in 1987, Singapore remained a society in which it was assumed that
everyone would marry, and marriage was a normal aspect of fully adult
status.
Both ethnicity and class affected the form and functioning of
families. Chinese and Indian families rested on cultural assumptions of
the permanence of marriage and of the household as an ongoing, corporate
group whose members, bound by duty, obligation, and subordination,
pooled and shared income. The continued efforts of Indian parents to
arrange the marriages or at least to influence the marital choices of
their offspring and the Tamil obligation to provide daughters with large
dowries reflected such cultural definitions of family and household. In
a similar manner, some Chinese combined the household with the family
enterprise, practicing a traditional entrepreneurial strategy that
included mobilizing the savings of all household members and allocating
them in accord with a long-term plan for family success. Such a strategy
might take the form of a thriving business with branches in the major
cities of Malaysia and Indonesia, or of sons and daughters employed in
the Singapore civil service, a large foreign bank, or a university in
Australia.
Malay families, on the other hand, gave priority to the individual
and to individual interests. They viewed relations between siblings as
tenuous and saw the household as a possibly short-lived coalition of
autonomous individuals linked by sentiments of mutual concern and
affection. Malays had traditionally had much higher rates of divorce and
adoption than other ethnic groups, and the distinction continued in the
1980s although the divorce rate was lower than in the l940s or through
the l960s. More significantly, for the Malays divorce was regarded as a
realistic and normal, although unfortunate, possibility in all
marriages. Because Malays did not define the household as a continuing
body, they did not make long-range strategic plans to maximize family
income and success. In Malay families, husbands, wives, and children
with jobs held separate purses and sometimes separate savings accounts.
It was thus difficult for Malays to establish family businesses as the
Chinese and the Indians did.
Class affected families in a manner generally similar to many other
industrialized societies. In all ethnic groups, lower-class or
working-class people tended to be dependent on kin outside the immediate
household for a wide range of services, and to operate wide networks of
mutual assistance and gift exchange. Throughout the 1980s, kin provided
the bulk of child care for married women working in factories. Such
relatives were paid for their services, but less than a stranger would
have been paid. The possibility of such support often determined whether
a woman took a job outside the home, and thus demonstrated the relation
between large numbers of kin and material comfort and security.
Substantial sums of money were passed back and forth on such occasions
as the birthdays of aged parents, the birth of children, or the move
into a new apartment. Family members were a major source of information
on and referrals to jobs for many unskilled or semiskilled workers.
Relations with the extended circle of relatives were not always
harmonious or happy, but they were important and necessary to the
welfare and comfort of most working-class families.
Middle- and upper-class households were less dependent on kin
networks for support. They maintained close ties with parents and
siblings, but did not need to rely on them. Indeed their relations with
their extended kin often were more amiable than those of the lower-class
households, where mutual need often was accompanied by disputes over
allocation of such resources as grandparents' childcare services, or of
the costs of supporting elderly parents and other dependent kin. Middle-
and upper-class households spent more leisure time with people who were
not their relatives and gained much of their social support from
networks based on common schooling, occupation, and associational
memberships. In such families, the bond between husband and wife was
close as they shared more interests and activities than most
workingclass couples and made more decisions jointly.
Marriages across ethnic lines occurred, but not often. Between 1954
and 1984, intermarriage rates remained at 6 percent of all marriages. None of the traditional cultures
encouraged marriage outside the group. The Hindu traditions of caste
endogamy and the Malay insistence on conversion to Islam as a condition
of marriage were major barriers to intermarriage. Shared religion
encouraged intermarriage, with marriages between Malays and Indian
Muslims the most common form of ethnic intermarriage. Interethnic
marriages included a disproportionate number of divorced or widowed
individuals.
Divorce rates in Singapore were low. Interethnic marriages were
somewhat more likely to end in divorce than were marriages within an
ethnic group. During the 1980s the divorce rate for Malays fell, while
it rose for the other ethnic groups. In 1987 there were 23,404 marriages
in Singapore, and 2,708 divorces, or 115 divorces for every 1,000
marriages. The figures included 4,465 marriages under the Muslim Law
Act, which regulated the marriage, divorce, and inheritance of Muslims,
and 796 divorces under the same act, for a Muslim divorce rate of 178
divorces for every 1,000 marriages. Marriages under the Women's Charter
(which regulated the marriage and divorce of non-Muslims) totaled
18,939, and divorces under that law were 1,912, for a non-Muslim divorce
rate of 100 per 1,000 marriages. The differential rates of divorce for
ethnic groups may have suggested greater differences than were in fact
the case. Situations that for Malay families resulted in prompt, legal
divorce were sometimes tolerated or handled informally by Chinese or
Indian families for whom the social stigma of divorce was greater and
the barriers to legal separation higher. For all ethnic groups, the most
common source of marital breakdown was the inability or unwillingness of
the husband to contribute to maintaining the household. This sometimes
led to desertion, which was the most common ground for divorce.
Singapore
Singapore - Religion
Singapore
Temples and Festivals
Singapore's immigrants commonly made their religious congregations a
form of social organization. From the foundation of the city, colonial
authorities had avoided interfering with the religious affairs of the
ethnic communities, fostering an atmosphere of religious tolerance. It
was characteristic of colonial Singapore that South Bridge Street, a
major thoroughfare in the old Chinatown, should also be the site of the
Sri Mariamman Temple, a south Indian Hindu temple, and of the Jamae or
Masjid Chulia Mosque, which served Chulia Muslims from India's
Coromandel Coast. The major religions were Chinese popular religion,
commonly although inaccurately referred to as Daoism or Buddhism;
Hinduism; Islam; Buddhism; and Christianity. Other religions included
smaller communities of Sikhs and of Jains from India; Parsis, Indians of
Iranian descent who followed the ancient Iranian Zoroastrian religion;
and Jews, originally from the Middle East, who supported two synagogues.
The Chinese practiced Chinese popular religion, a distinctive and
complex syncretic religion that incorporates some elements from
canonical Buddhism and Daoism but focuses on the worship of gods,
ghosts, and ancestors. It emphasizes ritual and practice over doctrine
and belief, has no commonly recognized name, and is so closely entwined
with Chinese culture and social organization that it cannot proselytize.
In Singapore its public manifestations included large temples housing
images of deities believed to respond to human appeals for guidance or
relief from affliction and use of the common Chinese cycle of
calendrical festivals. These occasions included the lunar New Year (in
January or February), a festival of renewal and family solidarity; Qing
Ming (Ching Ming in Wade-Giles romanization), celebrated by the solar
calender on April 5th (105 days after the winter solstice), to remember
the ancestors and worship at their graves; the fifteenth of the fifth
lunar month (April or May), in Singapore known as Vesak Day and
celebrated as marking the birth of the Buddha; the festival of the
hungry ghosts in the seventh lunar month, a major Hokkien holiday,
marked by domestic feasting and elaborate public rituals to feed and
placate the potentially dangerous souls of those with no descendants to
worship them; and the mid-autumn festival on the fifteenth of the eighth
lunar month, an occasion for exchanging gifts of sweet round mooncakes
and admiring the full moon. All Chinese temples held one or more annual
festivals, marked by street processions, performances of Chinese
traditional operas, and domestic banquets to which those who supported
the temple, either because of residential propinquity, subethnic
affiliation with a particular temple and its deity, or personal devotion
to the god, invited their friends and business associates. To prevent
the disruption of traffic and preserve public order, the government
limited the length and route of street processions and prohibited the
use of the long strings of firecrackers that had previously been a
component of all Chinese religious display. Some festivals or customs
that had little religious significance or were not practiced by the
southeastern Chinese migrants were promoted by the government's
Singapore Tourist Promotion Board for their spectacular and innocuous
content. These included the summer dragon boat races, originally held
only in China's Chang Jiang (Yangtze) River Valley, and the lantern
festival in which paper lanterns in the shape of animals or other
objects are carried through the streets by children or, if especially
impressive, displayed in parks and temples. In China the lantern
festival is celebrated in the first lunar month at the end of the New
Year season, but in Singapore it is combined with the mid-autumn
festival.
Canonical Buddhism was represented in Singapore as Sinhalese
Theravada Buddhism. This form of Buddhism prevails in Sri Lanka and
mainland Southeast Asia and differs from the Mahayana Buddhism of China,
Korea, and Japan in both doctrine and organization. Theravada Buddhism
was brought by Sinhalese migrants from Ceylon (contemporary Sri Lanka),
who also influenced the architectural style of Thai and Vietnamese
Theravada temples. These latter were staffed by Thai or Vietnamese
monks, some of whom were originally members of the overseas Chinese
communities of those countries and served a predominantly Chinese laity,
using Hokkien, Teochiu, Cantonese, or English. Singapore was also home
to a number of Chinese sects and syncretic cults that called themselves
Buddhist but taught their own particular doctrines and lacked properly
ordained Buddhist monks.
Hindus have been part of Singapore's population since its foundation
in 1819, and some of the old Hindu temples, such as the Sri Mariamman
Temple, were declared national historical sites in the 1980s and so
preserved from demolition. Singapore's Hindus adapted their religion to
their minority status in two primary ways--compartmentalization and
ritual reinterpretation. Compartmentalization referred to the Hindus
tendency to distinguish between the home, in which they maintained a
nearly completely orthodox Hindu pattern of diet and ritual observance,
and the secular outer world of work, school, and public life, where they
did not apply categories of purity and pollution. Singapore lacked the
tightly organized caste groups of communities found in India but
replaced them in large-scale temple festivals with groups representing
those of the same occupation or place of employment. The major Hindu
holidays were the Hindu New Year, in April or May; Thaipusam, a festival
during which penitents fulfilled vows to the deity Lord Subramanya by
participating in a procession while carrying kavadi, heavy
decorated frameworks holding offerings of milk, fruit, and flowers; and
Deepavali, the Festival of Lights. Deepavali, a celebration of the
victory of light over darkness and hence of good over evil, was a
national holiday.
Seven of the ten national holidays were religious festivals; two of
them were Chinese, two Muslim, two Christian, and one Hindu. The
festivals were the Chinese New Year; Vesak Day; Hari Raya Haji, the
Muslim pilgrimage festival; Hari Raya Pusa, which marked the end of the
fasting month of Ramadan and was a time of renewal; Christmas; Good
Friday; and Deepavali. Citizens were encouraged to learn about the
festivals of other religious and ethnic groups and to invite members of
other groups to their own celebrations and feasts. Public ceremonies
such as National Day or the commissioning of military officers were
marked by joint religious services conducted by the Inter-Religious
Organization, an ecumenical body founded in 1949 to promote
understanding and goodwill among the followers of different religions.
Religion and Ethnicity
In the 1980s, members of all ethnic groups lived and worked together,
dressed similarly, and shared equal access to all public institutions
and services. Religion, therefore, provided one of the major markers of
ethnic boundaries. Malays, for instance, would not eat at Chinese
restaurants or food stalls for fear of contamination by pork, and a
Chinese, in this case, could not invite a Malay colleague to a festive
banquet. Funerals of a traditional and ethnically distinctive style were
usually held even by families that were not otherwise very religiously
observant. The Community Associations and the Singapore Tourist
Promotion Board encouraged the public celebration of such ethnically
distinctive and appropriately colorful and noncontroversial festivals as
the Chinese lantern festival and the dragon boat races.
The marriages, divorces, and inheritances of members of religious
communities and the management of properties and endowments dedicated to
religious purposes were of concern to the government, which interacted
with some religious bodies through advisory boards dating back to the
colonial period. The Hindu Advisory Board, established in 1917, advised
the government on Hindu religion and customs and on any matters
concerning the general welfare of the Hindu community. It assisted the
Hindu Endowments Board, which administered the four major Hindu temples
and their property, in organizing the annual festivals at the temples.
The Sikh Advisory Board acted in the same way for the Sikhs.
The Singapore Muslim Religious Council (Majlis Ugama Islam Singapura)
played a very important role in the organization of Islamic affairs and
therefore of the Malay community. Authorized by the 1966 Administration
of Muslim Law Act, the council, composed of members nominated by Muslim
societies but appointed by the president of Singapore, was formally a
statutory board that advised the president on all matters relating to
the Muslim religion. It acted to centralize and standardize the practice
of Islam. The council administered all Muslim trusts (wafs);
organized a computerized and centralized collection of tithes and
obligatory gifts (zakat harta and zakat fitrah); and
managed all aspects of the pilgrimage to Mecca, including registering
pilgrims, obtaining Saudi Arabian visas, and making airline
reservations. The council also helped the government reorganize the
mosque system after redevelopment. Before the massive redevelopment and
rehousing of the 1970s and 1980s, Singapore's Muslims were served by
about ninety mosques, many of which had been built and were funded and
managed by local, sometimes ethnically based, communities. Redevelopment
destroyed both the mosques and the communities that had supported them,
scattering the people through new housing estates. The council, in
consultation with the government, decided not to rebuild the small
mosques but to replace them with large central mosques. Construction
funds came from a formally voluntary contribution collected along with
the Central Provident Fund deduction paid by all employed Muslims. The
new central mosques could accommodate 1,000 to 2,000 persons and
provided such services as kindergartens, religious classes, family
counseling, leadership and community development classes, tuition and
remedial instruction for school children, and Arabic language
instruction.
The government had regulated Muslim marriages and divorces since
1880, and the 1957 Muslim Ordinance authorized the establishment of the
centralized Sharia Court, with jurisdiction over divorce and inheritance
cases. The court, under the Ministry of Community Development, replaced
a set of government-licensed but otherwise unsupervised kathi
(Islamic judges) who had previously decided questions of divorce and
inheritance, following either the traditions of particular ethnic groups
or their own interpretations of Muslim law. The court attempted to
consistently enforce sharia law, standard Islamic law as set out in the
Quran and the decisions of early Muslim rulers and jurists, and to
reduce the high rate of divorce among Malays. In 1989 the Singapore
Muslim Religious Council took direct control of the subjects taught in
Islamic schools and of the Friday sermons given at all mosques.
Religious Change
Modernization and improved education levels brought changes in
religious practice. The inflexible work schedules of industrialism,
which tended to restrict communal ritual to evenings and Sundays, and
the lack of opportunity or inclination to devote years to mastering
ceremonial and esoteric knowledge, both contributed to a general
tendency toward ritual simplification and abbreviation. At the same
time, prosperous citizens contributed large sums to building funds, and
in the 1980s a wave of rebuilding and refurbishing renewed the city's
mosques, churches, Chinese temples, Buddhist monasteries, and Hindu
temples. Ethnic affiliation was demonstrated by public participation in
such annual rituals as processions, which did not require elaborate
training or study.
Immigrants tended to drop or modify religious and ritual practices
characteristic of and peculiar to the villages they had come from. Hindu
temples founded in the nineteenth century to serve migrants of specific
castes and to house deities worshipped only in small regions of
southeastern India became the temples patronized by all Hindu residents
of nearby apartment complexes. They offered a generic South Indian
Hinduism focused on major deities and festivals. Many Chinese became
more self-consciously Buddhist or joined syncretic cults that promoted
ethics and were far removed from the exorcism and sacrificial rituals of
the villages of Fujian and Guangdong. The movement away from village
practices was most clearly seen and most articulated among the Malays,
where Islamic reformers acted to replace the customary practices (adat)
of the various Malay-speaking societies of Java, Sumatra, and Malaya
with the precepts of classical Islamic law--sharia.
In 1988 the Ministry of Community Development reported the religious
distribution to be 28.3 percent Buddhist, 18.7 percent Christian, 17.6
percent no religion, 16 percent Islam, 13.4 percent Daoist, 9 percent
Hindu, and 1.1 percent other religions (Sikhs, Parsis, Jews). The
Christian proportion of the population nearly doubled between 1980 and
1988, growing from 10 percent to nearly 19 percent. The growth of
Christianity and of those professing no religion was greatest in the
Chinese community, with most of the Christian converts being young,
well-educated people in secure white-collar and professional jobs. Most
converts joined evangelical and charismatic Protestant churches
worshiping in English. About one-third of the members of Parliament were
Christians, as were many cabinet ministers and members of the ruling
party, which was dominated by well-educated, Englishspeaking Chinese.
The association of Christianity with elite social and political status
may have helped attract some converts.
By the late 1980s, some Buddhist organizations were winning converts
by following the Protestant churches in offering services, hymnbooks,
and counseling in English and Mandarin. A Buddhist Society at the
National University of Singapore offered lectures and social activities
similar to those of the popular Christian Fellowship. Some Chinese
secondary students chose Buddhism as their compulsory religious studies
subject, regarding Confucianism as too distant and abstract and Bible
study as too Western and too difficult. They then were likely to join
Buddhist organizations, which offered congenial groups, use of English,
and a link with Asian cultural traditions. In the late 1980s, other
Chinese whitecollar and skilled workers were joining the Japan-based
Soka Gakkai (Value Creation Society, an organization based on Nichiren
Buddhism), which provided a simple, direct style of worship featuring
chanting of a few texts and formulas and a wide range of social
activities. The more successful religious groups, Christian and
Buddhist, offered directly accessible religious practice with no
elaborate ritual or difficult doctrine and a supportive social group.
In the 1980s, the government regarded religion in general as a
positive social force that could serve as a bulwark against the
perceived threat of Westernization and the associated trends of
excessive individualism and lack of discipline. It made religious
education a compulsory subject in all secondary schools in the 1980s.
The government, although secular, was concerned, however, with the
social consequences of religiously motivated social action and therefore
monitored and sometimes prohibited the activities of religious groups.
The authorities feared that religion could sometimes lead to social and
implicitly political action or to contention between ethnic groups.
Islamic fundamentalism, for example, was a very sensitive topic that was
seldom publicly discussed. Throughout the 1980s, the authorities were
reported to have made unpublicized arrests and expulsions of Islamic
activists. The government restricted the activities of some Christian
groups, such as the Jehovah's Witnesses who opposed military service,
and in 1987 the government detained a group of Roman Catholic social
activists, accusing them of using church organizations as cover for a
Marxist plot. The charismatic and fundamentalist Protestant groups,
though generally apolitical and focused on individuals, aroused official
anxiety through their drive for more converts. Authorities feared that
Christian proselytization directed at the Malays would generate
resentment, tensions, and possible communal conflict. As early as 1974
the government had "advised" the Bible Society of Singapore to
stop publishing materials in Malay. In late 1988 and early 1989, a
series of leaders, including Prime Minister Lee Kuan Yew, condemned
"insensitive evangelization" as a serious threat to racial
harmony. Official restatements of the virtue of and necessity for
religious tolerance were mixed with threats of detention without trial
for religious extremists.
Singapore
Singapore - Health and Welfare
Singapore
Medical Services and Public Health
As indicated by their long life expectancy and low death rates,
Singaporeans generally enjoyed good health. Standards of nutrition and
environmental sanitation were high. The Ministry of the Environment's
Vector Control and Research Department was responsible for controlling
mosquitoes, flies, rats, and other disease-bearing animals; the Food
Control Department and the Hawkers Department inspected food producers
and outlets for cleanliness and sanitation. The Ministry of the
Environment's Public Affairs Department conducted educational campaigns
on such topics as environmental sanitation, control of mosquito-breeding
sites, proper disposal of refuse, and food handling. Educational efforts
were backed up by sanctions, which included fines of up to S$500 for
spitting or failing to flush public toilets.
The population was served by nine government hospitals with 7,717
beds and by twelve private hospitals with 2,076 beds. In 1987 the
Ministry of Health certified 2,941 physicians, 9,129 nurses, 653
dentists, and 487 pharmacists. Five of the nine government hospitals
were general hospitals, providing a complete range of medical services
and twenty-four hour emergency rooms, and the other four each had a
specialty: obstetrics and gynecology, dermatology and venereology,
psychiatry, or infectious diseases. In 1987 the Ministry of Health's
Community Health Service operated twenty-four clinics in major housing
complexes, offering primary medical treatment for injuries and common
diseases. The Maternal and Child Health Service provided preventive
health care for mothers and preschool children at twenty-three clinics,
while school children were served by the School Health Service.
Government hospitals and clinics charged fees for their services,
although the fees were generally low and the medical services were
heavily subsidized. The fees were intended to discourage frivolous use
of the medical system and to demonstrate that residents were responsible
for their own health costs, as Singapore was not a welfare state. After
1984 Singaporeans could pay for their medical expenses through the
Medisave Scheme, under which 6 percent of the monthly income of every
contributor to the Central Provident Fund could be set aside for the
medical expenses of the contributor and the contributor's spouse,
parents, grandparents, and children in all government or private
hospitals.
Mortality and Morbidity
The major causes of death in 1986 were heart disease, accounting for
24 percent of all deaths; cancer, 23 percent; cerebrovascular disease,
(stroke) 11 percent; and pneumonia, 8 percent. In 1988 two minor
outbreaks of dengue fever took place but were halted through prompt
control of arthropod-borne microorganisms, and a minor cholera epidemic
broke out among the inmates of a mental institution. In 1982 the World
Health Organization (WHO) declared Singapore malaria-free, and 161 of
the 165 cases of malaria reported in 1987 were determined to be
imported. In 1987 the most serious epidemic disease was hepatitis; 752
cases of acute viral hepatitis and 11 deaths were reported.
Noise-induced deafness and industrial-related skin disease the major
occupational diseases; there was also some concern over exposure of
workers to toxic and carcinogenic substances and to asbestos. The health
authorities paid special attention to patients with kidney failure, a
condition that which killed some 200 people a year. The number of deaths
reflected inadequate dialysis facilities and a shortage of organ donors.
The 1987 Human Organ Transplant Law gave doctors the right to remove the
kidneys of those killed in accidents unless the victim had objected in
writing or was a Muslim.
AIDS Policy
At the end of 1988, the Ministry of Health reported thirty-four cases
of acquired immune deficiency syndrome (AIDS) among Singaporeans; four
of these cases resulted in death. The first two cases were identified in
1985. Thereafter the incidence increased; five new cases were reported
in December 1988 alone. In 1987 the Ministry of Health established an
AIDS Task Force to inform health professionals of research on and
treatment programs for the disease. A National Advisory Committee, also
formed in 1987, with representatives from the Ministry of Health, other
ministries, the public media, hotels, and travel agencies concentrated
on educating the public about the disease. The Ministry of Health worked
with WHO, adapting its information and strategies to local
circumstances. All blood donors were routinely screened for AIDS, and
blood screening could be done at designated government clinics. In 1989
the Ministry of Health was sponsoring education programs on AIDS and
offering confidential counseling to people worried that they might be
infected. The ministry was trying to reach members of high-risk groups,
but many of them refused counseling from fear of being identified and
stigmatized.
Singapore
Singapore - Education
Singapore
The School System
The government frequently referred to Singapore's population as its
only natural resource and described education in the vocabulary of
resource development. The goal of the education system was to develop
the talents of every individual so that each could contribute to the
economy and to the ongoing struggle to make Singapore productive and
competitive in the international marketplace. The result was an
education system that stressed the assessment, tracking, and sorting of
students into appropriate programs. Educators forthrightly described
some students and some categories of students as better
"material" and of more value to the country than others. In
the 1960s and 1970s the education system, burdened with large numbers of
children resulting from the high birth rates of the previous decades and
reflecting the customary practices of the British colonial period,
produced a small number of highly trained university graduates and a
much larger number of young people who had been selected out of the
education systems following secondary schooling by the rigorous
application of standards. The latter entered the work force with no
particular skills. Major reforms in 1979 produced an elaborate tracking
system, intended to reduce the dropout rate and to see that those with
low academic performance left school with some marketable skills. During
the 1980s, more resources were put into vocational education and efforts
were made to match the "products" of the school system with
the manpower needs of industry and commerce. The combination of a school
system emphasizing testing and tracking with the popular perception of
education as the key to social mobility and to the source of the
certifications needed for desirable jobs led to high levels of
competition, parental pressure for achievement, and public attention and
concern.
In 1987 some 4 percent of the gross domestic product ( GDP) was
devoted to education. The government's goal for the 1990s was to
increase spending to 6 percent of GDP, which would match the levels of
Japan and the United States. Education was not compulsory, but
attendance was nearly universal. Primary education was free, and Malays
received free education through university. Students' families had to
purchase textbooks and school uniforms, but special funds were available
to ensure that no student dropped out because of financial need.
Secondary schools charged nominal fees of S$9.50 per month. Tuition at
the National University of Singapore for the 1989-90 academic year
ranged from S$2,600 per year for students in the undergraduate arts and
social sciences, business administration, and law courses to S$7,200 per
year for the medical course. The university-level tuitions were intended
to induce prosperous families to bear a share of the cost of training
that would lead to a well-paying job, but a system of loans, needbased
awards (bursaries), and scholarships for superior academic performance
meant that no able students were denied higher education because of
inability to pay.
The schools operated a modified British-style system in which the
main qualifications were the Cambridge University-administered General
Common Entrance (GCE) Ordinary level (O level) and Advanced level (A
level) examinations. Singapore secondary students took the same
examinations as their counterparts in Britain or in British system
schools throughout the world. All instruction was in English, with
supplementary teaching of the students' appropriate "mother
tongue"--Malay, Tamil, or Mandarin. The basic structure was a
six-year primary school, a four-year secondary school, and a twoyear
junior college for those preparing to enter higher education. As part of
the effort to reduce the dropout rate, some students progressed through
the system more slowly than others, spending more time in primary and
secondary school but achieving similar standards. The goal was that
every student achieve some success and leave school with some
certification. Both primary and secondary schools operated on double
sessions. Plans for the 1990s called for converting secondary schools to
single-session, all-day schools, a measure that would require
construction of fifty new schools.
As of June 1987, there were 229 government and government-aided
primary schools enrolling 266,501 students. Government-aided schools
originally were private schools that, in return for government
subsidies, taught the standard curriculum and employed teachers assigned
by the Ministry of Education. There were 157 secondary schools and
junior colleges, enrolling 201,125 students, and 18 vocational training
schools, enrolling 27,000 students. The 15 junior colleges operating by
late 1989 enrolled the "most promising" 25 percent of their
age cohort and were equipped with computers, laboratories and
well-stocked libraries. Some represented the elite private schools of
the colonial period, with their ancient names, traditions, and networks
of active alumni, and others were founded only in the 1980s, often in
the centers of the housing estates. In 1989 the government was
discussing the possibility of permitting some of the junior colleges to
revert to private status, in the interest of encouraging educational
excellence and diversity.
Singapore had six institutions of higher education: National
University of Singapore (the result of the 1980 merger of Singapore
University and Nanyang University); Nanyang Technological Institute;
Singapore Polytechnic Institute; Ngee Ann Polytechnic; the Institute of
Education; and the College of Physical Education. In 1987 these six
institutions enrolled 44,746 students, 62 percent male and 38 percent
female. Enrollment in universities and colleges increased from 15,000 in
1972 to nearly 45,000 in 1987, tripling in fifteen years. The largest
and most prestigious institution was the National University of
Singapore, enrolling 13,238 undergraduates in 1987. Only half of those
who applied to the National University were admitted, a degree of
selectivity that in 1986 brought parliamentary complaints that the
admission rate was inconsistent with the government's objective of
developing every citizen to the fullest potential.
The Ministry of Education tried to coordinate enrollments in
universities and polytechnic institutes and specific degree and diploma
courses with estimates of national manpower requirements. At the
university level, the majority of the students were enrolled in
engineering, science, and vocationally oriented courses. The Ministry of
Education and the government clearly preferred an education system that
turned out people with vocational qualifications to one producing large
numbers of general liberal arts graduates. The ministry attempted to
persuade students and their parents that enrollment in the three
polytechnic institutes, which offered diplomas rather than the more
prestigious degrees (a common distinction in the British system of
higher education), was not necessarily a second choice. In promoting
this choice, the ministry pointed to the good salaries and excellent
career prospects of polytechnic graduates who were employed by large
multinational corporations. Similar arguments were used to persuade
those who left secondary school with respectable O level level scores to
enroll in short courses at vocational and technical training institutes
and to qualify for such positions as electronics technicians or word
processors that were beyond the capabilities of those who had been
directed into vocational schools after the primary grades. Almost all of
the graduates of the demanding four-year Honors Degree Liberal Arts and
Social Science program at the National University of Singapore were
recruited into the upper levels of the civil service. Many graduates of
the ordinary three-year arts, social science, and science programs were
steered into teaching in secondary schools.
Education and Singaporean Identity
More clearly than any other social institution, the school system
expressed the distinctive vision of Singapore's leadership, with its
stress on merit, competition, technology, and international standards,
and its rejection of special privileges for any group. Singaporeans of
all ethnic groups and classes came together in the schools, and the
education system affected almost every family in significant and
profound ways. Most of the domestic political issues of the country,
such as the relations between ethnic groups, the competition for elite
status, the plans for the future security of the nation and its people,
and the distribution of scarce resources were reflected in the schools
and in education policy. Many of the settled education policies of the
1980s, such as the use of English as the medium of instruction, the
conversion of formerly Malay or Chinese or Anglican missionary schools
to standard government schools, or the attempted combination of open
access with strict examinations, were the result of long-standing
political disputes and controversy. In the determination of families and
parents that their children should succeed in school, and in the
universally acknowledged ranking of primary and secondary schools and
the struggle to enroll children in those schools that achieved the best
examination results, families expressed their distinctive values and
goals. The struggle for achievement in the schools, which often included
tutoring by parents or enrollment of young children in special private
supplementary schools to prepare for crucial examinations, also
demonstrated the system of social stratification and the struggle for
mobility that characterized the modern society. It was in the schools,
more than in any other institution, that the abstract values of
multiracialism and of Singaporean identity were given concrete form.
Singapore
Singapore - The Economy
Singapore
A FORMER COLONIAL TRADING PORT serving the regional economies of
maritime Southeast Asia, Singapore in the 1990s aspired to be a
"global city" serving world markets and major multinational
corporations. A quarter century after independence in 1965, the
city-state had become a manufacturing center with one of the highest
incomes in the region and a persistent labor shortage. As one of Asia's
four "little dragons" or newly industrializing economies,
Singapore along with the Republic of Korea (South Korea), Taiwan, and
Hong Kong was characterized by an export-oriented economy, relatively
equitable income distribution, trade surpluses with the United States
and other developed countries, and a common heritage of Chinese
civilization and Confucian values. The small island had no resources
other than its strategic location and the skills of its nearly 2.7
million people. In 1988 it claimed a set of economic superlatives,
including the world's busiest port, the world's highest rate of annual
economic growth (11 percent), and the world's highest savings rate (42
percent of income).
Singapore lived by international trade, as it had since its founding
in 1819, and operated as a free port with free markets. Its small
population and dependence on international markets meant that regional
and world markets were larger than domestic markets, which presented
both business managers and government policymakers with distinctive
economic challenges and opportunities. In 1988 the value of Singapore's
international trade was more than three times its gross domestic product
( GDP). The country's year-to-year economic performance fluctuated
unpredictably with the cycles of world markets, which were beyond the
control or even the influence of Singapore's leaders. In periods of
growing international trade, such as the 1970s, Singapore could reap
great gains, but even relatively minor downturns in world trade could
produce deep recession in the Singapore economy, as happened in 1985-86.
The country's dependence on and vulnerability to international markets
shaped the economic strategies of Singapore's leaders.
The economy in the 1980s rested on five major sectors: the regional
entrep�t trade; export-oriented manufacturing; petroleum refining and
shipping; production of goods and services for the domestic economy; and
the provision of specialized services for the international market, such
as banking and finance, telecommunications, and tourism. The spectacular
growth of manufacturing in the 1970s and 1980s had a major impact on the
economy and the society, but tended to obscure what carried over from
the economic structure of the past. Singapore's economy always depended
on international trade and on the sale of services. An entrep�t was
essentially a provider of services such as wholesaling, warehousing,
sorting and processing, credit, currency exchange, risk management, ship
repair and provisioning, business information, and the adjudication of
commercial disputes. In this perspective, which focused on exchange and
processing, the 1980s assembly of electronic components and manufacture
of precision optical instruments were evolutionary steps from the
nineteenthcentury sorting and grading of pepper and rubber. Both
processes used the skills of Singaporeans to add value to commodities
that were produced elsewhere and destined for consumption outside the
city-state.
The dependence on external markets and suppliers pushed Singapore
toward economic openness, free trade, and free markets. In the 1980s,
Singapore was a free port with only a few revenue tariffs and a small
set of protective tariffs scheduled for abolition in the 1990s. It had
no foreign exchange controls or domestic price controls. There were no
controls on private enterprise or investment, nor any limitations on
profit remittance or repatriation of capital. Foreign corporations were
welcome, foreign investment was solicited, and fully 70 percent of the
investment in manufacturing was foreign. The government provided foreign
and domestic enterprises with a high-quality infrastructure, efficient
and graft-free administration, and a sympathetic concern for the
problems of businesses.
The vulnerability inherent in heavy dependence on outside markets
impelled Singapore's leaders to buffer their country's response to
perturbations in world markets and to take advantage of their country's
ability to respond to changing economic conditions. Unable to control so
much that affected their nation's prosperity, they concentrated on those
domestic institutions that could be controlled. The consequence was an
economy characterized by a seemingly paradoxical adherence to free trade
and free markets in combination with a dominant government role in
macroeconomic management and government control of major factors of
production such as land, labor, and capital. The extraordinarily high
domestic savings rate provided reserves to weather such economic storms
as trade recessions and generated a pool of domestically controlled
capital that could be invested to serve the long-term interests of
Singapore rather than of foreign corporations. The high savings rate,
however, was the result of carefully formulated government programs,
which included a compulsory contribution of up to 25 percent of all
salaries to a government-controlled pension fund. The government held
about 75 percent of the country's land, was the largest single employer,
controlled the level of wages, and housed about 88 percent of the
population in largely self-owned apartments. It also operated a set of
wholly-owned government enterprises and held stock in additional
domestic and foreign firms. Government leaders, deeply aware of
Singapore's need to sell its services in a competitive international
market, continually stressed the necessity for the citizens to master
high levels of skills and to subordinate their personal wishes to the
good of the community. The combination of devotion to free-market
principles and the need for internal control and discipline in order to
adapt to the demands of markets reminded observers of many family firms,
and residents of the country commonly referred to it as Singapore Inc.
<"28.htm">PATTERNS OF DEVELOPMENT
<"29.htm">ECONOMIC ROLES OF THE GOVERNMENT
<"34.htm">LABOR
<"35.htm">INDUSTRY
<"36.htm">TRADE
<"37.htm">Tourism
<"38.htm">FINANCE
<"39.htm">AGRICULTURE
Singapore
Singapore - Economy - PATTERNS OF DEVELOPMENT
Singapore
Modern Singapore, founded as a trading post of the British East India
Company in 1819, achieved its initial economic success as an entrep�t
because of the island's location, harbor, and free port status. Although
Singapore at first served only as a center for trade and transshipment,
by the early twentieth century, primary goods, mainly rubber and tin
from the neighboring Malay Peninsula, were being imported for
processing. Singapore also became a regional center for the distribution
of European manufactured goods. After World War I, when the British
established a naval base on the island, Singapore became a key element
of the British Commonwealth of Nations military defense east of India,
thus adding the naval support industry to the island's economy.
In the period immediately after World War II, Singapore faced
enormous problems, including labor and social unrest, a decaying,
war-ravaged infrastructure, inadequate housing and community facilities,
a slow economic growth rate, low wages, and high unemployment made worse
by a rapidly expanding population. As late as 1959, the unemployment
rate was estimated at 13.5 percent. The struggle for survival in the
postwar period deeply affected the economic decision making of
Singapore's first generation leaders.
Mounting political pressure for independence from Britain culminated
in 1963 in the merger of Malaya, Singapore, and the British northern
Borneo territories of Sabah and Sarawak into the new nation of Malaysia.
A combination of political and ethnic differences between Singapore and
the national government, however, led in 1965 to Singapore's separation
from Malaysia and establishment as an independent nation. The economic
prospects of the new city-state at first appeared bleak. Upon separation
from Malaysia, Singapore lost its economic hinterland and jeopardized
its hopes for an enlarged domestic market to absorb the goods produced
by a small but growing manufacturing sector. Moreover, Indonesia's
policy of Confrontation (Konfrontasi) with Malaysia between 1963 and
1966 had substantially reduced Singapore's entrep�t trade.
Britain's announcement in 1968 of its intention to withdraw military
forces from Singapore by the early 1970s marked the beginning of a
greatly expanded, more intrusive role for the government in the economy.
From then on, the government no longer confined itself to such
traditional economic pursuits as improving the infrastructure, but
instead began to engage in activities that were or could have been the
domain of private enterprise. Britain's departure meant the loss,
directly or indirectly, of 38,000 jobs (20 percent of the work force) at
a time of already rising unemployment and rapid population growth; a
consequent reduction in the GDP; and an increase in Singapore's own
budgetary defense allocation to compensate for the British withdrawal.
Even so, the S$1,616 per capita income of Singapore in 1965 already was
quite high by developing country standards, an indication that
subsequent high growth rates were not merely a result of beginning at a
low base.
The period from 1965 to 1973 witnessed unprecedented economic growth
for the island nation, during which the average annual growth of real
GDP was 12.7 percent. Major credit for this development must be given to
the effective implementation of soundly conceived government policies,
which from the outset took full account of Singapore's strengths and
weaknesses. Furthermore, the time was right for structural change in the
economy. Enough capital had been accumulated to permit the domestic
production of goods that were more capital intensive. The government's
economic response to separation from Malaysia and the withdrawal of
British military forces included efforts to increase industrial growth
and solve the domestic problems of unemployment, population growth, and
housing. Growth was achieved because workers were added to the payroll
and provided with better machinery with which to work. Even more
remarkable, this growth was accomplished with an outstanding record of
price stability. Inflation was kept low by the government's conservative
fiscal policies, which included the maintenance of strict control over
the money supply.
Industrialization promised the most economic progress. The strategic
question was whether to rely principally on domestic entrepreneurs or to
make a conscious effort to attract foreign direct investment. The
decision to encourage the latter resulted both in a large share of
Singaporean manufacturing being foreignowned and a high degree of
export-led growth. Singapore's reliance on multinational corporations of
the world to provide the necessary investment meant less dependence on
the Southeast Asian region generally and neighboring countries
particularly.
The 1973 oil shock with the collapse of prices and the worldwide
recession it triggered brought the end of the super growth period. Even
so, Singapore's growth rate averaged 8.7 percent from 1973 to 1979,
which was high compared with other countries during that same period.
Manufacturing continued to grow as did transportation and
communications. Although the second worldwide oil crisis, beginning in
1979, set off the longest and deepest recession in the industrialized
countries since the Great Depression of the 1930s, Singapore was
seemingly untouched. If anything, its economy grew in 1980-81 while the
world economy was contracting. The real average GDP growth rate between
1979 and 1981 was 8.5 percent. Financial and business services joined
manufacturing as the major economic engines. During this period,
Singapore's function as a petroleum-servicing entrep�t made it more
like an oil producer than an oil consumer.
For the first two decades of its independence, Singapore enjoyed
continuous high economic growth, largely outperforming the world
economy. Its GDP growth rate never fell below 5 percent and rose as high
as 15 percent. At the same time, Singapore managed to maintain an
inflation rate below world averages.
Given Singapore's dependence on the world economy, however, the
consequences of declining foreign demand were inevitable. The 1985
recession was the worst in the nation's history. Singapore staggered
under a year of negative growth (-1.5 percent), then recovered slightly
in 1986 (+1.9 percent). The causes lay both outside and within the
country. Externally, worldwide slumps in petroleum-related and
marine-related sectors were reflected in reduced demand for Singapore's
goods and services and raised the specter of worldwide overcapacity in
shipbuilding and shiprepairing . Furthermore, the slowdown in demand for
semiconductors and electronics in the United States sharply reduced
demand for Singaporean components and parts.
Internally, the construction boom--which had produced a glut of
hotels, shopping centers, and apartments--began to be reversed. Domestic
demand also weakened as a result of a rise in domestic savings, which
was not matched by a rise in productive domestic investment. The
situation was complicated by a loss of international competitiveness and
a profit squeeze attributed to labor costs rising faster than
productivity.
The government responded promptly and firmly by lowering employer
contributions to the Central Provident Fund, freezing overall wage
levels for 1986 and 1987, reducing corporate income taxes from 40 to 30
percent, reducing personal income taxes in line with corporate taxes,
and introducing an across-the-board investment allowance of 30 percent
to encourage greater investment in equipment and machinery. These
measures were highly successful; costs dropped 30 percent and
productivity climbed. By 1988 Singapore's economy had rebounded.
Singapore
Singapore - ECONOMIC ROLES OF THE GOVERNMENT
Singapore
Budgeting and Planning
Although Singapore billed itself as a free-enterprise economy, the
economic role of government was pervasive. As governing body for both
the nation and the city, the government was responsible for planning and
budgeting for everything from international finance to trash collection.
The government owned, controlled, regulated, or allocated land, labor,
and capital resources. It set or influenced many of the prices on which
private investors based business calculations and investment decisions.
State intervention in the economy had a positive impact not only on
private business profitability but also on the general welfare of the
population. Beyond the jobs created in the private and public sectors,
the government provided subsidized housing, education, and health and
recreational services, as well as public transportation. The government
also managed the bulk of savings for retirement through the Central
Provident Fund and Post Office Savings Bank. It also decided annual wage
increments and set minimum fringe benefits in the public and private
sectors. State responsibility for workers' welfare won the government
the support of the population, thus guaranteeing the political stability
that encouraged private investment. In general, state intervention in
the economy managed to be probusiness without being antilabor, at least
regarding material welfare.
Budgeting and taxation were frequently used for attaining economic
goals. In the postrecession period, budgetary changes primarily
benefited business. For example, the fiscal year ( FY) 1988 budget
included an overseas investment incentive program, administered by the
Economic Development Board, allowing tax write-offs for losses from
approved overseas investments. Other concessions such as suspension of
taxes on utilities and a 50 percent rebate on property taxes were in
effect between 1985 and 1988 to counteract the economic slump.
Budgeting and taxation also were often used to achieve or reinforce
social goals such as population control. Until 1984 the government
encouraged limiting of families to two children by levying higher
medical and education costs for additional children. In 1986, however,
tax rebates were introduced to encourage collegeeducated women to have
third and fourth children.
<"30.htm">Economic Boards
<"31.htm">Land Management and Development
<"32.htm">Forced Savings and Capital Formation
<"33.htm">State-Owned Enterprises
Singapore
Singapore - Economic Boards
Singapore
Under the appropriate government ministries, statutory boards-- a
concept carried over from colonial days--were established to manage
specific parts of the economy and foster overall and sectoral
development. Each worked somewhat autonomously, using a hands-on
approach to the problems in the areas in which it operated.
Economic Development Board
The Economic Development Board was established in 1961 to spearhead
Singapore's industrialization. Initially its function was to promote
industrial investment, develop and manage industrial estates, and
provide medium- and long-term industrial financing. The latter function
was taken over in 1968 by the newly created Development Bank of
Singapore. When the limits of import substitution became evident, given
the small domestic market, policy was redirected toward promoting an
export-oriented, labor-intensive industrialization program. After 1986
the board's portfolio was enlarged to include the promotion of services
in partnership with other government agencies responsible for the
various service sectors and the development of local small- and
medium-sized enterprises. In the first two decades following
independence, the board evolved industrial strategies in response to
changes in the international and domestic business environments, as well
as negotiating the public-private consensus necessary for implementing
them. The board was not an economic tsardom but, rather, a consensus
maker among agencies and corporations that commanded larger financing.
In 1989 the Economic Development Board focused its attention on
attracting investments in manufacturing and other high value-added
services, which met the technological skills and employment needs of
Singapore's future economic development.
Small Enterprise Bureau
The Small Enterprise Bureau was established in 1986, following the
economic slump, when the government realized the importance of
developing and upgrading local small- and medium-sized enterprises. The
bureau worked closely with the Economic Development Board and managed a
number of assistance programs, some of which predated the bureau.
Emphasis was placed on helping local firms to improve and modernize
their plants and technology, product design, management skills, and
marketing capabilities. Launched in 1976, the Small Industry Finance
Scheme provided low-cost financing to local smalland medium-sized
enterprises in manufacturing and related support services. In 1985 this
program was extended to the nonmanufacturing sector, and in 1987 some
1,125 loans amounting to S$297 million were approved by the Economic
Development Board under the plan. The Small Industry Technical
Assistance Scheme, introduced in 1982, provided grants to defray part of
the cost of engaging short-term consultants and increasing or
establishing in-service training for employees.
National Productivity Board
The National Productivity Board (NPB) was established in 1972 to
improve productivity in all sectors of the economy. Increasing
individual and company productivity at all levels was a government
priority, given Singapore's full employment picture and relatively high
wages. Greater worker productivity than the country's neighbors and
competitors was viewed by the government as a necessity as well as one
of Singapore's major advantages.
The National Productivity Board followed a "total
productivity" approach, which emphasized productivity measurement,
product quality, a flexible wage system, worker training, and assistance
to small- and medium-sized enterprises. In order to promote productivity
in both the public and private sectors, the board used mass media
publicity, seminars, conventions, and publications to remind
Singaporeans that productivity must be a permanent pillar of the
economy. The board sponsored a productivity campaign each year with such
slogans as the one for 1988, "Train Up--Be the Best You Can
Be."
The National Productivity Board offered management guidance services
to small- and medium-sized enterprises to assist them in improving their
productivity and efficiency, as well as referring companies to private
management consultancy services available in Singapore. Beginning in the
early 1980s, the board also spearheaded campaigns to introduce
productivity management techniques used extensively by Japanese business
and industry, such as quality control circles.
Trade Development Board
Changes in world trade patterns and in what the government viewed as
an increasingly protectionist international trade environment prompted
the establishment of the Trade Development Board in 1983 as a national
trade promotion agency. Based on the recommendations of specialists, the
board formulated policies reflecting the needs of traders in general, as
well as the specific needs of particular trade sectors. Initial areas of
focus were trade facilitation of electronics, printing and publishing,
textiles, and timber products. The Trade Development Board reviewed
existing marketing policies, strategies, and techniques and explored new
opportunities in both traditional and nontraditional markets. The board
assisted both local and foreign companies interested in using Singapore
as a base for such trading activities as warehousing and distribution.
The Trade Development Board also helped Singapore companies market their
products by assisting them in improving their product designs.
Singapore
Singapore - Land Management and Development
Singapore
One of the government's most important roles was the oversight of
land use and development. This was a particularly critical issue given
the country's minute size and dense population; a total land area of 636
square kilometers and a population density of 4,166 per square kilometer
made Singapore one of the most densely populated countries in the world.
As pressure for economic growth increased, optimization of land use
became more critical.
Housing and Development Board
Central to the issue of land management was another statutory board,
the Housing and Development Board, established in 1960. Between 1960 and
1985, the government-owned board completed more than 500,000 high-rise,
high-density public housing apartments-- known as housing estates--along
with their related facilities were completed. By comparison, the British
colonial government's Singapore Improvement Trust had completed only
23,000 apartments in its thirty-two years of existence (1927-59). From
1974 to 1982, the Housing and Development Board built and marketed
middle-income apartments, an activity which became a function of the
board after 1982.
By 1988 the Housing Development Board was providing housing and
related facilities for 88 percent of Singaporeans, or some 2.3 million
people--a feat that has been called urban Singapore's equivalent of
"land reform." Government encouragement of apartment ownership
was both an economic and a "nation building" goal because
individual ownership would ultimately pay for the program while giving
citizens a "stake in Singapore." The board also provided
estate management services and played an active role in promoting the
advancement of construction technology. As one of the country's major
domestic industries, housing construction served as an important
economic pump primer.
Home owners were encouraged to use their Central Provident Fund
savings to pay for the apartments. The factors determining the selling
prices of apartments included location, construction cost, ability of
the applicants to pay, and the practical limits to government subsidies.
Resettlement policies aimed at equitable payments, minimal readjustment,
and real improvement in housing conditions. In social terms, attention
was paid to providing an environment conducive to community living,
integrating the population, preserving the traditional Asian family
structure, and encouraging upward social mobility by providing
opportunities for home upgrading.
Starting with a capital expenditure of S$10 million in 1960, the
Housing and Development Board's annual capital expenditures rose to
about S$4 billion by 1985. The board's capital budget, with funds
obtained in the form of low-interest government loans, represented 40
percent of the government's capital budget. Selling prices, rent rates,
and maintenance charges were determined by the government, and the board
received an annual subsidy of 1 to 2 percent of the government's main
operating expenditure.
Urban Renewal Authority
In 1974 the Housing and Development Board's Urban Renewal Department
was made a statutory board and named the Urban Redevelopment Authority.
Responsible for slum clearance and comprehensive development of the
city's Central Area, the authority was to plan, guide, and implement
urban renewal. The Urban Renewal Authority drew up long-term land-use
plans, which it implemented through its own development projects as well
as the Sale of Sites Programme. The latter, a key instrument in the
government's comprehensive redevelopment plans, represented a
partnership between the public and private sectors. The public sector
provided initiative, expertise, and infrastructural services; the
private sector contributed financial resources and entrepreneurship to
facilitate the completion of projects. Between 1967 and 1983, some 166
parcels of land were turned into 143 projects for residential, office,
shopping, hotel, entertainment, and industrial developments.
Jurong Town Corporation
The primary responsibility for acquiring, developing, and managing
industrial sites, however, belonged to the Jurong Town Corporation,
established in 1968. The corporation provided manufacturers with their
choice of industrial land sites on which to build their own factories or
ready-built factories for the immediate start-up of manufacturing
operations. In the 1950s, when the idea of establishing an industrial
estate was first conceived, Jurong was an area of dense tropical forests
and mangrove swamps on the southwestern quadrant of the island, and it
was not until 1960 that the government decided to undertake the project.
During the first few years, entrepreneurial response was disappointing,
but after independence the pace of development accelerated. By 1989
Jurong had quadrupled its original size, and the corporation also
managed twenty-three other industrial estates, including the Singapore
Science Park, a research and development park adjacent to the National
University of Singapore. Although the emphasis in the 1970s had been on
the development of labor-intensive industries, in the 1980s priority was
given to upgrading facilities to make them more attractive for the
establishment of high value-added and high technology industries.
The industrial estates were designed to be self-contained urban
centers and included such facilities as golf courses, banks, shopping
centers, restaurants, child-care centers, and parks. As of 1988, they
contained some 3,600 factories employing a total of 216,000 workers. The
Jurong Town Corporation also provided infrastructure and support
facilities, including the Jurong Industrial Port, which was the
country's main bulk cargo gateway, and the Jurong Marine Base, which
serviced offshore petroleum operations.
The Jurong Town Corporation shared responsibility for coastal
planning and development control with the Housing and Development Board,
the Urban Renewal Authority, and the Port of Singapore Authority. The
coastal zone, dominated by its entrep�t facilities, was the traditional
foundation on which Singapore's economy was built. Between 1965 and
1987, the coastal zone was enlarged by about fifty square kilometers
through reclamation of tidal flats, shallow lagoons, and wetlands. The
two largest landfill operations were the East and the West Coast
Reclamation schemes adjoining the Central Business District. The former
was the Housing and Development Board's largest project, in which a
"sea city" almost the size of the present-day downtown area
had been developed by both the private and public sector. Experts
estimated that in the 1980s Singapore, including the offshore islands,
had the potential of increasing its existing land resources by about 10
percent.
Singapore
Singapore - Forced Savings and Capital Formation
Singapore
Singapore's much-vaunted savings rate--and much of the funding for
development, particularly public housing--resulted in large measure from
mandatory contributions to the Central Provident Fund, as well as
voluntary deposits in the Post Office Savings Bank. The Central
Provident Fund was set up in 1955 as a compulsory national social
security savings plan to ensure the financial security of all workers
either retired or no longer able to work. Both worker and employer
contributed to the employee's account with the fund. The rate of
contribution, which had gradually risen to 50 percent of the employee's
gross wage (coming equally from employer and employee), was lowered to
35 percent in 1986. In 1987 new long-term contribution rates were set
calling for 40 percent for employees below fifty-five years of age, 25
percent for those fifty-five to fifty-nine, 15 percent for those sixty
to sixty-four, and 10 percent for those over sixty-five, with equal
contributions coming from employee and employer. A series of transition
rates leading to the new long-term rates were first applied in 1988. The
contributions were tax-exempt and subject to maximum limits based on a
salary ceiling. Beginning in 1986, the government paid a market-based
interest rate on Central Provident Fund savings (3.19 percent per year
in June 1988).
Every employed Singaporean or permanent resident was automatically a
member of Central Provident Fund, although some self-employed people
were not. Membership grew from 180,000 in 1955 to 2.08 million in 1989.
At the end of 1988, the 2.06 million members of the Central Provident
Fund had S$32.5 billion to their credit. That same year, a total of
S$2,776 million was withdrawn to purchase residential properties; S$9.8
million was paid under the Home Protection Insurance Scheme; S$1,059
million was paid under the Approved Investments Scheme; and S$13.7
million was withdrawn for the purchase of nonresidential properties.
Each member actually held three accounts with the Central Provident
Fund: Ordinary, Special, and, since the mid-1980s, Medisave Accounts.
The first two were primarily for old age and contingencies such as
permanent disability. The Ordinary Account, in addition, could be used
at any time to buy residential properties, under various Housing and
Development Board programs, and for home protection and dependents'
protection insurance. Two further programs were established in 1987: a
Minimum Sum Scheme, which established a base amount to be retained in
the account against retirement, and a Topping-up Extension under which,
as well as adding to their own, members could demonstrate "filial
piety" by adding to their parents' accounts. Since the late 1980s,
members could use their accounts to buy approved shares, loan stocks,
unit trusts, and gold for investment. Part of the rationale for the
latter was to allow Singaporeans to diversify their savings and to gain
experience in financial decision making.
Although comparable to social security programs in some Western
countries, the Central Provident Fund's concept and administration
differed. Rather than having the younger generation pay in while the
older generation withdrew, whatever was put into the Central Provident
Fund by or for a member was guaranteed returnable to that person with
interest.
Thus, at the individual level, Central Provident Fund savings
promoted personal and familial self-reliance and financial protection,
an economic attitude constantly encouraged by government leaders.
Collectively, the Central Provident Fund savings assured the government
of an enormous, relatively cheap "piggy bank" for funding
public-sector development; the savings also served as a mechanism for
curtailing private consumption, thereby limiting inflation. The result,
according to some critics, was that the city-state had become
overendowed with buildings, with too few productive businesses to put in
them. They also noted that the bloated size of the Central Provident
Fund (S$32.5 billion in 1988, equivalent to 82 percent of the GDP) was
the most important factor behind the unwieldiness of public savings.
Some analysts advised that the fund was beginning to outlive its
usefulness and should be dismantled and replaced by private pension
funds and health insurance plans. As a result, they stated, savings
would be channelled to private businessmen rather than to bureaucrats.
Singapore
Singapore - State-Owned Enterprises
Singapore
Over time, the statutory boards not only became major actors in the
economy but also formed subsidiary companies to add flexibility to their
own operations. For example, in 1986 the Singapore Broadcasting
Corporation formed a subsidiary to produce commercials on a
fee-for-service basis. The government entered other areas of the economy
that it considered appropriate, exerting leadership, assuming risk, and
not hesitating to withdraw its support or close down unprofitable
companies.
Numerous state and quasi-state companies were created either directly
by ministries or, more often, organized under three wholly owned
government holding companies (Temasek Holdings (Private) Limited, MND
Holdings, and Sheng-Li Holding Company), which provided a wide range of
goods and services. Joint ventures between the government and both
domestic and foreign partners produced several industrial products,
including steel and refined sugar. In addition, the National Trades
Union Congress (NTUC), which was closely tied to the government, ran
many cooperative businesses, including supermarkets, taxis, and a travel
agency.
Although these companies collectively contributed significantly to
the growth of the economy, neither their total amount of profits nor
their rate of return on investment could be documented. In 1983 some 450
such companies, excluding subsidiaries of the statutory boards, employed
58,000 workers, or 5 percent of the labor force. In 1986 there were
approximately 500 such companies still active. These different
institutional forms permitted versatility.
Singapore
Singapore - LABOR
Singapore
Singaporeans themselves were universally viewed as the nation's best
natural resource. In 1989, however, the work force was a shrinking
resource. The high rate of economic growth combined with an increasing
number of Singaporeans over the retirement age of fifty-five (nearly 12
percent) and a lower-than-replacement birth rate had resulted in a
significant labor shortage. By the end of the century, the labor market
was projected to be even tighter. According to the Ministry of Health,
the fifteen to twenty-nine age-group would decline 25 percent, from
816,000 in 1985 to 619,000 in the year 2000.
In 1987 and 1988, slightly more than six Singaporeans out of ten were
working or looking for work. Men's rate of participation, 79 percent,
remained steady. Women, however, responding to job opportunities in the
manufacturing and commercial sectors, were increasingly entering the
labor market (48 percent in 1988, up from 47 percent in 1987, 40 percent
in 1978, and 24.6 percent in 1970). Job-switching was rampant,
particularly in manufacturing, where a 1988 survey showed that three out
of four new workers quit within the month they were hired. Higher wage
and input costs, as well as job-switching, resulted in a decline in the
growth of manufacturing productivity (2.4 percent in 1988 compared with
3.7 percent in 1987 and 13.6 in 1986). The labor market, then, was at
the center of challenges facing the Singaporean economy. The nature of
the concern about the labor market had been almost totally reversed
since independence. The early 1960s were a time of labor unrest, and
unemployment was still about 10 percent by 1965. By the late 1960s,
however, there was substantial industrial peace, which had continued
through the 1970s and 1980s. With unemployment at a very manageable 3.3
percent in 1988, the government's attention was focused on other aspects
of the labor market.
Industrial Relations and Labor Unions
Industrial relations in Singapore reflected the symbiotic
relationship between the labor movement and the dominant political
party, the People's Action Party (PAP), a relationship rooted in a
political history of confrontation that evolved into consensus building.
Trade unions were a principal instrument in the anticolonial struggle
used by both the democratic socialist PAP and the communists with whom
they cooperated uneasily. In 1961 the Singapore Trade Union Congress
split into the left-wing Singapore Association of Trade Unions (SATU)
and the noncommunist National Trades Union Congress (NTUC). The NTUC
quickly became the leading trade union organization, largely because of
its effectiveness and government support. Moreover, in 1963, when SATU
led a general strike against the government, the pro-communist trade
organization was banned and many of its leaders were arrested.
Strong personal ties between leaders of the PAP and the NTUC formed
the background of the symbiotic relationship, which was
institutionalized by formal links. In 1980 NTUC Secretary General Ong
Teng Cheong was made a minister-without-portfolio, and a NTUCPAP Liaison
Committee comprising top leaders of both organizations was established.
As the "second generation" political leaders assumed more
government leadership following the 1984 election, Ong was named second
deputy prime minister. Following the September 1988 general elections,
the NTUC reaffirmed its close relationship with the PAP by expelling
officers of NTUC-affiliated unions who had run for Parliament on
opposition tickets. The NTUC and the PAP shared the same ideology,
according to NTUC officials, so that active support of the opposition
was inconsistent with membership in NTUC-related institutions. Workers
who did not support the PAP were advised to form their own unions.
The legal-institutional framework also exerted control over labor
conditions. In mid-1968, in an attempt to woo private foreign
investment, Prime Minister Lee Kuan Yew successfully pushed through
Parliament a new employment bill and amendments to the 1960 Industrial
Relations Act. In order to make factors such as working hours,
conditions of service, and fringe benefits predictable, and thus make
businesses sufficiently attractive for investors, trade unions were
barred from negotiating such matters as promotion, transfer, employment,
dismissal, retrenchment, and reinstatement, issues that accounted for
most earlier labor disputes. To spread work and help alleviate the
effects of unemployment, overtime was limited and the compulsory
retirement age was set at fifty-five. Lee's actions, which the militant
unions opposed but could do little about, were part of the government's
efforts to create in Singapore the conditions and laissez-faire
atmosphere that had enabled Hong Kong to prosper. Such measures, in the
government's view, were necessary to draw business to the port. Lee
stressed survival, saying: "No one owes Singapore a living."
Rapid economic growth in the late 1960s and early 1970s reduced
unemployment and resulted in the amendment of these laws. A National
Wages Council was formed in 1972 and many of its recommendations
adopted. By 1984 a twelve-hour shift was permitted. In order to enlarge
the limited labor pool, in 1988 changes were introduced in Central
Provident Fund policies reducing payment rates for those over
fifty-five, thereby encouraging employers to raise the retirement age to
sixty. The discipline imposed on, and expected of, the labor force was
accompanied by provisions for workers' welfare. The Industrial
Arbitration Court existed to settle disputes through conciliation and
arbitration. The court, established in 1960, played a major role in
settling labor-management disputes through binding decisions based on
formal hearings and through mediating voluntary agreements. Adjudication
of disputes between employers and nonunion workers came under the
separate jurisdiction of the Labour Court. To help job seekers, the
government maintained a free employment service serving both job seekers
and employers. A comprehensive code governed the safety and health of
workers and provided a system of workers' compensation. Under the
Ministry of Labour, the Factory Inspectorate enforced these provisions
in factories, where more than 35 percent of Singapore's workers were
employed in 1988.
The trade unions' role and structure also had been modified. In the
1970s, the NTUC began establishing cooperatives in order to promote the
welfare of its members. In the 1980s, omnibus unions were split along
industry lines and further split into house unions to facilitate better
labor-management relations and promote company loyalty. In the 1982
Amendment to the Trade Union Act, the role of trade unions was defined
as promoting good industrial relations between workers and employers;
improving working conditions; and improving productivity for the mutual
benefit of workers, employers, and the country.
Union membership declined steadily beginning in the late 1970s. In
1988 there were some 83 registered unions, with about 1,000 branch
locals, representing one-quarter of the organizable work force. This
number was down from ninety unions in 1977. Increasing emphasis on
developing white-collar, capital-intensive, and service-oriented
industries was partly responsible for the union membership decline. The
unions were countering the decline by offering attractive packages to
bring in new members.
Wage Policies
Following the rapid economic growth of the late 1960s and early
1970s, signs of a tight labor market emerged along with a concern that
wages might escalate. In response, the government in 1972 established
the National Wages Council, a tripartite forum with representation from
the employers' federations, trade unions, and the government. As a
government advisory body, the council recommended annual wage increases
for the entire economy; ensured orderly wage development so as to
promote economic and social progress; and assisted in the development of
incentive schemes to improve national productivity.
The wage guidelines were not mandatory but were followed by the
public sector (by far the largest employer) and widely implemented in
the private sector. The influence of these recommendations generally was
not applicable to private-sector professional and managerial workers,
whose wages were determined more by international forces, but was more
important for non-professional white-collar workers. For blue-collar
workers, who constituted about 40 percent of the labor force in both the
public and private sector, union influence was more crucial than the
National Wages Council's recommendations, but market forces were even
more important.
Between 1973 and 1979, actual wage increases followed the council
recommended wage increases closely. In 1979 the "wage correction
policy," in which there were three years of high-wage
recommendations, was designed to force an increase of the productivity
of higher value-added operations, to reduce the reliance on cheap
unskilled foreign labor, and to raise labor productivity. From 1980 to
1984, however, actual wage increases exceeded the recommendations by an
average of 2.4 percentage points per year, as the increasingly heavy
demands for labor apparently outstripped its supply. Additionally,
collective agreements for unionized workers lasted for two or three
years with built-in wage increases. Although starting pay was relatively
low, large gaps in wages were institutionalized through longevity of
employment and annual raises.
The effect of wage increases, compounded by a further raise in the
mandatory Central Provident Fund component of wages, was to price
Singapore out of the market. High wages were a major contributor to
Singapore's 1985 recession. Consequently, in 1986 and 1987 the
government instituted a wage restraint policy: wages were frozen and the
employer's contribution to the fund substantially reduced. The policy's
relative success could be attributed to close government-labor ties and
to the tripartite forum of the National Wages Council.
Proposals for wage reform--a "flexi-wage policy"--were
announced in mid-November 1986 and became effective with the enactment
of the 1988 Employment (Amendment) Act. Under this plan, the basic wage
remained relatively stable with adjustments for good or bad years made
by increasing or reducing the annual bonus. Negotiating the size of the
bonus--frozen to the equivalent of one month's salary since 1972--was
left to employers and unions, who would be able to bargain for its
retention, abolition, or modification. Profit-sharing, productivity
incentive, and employee share plans were encouraged to ensure that high
wage payments awarded in fat years were not perpetuated in lean years
and that individual as well as company productivity, growth,
profitability, competitiveness, and prospects for the industry were
taken into account. The government was anxious that wages not increase
precipitously. This concern was shared by management, which worried
about shrinking profit margins resulting from higher operating costs.
Workers, on the other hand, wanted to share in the benefits of the
economic boom after giving up wage increases to help cope with the 1985
recession.
Foreign Labor
Two groups comprised foreign nonresident labor in Singapore. The
majority were unskilled work-permit holders who could only enter and
work in the country if their prospective employers applied for work
permits for them. Skilled workers and professionals on employment passes
comprised the other group.
Work permits were for a short duration with no guarantee of automatic
renewal. Malaysia, particularly the southernmost state of Johor, was the
traditional source of such workers. Singapore's tight immigration policy
was relaxed as early as 1968 to allow in these workers. At the peak of
the economic boom in 1973, noncitizen work-permit holders reportedly
accounted for about one-eighth of the total work force. Large numbers of
these "guest workers" were repatriated during the 1974-75
world recession because of retrenchments, particularly in the
labor-intensive manufacturing industries.
With the tightening of the labor market in 1978-79, it became more
difficult to fill less desirable jobs with domestic labor or labor from
Malaysia, which also had a tight job market. Foreign workers were then
recruited from Indonesia, Thailand, Sri Lanka, India, Bangladesh, and
the Philippines. By 1984 workers from South Korea, Hong Kong, Macao, and
Taiwan were being allowed in, on the basis that their Confucian cultural
background might enable them to adapt more readily than immigrants from
other cultures.
The increase in foreign workers was remarkable; by 1980 they
comprised 7 percent of the total compared with 3 percent a decade
earlier. No figures on foreign labor were published after 1980.
According to the 1980 census, 46 percent of the foreign workers were in
manufacturing, 20 percent in construction, and 9 percent in personal and
household services. The recession led to a repatriation of some 60,000
foreign workers in 1985, two-thirds of the total employment decline. The
foreign worker levy was raised to S$250 per month in July 1989, and the
maximum foreign worker dependency at the firm level was reduced from 50
percent to 40 percent. Both measures were designed to encourage firms to
speed up automation of labor-intensive operations in order to reduce
reliance on foreign workers.
Manpower Training
The main goals of manpower training were to increase the average
skill level of the labor force and, at the same time, provide sufficient
numbers of workers with the specialized skills necessary to meet future
industrial needs. Beginning in the late 1970s, the government placed
increased stress on education in order to achieve the objective of
industrial restructuring. As of 1987, however, Singapore's work force
was less educated than that of some of the countries with which it
competed. Five percent of the work force had university educations
compared with 19 percent for the United States and Japan and 6 percent
for Taiwan. Some 11 percent had received post-secondary schooling other
than in universities, compared with 46 percent for Japan, 23 percent for
Taiwan, and 16 percent for the United States.
In the early 1980s, government studies showed that about half of the
work force had primary-level education or less, and many older workers
had low levels of English language skills. To remedy this situation, the
Basic Education for Skills Training (BEST) program was introduced in
1984 to provide opportunities for workers who had not completed primary
education to improve their English and math. By 1989 some 116,300
workers (half the target group) had had some BEST training. Time was
also solving the problem as younger people received more education and
the older, less-educated workers passed out of the work force; between
1979 and 1984, entrants to the work force with only primary-level
education or less declined from 43 percent to 26 percent. The government
needed, however, to ensure that this better-educated work force was
trained in the necessary skills to complete the transformation of
Singapore from a labor-intensive economy to a high-technology
city-state--a "technopolis."
A further problem in achieving this transition resulted from
"government brain drain." Each year 50 to 60 percent of new
university graduates were absorbed by the government, including
government-owned companies and the statutory boards. A system of
awarding undergraduate scholarships, which often tied the awardees to
eight years of government service, assured that the public sector
absorbed many of the top-ranking students. Some critics thought that
this concentration of the country's valuable human resources in the
public sector might be to the long-run detriment of entrepreneurial and
private-sector development.
Singapore
Singapore - INDUSTRY
Singapore
Industrialization Policy
The manufacturing sector was a mainstay of Singapore's economic
growth despite the absence of natural resources or an agricultural base.
By the mid-1970s, the country had undergone a quarter-century of rapid
industrial advance based on low-cost labor, low- to middle-level
technology, and a rapid increase in exports. At that time, Singapore's
planners settled on a policy emphasizing high technology, particularly
information technology. In 1988 Singapore's 3,694 manufacturing
establishments, employing 352,600 workers, were responsible for 29
percent of the GDP. Industrial production, valued at S$14,509.7 million,
was fractionally higher than earnings from financial and business
services, double those from commerce, and nearly equal to the total of
commerce and transport and communications. This represented a 20-percent
increase over 1987. The manufacturing sector's continuing success was
largely a function of Singapore's ability to attract foreign investment
through a favorable business climate and then provide investors with an
educated, trained, and disciplined labor force.
Singapore entered nationhood with a mixed legacy. The industrial
sector was small, its productivity low. Manufacturing in 1960 was a mere
11.4 percent of the GDP; commerce, far and away the largest sector,
accounted for 32 percent. The industrial policy in 1959 sought to
promote industrialization as a way of diversifying from Singapore's
traditional role as an entrep�t. Reliance was placed on private
enterprises whose basic decisions were determined on the expectation of
a common market with the neighboring Federation of Malaya. A system of
import quotas was introduced for a limited number of goods, along with
controls on how many enterprises could enter a particular field.
Circumstances altered strategies. After separation from Malaysia in
1965, quotas were mainly replaced by a low level (for developing
countries) of protective import tariffs. A traditional import
substitution strategy was implemented.
In 1968, when the British announced their intention to withdraw from
their Singapore bases, import substitution was succeeded by a strategy
promoting export-oriented, labor-intensive industrialization. At that
time, the government began its central role in formulating and
implementing the industrialization program through the Economic
Development Board.
The new approach became official policy in 1967 with the government's
proclamation of the Export Expansion Incentives (Relief from Income Tax)
Act and was further enhanced by the 1968 Employment Act. Direct foreign
investment was welcomed both to help Singapore penetrate export markets
and to bring in advanced technology. As early as 1970, when full
employment was attained, there was some thought given to upgrading the
industrial structure in order to provide more higher paying jobs. By
1979 efforts to upgrade the overall industrial structure and to
accelerate the trend toward skill- and technology-intensive, higher
value-added economic activity were intensified. The government
implemented the large, three-year wage increases recommended by the
National Wages Council, which began the easing out of labor-intensive,
low valueadded activities in Singapore.
The machinery industry was increasingly in the forefront of
technological innovation as a result of the Economic Development Board's
promotion of computer-controlled production, industrial robots, and
flexible manufacturing systems. The industry's output increased by 17
percent in 1987 and 20 percent in 1988.
Domestic enterprises played a lesser role in industrialization. The
government argued that the emphasis on large industry was a more
effective stimulus to increased productivity and long-range economic
development. Major promotional efforts sponsored by the government were
focused on high-productivity projects, creating industries that
officials claimed would not otherwise have been established in
Singapore. Although institutional assistance for small-scale local
industry, the majority of enterprises, was provided through a subsidiary
of the Economic Development Board, the effectiveness of this aid was
limited until after the mid-1980s recession, when greater emphasis was
placed on encouraging and upgrading small-scale local industry.
Following a decline in the textile industry in the mid-1980s
resulting from increased international competition, automation and the
upgrading of product lines were encouraged. What had originally been a
textile industry and then a mass-market clothing industry was encouraged
to target high-fashion markets. A 10 percent growth in the fashion
industry in 1987 reflected both the new trend and a strong market among
Western trading partners.
Information Technology
After 1979 there was a single-minded emphasis among policy makers on
escalating the level of technology in order to implement the succeeding
phases of Singapore's industrial revolution. They relied on information
technology as the strategy's principal instrument. The
Telecommunications Authority of Singapore (Telecoms) was a key to the
strategy because of the high caliber of its services and products and
because Telecoms and the telecommunications industry had an important
role in the progress of every industry in Singapore.
A second key was computers and related electronics, which in the late
1980s constituted Singapore's largest industry, measured both in numbers
of jobs and in value added by manufacturing. In 1981 the 65,000 to
70,000 electronics workers comprised about 7 percent of the labor force;
gross production of electronics at about S$5.9 billion was about 15
percent of total manufacturing output. By 1987 electronics accounted for
28 percent of manufacturing employment and contributed 31 percent or
S$11 billion in output. By 1989, Singapore had become the world's
largest producer of disk drives and disk drive parts. Other related
products included integrated circuits, data processing equipment,
telecommunications equipment, and radio receivers.
The electronics industry began a calculated transition away from
labor-intensive products toward higher technological content and
worker-skilled products in 1974. Potential investors were encouraged to
look elsewhere for low-wage, unskilled labor. Aside from producing high
value-added exports, the computer and electronics industries played a
vital role in raising manpower productivity in other
technology-intensive industries through computerization and computer
communications. The National Computer Board was formed in 1981 to
establish Singapore as an international center for computer services, to
reduce the shortage of trained computer professionals, and to assure
standards of international caliber at all levels.
Copyright and "intellectual property" issues served as an
impediment to computer and other industrial development in the early
1980s, when Singapore, as well as other Asian countries, was known for
producing pirated versions of everything from computers and computer
software to designer handbags. Following threats by their major Western
trading partners to impose trade sanctions and by international computer
and software companies not to do business, Singapore passed its first
copyright law in 1986. There was fairly rigorous enforcement in areas in
which Western pressure was applied (computer software, films, and
cassette tapes), and nearly full compliance in the book trade, which had
not been as serious a problem. The Asian "copyright
revolution" (Singapore's was one of several such laws enacted in
the region) was significant as a realization by those countries that
they had joined the international knowledge network as producers as well
as consumers.
By the mid-1980s, the small but growing printing and publishing
industry had entered the high-technology world with computerized
typesetting, color separation, and book binding. Its high-quality
printing facilities and sophisticated satellite telecommunications
network made Singapore a regional publishing and distribution center in
1989.
Petroleum
Petroleum and petrochemicals were another base of Singapore's
industrial and economic life. In the late 1980s, Singapore was the
world's third largest oil-trading center and also the third largest
center for petroleum refining. It was the second largest builder of
drilling rigs, and its facilities for repairing and maintaining rigs and
tankers were the most competitive in East Asia.
When oil prices began eroding in 1981 and collapsing toward the end
of 1985, Singapore felt both negative and positive consequences. The
collapse of oil prices dealt a severe blow to oil exploration. The
impact was felt widely and immediately in everything from reduced orders
for rig construction to lowered occupancy of luxury apartments as
foreign petroleum workers returned home. With both of its immediate
neighbors, Indonesia and Malaysia, heavily dependent on oil and gas
exports for revenue, Singapore had a resulting loss of trade in both
goods and services.
Singapore benefited, however, from the availability of cheaper
energy, which in 1986 amounted to a savings of about S$2.5 billion
(US$1.12 billion). Furthermore, Singaporean refineries invested in the
equipment and technology necessary to enable them to refine a wide
variety of crude oils and obtain a greater proportion of highvalued
products from the refining process. Petroleum refining alone made up 28
percent of Singapore's manufacturing output in 1985, although by 1988 it
had dropped by half as a result of a decline in petroleum production and
growth in other industries. Singapore also benefited indirectly when
large oil importers such as Japan and the United States obtained higher
real incomes from lower oil prices, enabling them to increase their
imports from Singapore and other countries.
Singapore
Singapore - TRADE
Singapore
Foreign Trade
Trade in goods and services was Singapore's life blood as truly in
1989 as it was in the early twentieth century or a century earlier when
the British East India Company first began business there. Trade, along
with domestic savings and foreign investment, remained key to the
country's growth. Singapore traditionally had a merchandise-trade
balance deficit (in part at least because food was imported), which it
customarily offset with a surplus on the services account. It was one of
the world's few countries where total international trade (domestic
exports and reexports plus imports) was greater than total GDP. In 1988
trade (S$167.3 billion) was more than three times GDP (S$48 billion),
and two-thirds of the goods and services Singapore produced were
exported.
Singapore, however, was more than simply a trade and manufacturing
center in the late 1980s. Trade and manufacturing were closely tied to
the country's expanding business services and international financial
market; each enhanced the other. In addition to the more than 650
multinational companies that had set up manufacturing plants and
technical support facilities, several thousand international financial
institutions, service companies, and trading firms also maintained a
presence in Singapore. The increasing internationalization of the
economy and the continuing centrality of external trade meant that world
trade fluctuations and the state of the global economy were significant
factors-- largely out of the country's direct control--in what happened
to Singapore's trade and wider economy.
As a British colony in the nineteenth and early twentieth centuries,
Singapore was an entrep�t for the exchange of raw materials from
Southeast Asia--mainly present-day Indonesia and Malaysia--for European
merchandise. Newly independent Singapore's decision in 1965 to emphasize
industrial development and the growing success of that plan gradually
resulted in a significant change in the nature of trade. By the
mid-1970s, the proportion of reexports and domestic exports had been
roughly reversed, with reexports accounting for less than 41 percent.
In the 1980s, the somewhat diminished entrep�t trade remained
important, and Singapore continued to act as a regional processing and
distribution center. Reexports' share of total exports averaged 35
percent from 1980 to 1987. Although primary commodities (crude rubber,
nonferrous metals, and to a lesser extent palm and coconut oil) were
still a factor in trading activities, machinery and transportation
equipment dominated. Singapore also served as a back door to trade with
Asian communist countries for third countries, such as Indonesia.
Between 1980 and 1984, total exports grew an average of 5.5 percent
per year. The strongest impetus came from the newer electrical and
electronics industries. The trade deficit declined steadily after 1982,
reflecting lower commodity prices paid to foreign producers, greater
levels of internal efficiency, and industrial upgrading. In 1985,
however, total exports decreased by 2.26 percent. Higher value-added
exports declined, both as a function of weaker demand and a worldwide
saturation in many areas, such as computer peripherals. Petroleum
exports, still a major sector, virtually stagnated.
Trade, along with the rest of the economy, reasserted itself by 1987,
resulting partly from government economic decisions and partly as a
reflection of rising world commodity prices. In 1988 Singapore's total
trade amounted to about S$167.3 billion (US$80.8 billion), with a global
trade deficit of about S$8.18 billion. Singapore's GDP grew by 10.8
percent in 1988, the best growth rate in fifteen years. Disk drives were
the largest non-oil item exported, worth S$4.89 billion. Other major
exports were integrated circuits, data processing equipment and parts,
telecommunications equipment, radio receivers, clothing, and plastics.
By early 1989, signs of slowing down and leveling off had appeared
with the first export declines in eighteen months. Analysts agreed the
weak external demand for electronics and computer parts resulted, in
part, from an oversupply on the world market of disk drives,
semiconductors, and related items. Imports surged, however, widening the
trade deficit sharply.
Although their volume was not large, food products were a significant
aspect of Singapore's trade. The urban nation produced only a small
proportion of its own food, requiring it to import large quantities.
Some food products, such as soy sauce and juices, were processed in
Singapore for export, and Singapore continued its historical role as the
regional center for the spice trade.
Trading Partners
Along with the changes in the composition of trade that had taken
place since independence, there also were changes in direction. The
preeminence of Britain as supplier of manufactures declined after
independence, and by the early 1970s the United States and Japan had
become Singapore's two leading sources of industrial products. Malaysia
and Indonesia remained the principal sources of such primary imports as
crude rubber, vegetable oils, and spices and an important destination
for manufactured exports, including both the products of Singapore and
of the entrep�t trade.
Singapore did not report trade with Indonesia. The omission dated
from the period of the Indonesian Confrontation in the mid1960s and
continued, according to some observers, because Singapore was afraid
that if the Indonesian government knew the volume of the trade, it might
try to curtail it. Estimates were difficult because a substantial part
of the trade was viewed by Indonesia as smuggling and was, therefore,
unlisted, although in Singapore's open export market it was legal.
Nevertheless, trade with Indonesia could be presumed, based partly on
Indonesian trade figures, to have assumed a gradually larger role
starting in the mid-1970s.
As Singapore became more export oriented, its trading patterns became
increasingly complex and interdependent. By the late 1980s, Singapore's
trade links were strongest with the countries of the Organisation for
Economic Co-operation and Development ( OECD), especially the United
States, Japan, and the countries of the European Economic Community (
EEC) or of the Association of Southeast Asian Nations ( ASEAN).
Singapore's drive to industrialization had drawn it increasingly towards
the OECD countries for foreign investment, technology, and markets. To a
large extent, this shift had meant decreasing reliance on its ASEAN
neighbors, particularly for markets and supplies. The other Asian NIEs,
Hong Kong, Korea, and Taiwan, were sometimes viewed as Singapore's
competitors. On the other hand, Singapore engaged in considerable and
growing trade with them, particularly with Taiwan, and all three were a
source of skilled labor.
United States
By the 1980s, the United States had become Singapore's most important
trading partner and, as such, crucial to the country's welfare.
Singaporean officials often stated that a 1 percent drop in the United
States economy had a 1.4 percent effect on Singapore's gross national
product ( GNP). Consequently, in the 1980s Singapore was critically
concerned about protectionist policies and budget deficits in the United
States. In 1988 Singapore's total exports to the United States amounted
to S$18.8 billion, up 28 percent over the previous year, and accounted
for 24 percent of the nation's total exports. Of that total, about 80
percent were Singaporean manufactures, including disk drives, integrated
circuits, semiconductors, parts for data processing machines, television
sets, radios and radio cassette players, and clothing. Reexports to the
United States also were an important part of the trade. Singapore's
exports to the United States outstripped its imports from there,
although the United States was, after Japan, Singapore's second largest
supplier.
Until 1989 Singapore and the three other NIEs enjoyed trade
preferences with the United States under the United States Generalized
System of Preferences (
GSP). This system was originally instituted to aid
developing economies, but in 1989, the four Asian NIEs were removed from
the program because of what some observers have seen as their major
advances in economic development and improvements in trade
competitiveness. The United States had been trying for some time to
wrest trade and currency concessions from all four countries (but
primarily South Korea), which had not been forthcoming. Although
Washington presented the decision more as an economic graduation
ceremony, observers noted that the move reflected United States
frustration over its continuing trade deficit despite considerable
devaluation in the United States dollar.
The removal of the GSP affected less than 15 percent of Singapore's
exports to the United States, among them telephones, office machines,
wood furniture, and medical instruments, which faced duties of 5 to 10
percent. Ironically, United States firms based in Singapore were among
the hardest hit. More than 50 percent of Singapore's exports to the
United States came from American firms with operations there, such as
ATandT, Digital Equipment, Hewlett-Packard, Rockwell International, and
Travenol Laboratories. Singaporean companies, as well as Japanese and
European firms with operations in Singapore, were also affected by the
removal of the GSP. In early 1988, some 4,000 NTUC members gathered
outside the United States Embassy in Singapore to protest the decision,
and the Singaporean government expressed regret.
Japan
Japan's place in Singapore's business picture was underscored by the
fact that, in the 1980s, Japanese were the largest resident expatriate
community in the city. Japan was the country's single largest supplier,
accounting in 1987 for 25.3 percent of total imports, and Singapore's
largest trade deficit was with Japan. Buyback arrangements for products
manufactured by Japanese firms in Singapore also accounted for a
significant part of the trade. Oil accounted for 40 percent of
Singapore's exports to Japan in 1988. Singaporean observers noted by
1989 a significant difference in the market orientation between Japanese
firms and United States-owned multinationals. Japanese firms in
Singapore were producing primarily for the United States and other
third-country markets, rather than for the Japanese home market. The
United Statescontrolled multinationals, on the other hand, produced
mainly for their own home market. Many of these same observers, both
official and unofficial, also expressed the sentiment that the world
export market in the 1990s, would "belong to Japan."
Association of Southeast Asian Nations
The Association of Southeastern Asian Nations (ASEAN) was founded in
1967 primarily as a forum for discussing issues of mutual concern among
neighboring Southeast Asian countries rather than as a trading union
similar to the EEC. In part, this orientation was because, other than
Singapore, most of the ASEAN countries had similar products, tending to
make them more competitive than cooperative. Although trade relations
among the ASEAN countries remained largely bilateral, there was some
informal economic cooperation, including joint representations to
foreign governments on economic issues of common concern. In 1989 the
possibility of a more formalized economic entity was at least being
considered by the ASEAN members.
In 1988 Malaysia was Singapore's largest ASEAN trading partner and
third largest overall trading partner, after the United States and
Japan. The Malaysian market was the single largest ASEAN destination for
Singapore's exports and its second largest export market overall. In the
late 1980s, Singapore established increasingly close economic and
industrial ties with Malaysia's Johor state, which had served as
Singapore's hinterland in colonial times. To alleviate its land shortage
as well as its labor shortage and high labor costs, Singapore, began to
transfer labor-intensive industries to sites across the causeway
connecting it to Malaysia's southernmost state. Johor, in turn, hoped
"economic twinning" with Singapore would boost its long-term
development. By early 1987, there were 217 Singaporean companies or
Singapore-based multinationals in Malaysia, having total investments of
slightly more than S$200 million.
Singapore's much smaller markets with the other ASEAN countries also
were growing. In 1989 Singapore recorded its highest growth in bilateral
ASEAN trade with Thailand, which replaced Taiwan as its fifth largest
trading partner. Intra-ASEAN trade generally might have been
underestimated, partly because of the volume of informal trade,
including smuggling, and partly because so much of it was controlled by
the Chinese community in each country. Keeping business within the
family, clan, or dialect group was a central Chinese business practice
that persisted across national boundaries.
Other Trading Partners
Beginning in the mid-1980s, Singapore--which for two decades had
sharply curtailed many forms of contact with China--began promoting
itself as an alternative to Hong Kong as a "Gateway to China."
In 1989 Singapore was estimated to be the fourth-largest foreign
investor in the special economic zones of southern China and that
country's fifth-largest trading partner; Singapore's companies were
estimated to have about S$1 billion directly invested in China. Since
most such investments were made in conjunction with Hong Kong-based
companies, the real extent of Singapore's exposure to China may have
been considerably higher.
Non-oil trade with the various EEC countries, which had been steady
during the early 1980s, strengthened in 1987 and 1988. Nearly
three-quarters of this increase was in exports of disk drives and
integrated circuits, particularly to the Federal Republic of Germany
(West Germany), Great Britain and the Netherlands. Overall, however,
Singapore had a small trade deficit with Western Europe in 1988.
Singapore
Singapore - Tourism
Singapore
Tourism had been an important sector of Singapore's economy for more
than a decade, averaging 16 percent of total foreign exchange earnings
and 6 percent of GDP between 1980 and 1985. Tourist arrivals had dropped
sharply in 1983, however, the first decline in over twenty years. The
decrease resulted both from the regional and world economic downturn at
that time and from travel restrictions instituted by neighboring
countries to preserve their own foreign exchange. Observers noted also
that Singapore was losing its "oriental mystique and charm."
In its effort to build a modern city, it had torn down old buildings and
curtailed traditional street activities, aspects considered by tourists
to be part of Singapore's attraction. In 1984 the government established
a Tourism Task Force to recommend ways to attract more visitors, and the
following year the budget of the Singapore Tourist Promotion Board was
increased by 60 percent. Steps were taken to preserve areas of special
architectural, historical, or cultural interest. Sentosa Island, off the
southern coast, was developed as a resort and recreation center,
complete with museums, parks, golf courses, lagoons, beaches, trails,
and gardens, all connected by monorail. Singapore also began billing
itself as the "hub of Southeast Asia" and marketing sidetrips
to destinations in neighboring countries. As with other economic
activities, tourism was viewed as a high value-added industry. Although
increasing the absolute number of visitor arrivals was the main target,
a further aim was to attract the high-spending, business visitors
attending conventions and trade exhibitions, which Singapore hosted in
large numbers.
Tourist arrivals recovered quickly from the 1983 downturn, reaching 3
million in 1985. In 1987 tourist arrivals reached 3.7 million, a 15
percent increase over the previous year. In 1988 arrivals rose another
14 percent to nearly 4.2 million. Singapore's top tourist-generating
markets in 1987 were ASEAN (29 percent), Japan (15 percent), Australia
(9 percent), India (7 percent), the United States (6 percent), and
Britain (5 percent). Although a building boom had caused a glut of hotel
rooms in the mid-1980s, by early 1989 occupancy was running at about 80
percent.
Singapore
Singapore - FINANCE
Singapore
The country's rapid development was closely linked to the
government's efficient financial management. Conservative fiscal and
monetary policies generated high savings, which, along with high levels
of foreign investment, allowed growth without the accumulation of
external debt. In 1988 Singapore had foreign reserves worth about S$33
billion, which, per capita, put it ahead of Switzerland, Saudi Arabia,
and Taiwan. That same year, the domestic savings rate rose to one of the
highest in the world (42 percent), as gross national savings, comprising
public and private savings, totaled S$20.9 billion, 19 percent higher
than in 1987. By the mid-1980s, however, domestic demand had been so
stunted that it became increasingly difficult to find productive areas
for investment. In the recession year of 1986, for the first time, gross
national savings exceeded gross capital formation. This was in spite of
a 15 percent cut in the employers' contribution to the Central Provident
Fund. As a result, already depressed domestic demand was depressed even
further, falling by 1 percent in 1986 after a decline of 3 percent the
previous year.
Singapore's foreign reserves were, in fact, the country's domestic
savings held overseas. Since the source of the domestic savings was in
large measure the compulsory savings held by the Central Provident Fund,
Singapore had a huge domestic liability. The fund claims, standing in
1988 at S$32 billion, almost equalled Singapore's foreign reserves. But
since they were fully funded and denominated in Singapore dollars, the
country was relieved of the problems of showing either a budget deficit
or an external debt.
Indeed, for many years, the government had pointed out that its
foreign reserves, managed by the Government of Singapore Investment
Corporation, were larger than that of wealthier, more populous
countries. The reserves issue became politicized after 1987 when Lee
Kuan Yew proposed a change in the country's government to an executive
presidency in which the president (presumably Lee himself) would have
veto power over Parliament's use of the reserves. In 1986 the
government-sponsored Report of the Economic Committee admitted that
"over saving" was a problem. Not until 1988, however, were
some tentative steps taken to invest the surpluses directly in
productive resources. This process included a one-time transfer to
government revenue of S$1.5 billion from the accumulated reserves of
four statutory boards.
The country's public sector financial system was structurally complex
and difficult to follow owing to different accounting practices. Funds
essentially were derived from three sources: tax revenue (directly on
income, property, and inheritance; indirectly as excise duties, motor
vehicle taxes; stamp duties, and other taxes), nontax revenue
(regulatory charges, sales of goods and services, and interest and
dividends); and public sector borrowing. The statutory boards had
separate budgets, although they played a major role in infrastructure
creation. Government companies also were not included in public finance
reporting.
After 1975 the government consistently had substantial current as
well as overall surpluses. From 1983 to 1985, total government
expenditure averaged 59.8 percent of current revenue. In fact, the
overall surplus exceeded even the net contributions to the Central
Provident Fund. The seven major statutory boards also had consistent
current surpluses. Economic theoretician and member of Parliament
Augustine Tan suggested that Singapore's public spending and public
savings were much too large. According to Tan, the government tended to
err on the side of financial surplus, despite frequent forecasts of
deficit, because the government consistently underestimated tax revenues
and overestimated expenditures. These surpluses then put upward pressure
on the exchange rate and eroded manufacturers' competitiveness.
Currency, Trade, and Investment Regulation
Singapore had an exceptionally open economy. Fundamentally strong,
the currency reflected a sound balance of payments position, large
reserves, and the authorities' conservative attitude. From 1967 until
June 1973, the Singapore dollar was tied to the United States dollar,
and thereafter the currency was allowed to float.
The Monetary Authority of Singapore, the country's quasicentral bank,
pursued a policy of intervention both domestically and in foreign
exchange markets to maintain a strong currency. This multifaceted
strategy was designed to promote Singapore's development as a financial
center by attracting funds, while inducing low inflation by preventing
the erosion of the large Central Provident Fund balances. Furthermore,
the strong currency complemented the high wage industrial strategy,
forcing long-term quality rather than short-term prices to be the basis
for export competition.
Given Singapore's dependency on imports, however, setting an exchange
rate always generated controversy. The 1986 Report of the Economic
Committee did not clarify official thinking. It recommended that the
exchange rate should "continue to be set by market forces, but its
impact on [Singapore's] export competitiveness and tourist costs should
be taken into account. The [Singapore] dollar should, as far as
possible, be allowed to find its own appropriate level, reflecting
fundamental economic trends."
After 1978, when the government abolished all currency exchange
controls, Singaporean residents (individuals and corporations) were free
to move funds, import capital, or repatriate profits without
restriction. Likewise, trade regulations were minimal. Import duties
applied only to a few items (automobiles, alcohol, petroleum, and
tobacco), and licenses were required only for imports originating from a
few Eastern bloc countries. There were no export duties. As the
government played an active part in promoting exports, there was an
extensive system of supports including an export insurance plan.
The government promoted investment vigorously through a whole range
of tax and investment allowances and soft loans aimed at attracting new
investment or at helping existing businesses upgrade or expand. There
was no capital gains tax. Special incentives existed for foreigners,
including concessionary tax arrangements for some nonresidents, relief
from double taxation, and permission to buy commercial and certain
residential property. In 1985 extensive tax reductions were introduced
to reduce business costs.
Financial Center Development
As a result of its strategic location and well-developed
infrastructure, Singapore traditionally had been the trade and financial
services center for the region. In the 1970s, the government identified
financial services as a key source of growth and provided incentives for
its development. By the 1980s, the focus was on further diversification,
upgrading, and automation of financial services. Emphasis was placed on
the development of investment portfolio management, securities trading,
capital market activities, foreign exchange and futures trading, and
promotion of more sophisticated and specialized fee-based activities.
Consequently, by the mid-1980s, Singapore was the third most
important financial center in Asia after Tokyo and Hong Kong. The
financial services sector, having sustained double digit growth over the
previous decade, accounted for some 23 percent of GDP and employed
approximately 9 percent of the labor force. In 1985, however, growth in
the sector slowed to just 2.6 percent, and in December of that year the
Stock Exchange of Singapore suffered a major crisis, which forced it to
close for three days. In view of the troubled domestic economy,
observers worried that Singapore's future as a financial center looked
somewhat problematic. Furthermore, international financial market
deregulation threatened to create an environment in which it would be
more difficult for Singapore to thrive, especially given its high cost
structure and somewhat heavy-handed regulatory environment. The
government took steps to correct some of the problems, and by 1989
Singapore's financial service sector could again be described as
"booming."
The financial sector included three types of commercial banks (full
license, restricted, and offshore), representative offices, merchant
banks, discount houses, and finance companies. In 1988 there were 13
local, 64 merchant, and 134 commercial banks. All banks in Singapore
were administered by the Monetary Authority of Singapore and were
required to hold a statutory minimum cash balance against their deposit
and other specified liabilities with the authority.
The Development Bank of Singapore was established in 1968 to provide
financial services supporting industrialization and general economic
development. Owned jointly by the government (49 percent) and private
sector shareholders, it had evolved from a long-term financing
institution to a multiservice bank. The largest Singaporean commercial
bank in terms of assets in 1989, the Development Bank was listed on the
stock exchanges of both Singapore and Malaysia. Through its
subsidiaries, it also provided specialized financial and insurance
services, factoring, stockbroking, merchant banking, and venture capital
investment management services. The Development Bank was the
city-state's largest source of long-term finance, including equity and
venture capital financing, medium- and long-term loans, and guarantees.
The Singapore Foreign Exchange Market had grown remarkably since the
1985 recession. As an international financial center, the country had
benefited from the worldwide increase in business as well as from the
related expansion in the financially liberated Japanese market. Major
currencies--the United States dollar, the Japanese yen, the West German
deutsche mark, and the British pound sterling--were actively traded.
Volumes in such other currencies as the Australian dollar had risen as
well. Average daily turnover was US$45 billion in 1988 compared with
US$12.5 billion in 1985.
Singapore established the Asian dollar market as the Asian equivalent
of the Eurodollar market in 1968 when the local branch of the United
States-based Bank of America secured government approval to borrow
deposits of nonresidents, mainly in foreign currencies, and use them to
finance corporate activities in Asia. At the time, expanding economic
development in Southeast Asia was rapidly increasing the demand for
foreign investment funds, and the desirability of a regional center able
to carry out the necessary middleman function was apparent. Singapore
offered the ideal location. The Asian dollar market was essentially an
international money and capital market for foreign currencies, and its
assets grew from US$30 million in 1968 to US$273 billion in November
1988. To operate in the market, financial institutions were required to
obtain approval from the Monetary Authority of Singapore and to set up
separate bookkeeping entities called Asian currency units for
transactions in the market. Funds were obtained mainly from external or
nonresident sources--central banks, foreigners seeking a stable location
such as Singapore to deposit cash, multinational corporations, and
commercial banks outside Singapore.
In 1973, to stimulate the expansion of the Asian dollar market, the
Monetary Authority of Singapore established the so-called offshore
banking system, designed to concentrate on that market and its foreign
exchange operations. Beginning in 1983, funds managed in Singapore on
behalf of nonresidents and invested offshore or in the local stock
market were exempt from tax. The fees earned for managing such offshore
funds were taxed at a concessionary rate of 10 percent.
Inaugurated in 1973, the Stock Exchange of Singapore was governed by
a committee comprising four elected stockbroker members and five
appointed nonbroker members. In late 1988, the 327 companies listed on
the main board of the exchange were classified into six groups:
industrial and commercial, finance, hotel, property, plantation
(farming), and mining. The market underwent a major, prolonged
reorganization following the December 1985 collapse of a Singaporean
company, Pan Electric, which revealed a massive web of forward share
dealings based on borrowed money. The collapse resulted in a tighter
regulation of the financial futures market and the securities industry.
In 1986 the Securities Industry Council was established to advise the
minister for finance on all matters relating to the securities industry.
In 1987 the government introduced tax incentives to encourage the
trading of international securities in Singapore. The National
Association of Securities Dealers (NASDAQ) in the United States and the
Stock Exchange of Singapore established a link to facilitate the trading
of NASDAQ stocks in Singapore by providing for the exchange of price and
trading information on a selected list of NASDAQ stocks between the two
exchanges. A move by the Singapore exchange to a new, spacious location
in 1988 brought a transformation in trading methodology, including
partial automation of the trading system, which until then had adhered
to the traditional outcry auction system.
By 1987 Singapore's stock market, fuelled by bullish sentiments sent
indices soaring to new highs--a recovery from the December 1985 crisis.
All gains, however, were wiped out by the crash of world stock markets
in October 1987, a crash from which the Singapore exchange had made
substantial recovery by mid-1989.
Singapore also expanded other international financial markets in the
late 1980s. Trading in gold futures originally was undertaken in the
Gold Exchange of Singapore, which was established in 1978 and
reorganized in 1983. The scope of its activities was widened to include
financial futures trading, and it was renamed the Singapore
International Monetary Exchange (SIMEX). Starting in 1984, the financial
futures market featured a mutual offset arrangement between SIMEX and
the Chicago Mercantile Exchange, which allowed contracts executed on one
exchange to be offset on the other without additional transactional cost
for market participants. The linkage was the first of its kind in the
world and greatly facilitated round-the-clock trading in futures
contracts. In 1988 six forms of futures contracts were traded:
international gold futures; the Eurodollar time deposit interest rate;
the Nikkei Average Stock Index; and three currency exchange rates--US
dollar/West German deutsche mark, US dollar/Japanese yen, and US
dollar/British pound sterling. Trading volume on the SIMEX had grown
steadily.
The restructured Government Securities Market was launched in May
1987, auctioning at market rates taxable Singapore government securities
ranging in maturity from three months to five years. Previously,
long-term government stock was sold to a captive market of banks,
insurance companies, and a few individuals and nonprofit organizations.
International Financial Organizations
In 1966 Singapore became a member of the International Monetary Fund
( IMF), the World Bank, and the Asian Development Bank. Two years later,
Singapore joined the International Finance Corporation, an affiliate of
the World Bank. Singapore's loans from the World Bank and the Asian
Development Bank had been used to finance development projects relating
to water supply, electric power generation and distribution, sewerage,
telephone services, educational services, and environmental control. A
total of fourteen loans were secured from the World Bank between 1963
and 1975 and fourteen from the Asian Development Bank between 1969 and
1980. There were no further loans in the 1980s. Singapore's estimated
outstanding borrowings from the World Bank and the Asian Development
Bank in late 1988 totalled US$35.1 billion and US$45.4 million,
respectively. Its 1988 quota of IMF special drawing rights
(SDR)--related to its national income, monetary reserves, trade balance
and other economic indicators--was SDR 92.4 million.
Singapore
Singapore - AGRICULTURE
Singapore
Orchard Road, now one of Singapore's most up-scale thoroughfares, got
its name because it originally was lined with fruit orchards and
vegetable gardens. Although contemporary Singapore still maintained a
tiny agricultural base, by 1988 urbanization had reduced the land area
used for farming to only about 3 percent of the total. Nonetheless, with
intensive production, the farming sector met part of the domestic demand
for essential fresh farm produce: poultry, eggs, pork, some vegetables,
and fish. In 1988 there were 2,075 licensed farms occupying only 2,037
hectares of land, with a total output of some S$362 million worth of
farm produce. A decade earlier farm holdings had covered 1,280 hectares.
The Primary Production Department, under the Ministry of National
Development, ensured an adequate and regular supply of fresh produce and
provided support for agro-industries, including research and development
aimed at improving commercial and hightechnology farming. The department
projected in 1988 that a total of 2,000 hectares of land in ten
agro-technology parks would be developed and rented out for long-term
farming over the next decade.
The government began phasing out pig farming in 1984 because of odor
and environmental pollution. Some 200 pig farms raising about 500,000
pigs in 1987 were scheduled to be reduced to 22 farms with 300,000 pigs
by 1990. Imports from Malaysia, Indonesia, and Thailand would be
increased to meet domestic needs. Some 1,000 poultry farms kept a total
of about 2.2 million layers, 1.6 million broilers, 245,000 breeders, and
645,000 ducks. Singapore remained free of major animal diseases.
Singapore grew 5.6 percent of its total supply of 180,000 tons of
fresh vegetables in 1988 and imported the rest from Malaysia, Indonesia,
China, and Australia. The main crops cultivated locally included
vegetables, mushrooms, fruit, orchids, and ornamental plants. About 370
vegetable farms produced an estimated 10,000 tons of vegetables, and
mushroom cultivation expanded rapidly after the mid-1980s. The Mushroom
Unit of the Primary Production Development conducted research on
mushroom cultivation and advised commercial mushroom growers, who
produced a variety of mushrooms for the local market.
Noted for its orchids, Singapore exported flowers worth S$13.8
million in 1988, mainly to Western Europe, Japan, Australia, and the
United States. Singapore's 153 orchid farms produced another S$2.2
million worth of flowers for the domestic market.
Local fishermen provided about 13 percent of the country's
110,000-ton fresh fish supply in 1988, using three major fishing
methods--trawling, gill-netting, and long-lining. There were about 1,170
licensed fishermen operating nearly 400 fishing vessels, most of which
were motorized. The Jurong Port and Market Complex was a major fish
landing point for both domestic and foreign vessels and handled 84
percent of the total fresh fish supply in 1988. Many foreign vessels
brought their catches there for processing and reexport. Fresh fish
arrived also by truck from Malaysia and Thailand and by sea and air from
other neighboring countries.
Fish farming was a small but growing field. In 1988 seventyfour
licensed marine fish farms raised mainly high-value fish such as grouper
and sea bass in a total of forty hectares of coastal waters. Many of the
farms had also introduced prawn farming in floating cages. Exports of
ornamental fish for aquariums amounted to S$60 million in 1988. Some 400
licensed aquarium fish farms operated in Singapore in 1988, including 36
commercial farms operating in the Tampines Aquarium Fish Farming Estate.
Singapore
Singapore - Government
Singapore
AFTER TWENTY-FIVE YEARS OF INDEPENDENCE, Singapore enjoyed a
reputation for political stability and honest, effective government.
Probably the world's only ex-colony to have independence forced upon it,
Singapore responded to its unanticipated expulsion from Malaysia in
August 1965 by concentrating on economic development and by fostering a
sense of nationhood. Though the survival of the miniature state was in
doubt for a time, it not only survived but also managed to achieve the
highest standard of living in Southeast Asia. The country also enjoyed a
rare political continuity; its ruling party and prime minister triumphed
in every election from 1959 to 1988. Singapore's government had an
international reputation for effective administration and for ingenious
and successful economic policies. It was also known for its
authoritarian style of governance and limited tolerance for opposition
or criticism, qualities the government deemed necessary to ensure
survival in a hostile world and which its domestic and foreign critics
claimed indicated a refusal to consider the opinions of its citizens or
anyone outside the closed circle of the aging leadership. In the early
1990s, the leadership would face the issues of political succession and
of modifying the relationship between the state and the increasingly
prosperous and well-educated society it had created.
Singapore
Singapore - Form of Government
Singapore
The Republic of Singapore is a city-state with a governing structure
patterned on the British system of parliamentary government. In 1989
legislative power was vested in a unicameral Parliament with eighty-one
members who were elected for five-year terms (or less if the Parliament
was dissolved prematurely). Members of Parliament were elected by
universal adult suffrage from forty-two single-member constituencies and
thirteen group representation constituencies. Voting was compulsory for
all citizens above the age of twenty-one. The group representation
constituencies elected a team of three members, at least one of the whom
had to be Malay, Indian, or a member of one of Singapore's other
minorities. The group representation constituencies, introduced in the
1988 general election, were intended to ensure multiracial parliamentary
representation to reflect Singapore's multiracial society. In another
departure from the British model, members of Parliament elected on a
party ticket had to resign if they changed parties. A 1984 amendment to
the Parliamentary Elections Act provided for the appointment to
Parliament of up to three nonconstituency members if the opposition
parties failed to win at least three seats in the general election. The
nonconstituency members were chosen from the opposition candidates who
had polled the highest percentage of votes. The seventh Parliament,
elected on September 3, 1988, and meeting for the first time on January
9, 1989, included one elected opposition member and one nonconstituency
member.
Singapore had only one level of government--national government and
local government were one and the same. The form of the government
reflected the country's unusually small area and modest total population
of 2.6 million. Below the national level, the only recognized
territorial divisions were the fifty-five parliamentary constituencies.
Members of Parliament thus performed some of the same functions as
municipal aldermen in foreign cities and often won political support by
helping to find jobs for constituents or doing other favors requiring
intercession with the powerful civil bureaucracy. The single-member
constituencies varied in population from 11,000 electors to as many as
55,000; some of the variability reflected population movement away from
the old urban core and out to new housing developments.
As in all British-style polities, the government was headed by a
prime minister who led a cabinet of ministers of state selected from the
ranks of the members of Parliament. The cabinet was the policy-making
body, and its members directed the work of the permanent civil servants
in the ministries they headed. In 1989, the cabinet comprised fifteen
members. Below the prime minister were a first deputy prime minister and
a second deputy prime minister. They were followed by the ministers in
charge of such functional departments as the Ministry of Finance or the
Ministry of Defence and by two ministers without portfolio. The prime
minister could reassign his cabinet members to new portfolios or drop
them from the cabinet, and successful ministers headed several
progressively more significant ministries in their careers. There were
thirteen ministerial portfolios in 1989: defence, law, foreign affairs,
national development, education, environment, communications and
information, home affairs, finance, labour, community development, trade
and industry, and health. Some portfolios were split between different
ministers. The first deputy prime minister (Goh Chok Tong) was also
first minister for defence. The minister for communications and
information (Yeo Ning Hong) also served as second minister for defence
(policy). The minister for trade and industry (Brigadier General
(Reserve) Lee Hsien Loong) was concurrently second minister for defence
(services). The foreign affairs and law portfolios were similarly
divided.
The cabinet met once or twice a week; its meetings were private and
confidential. Administrative and staff support to the prime minister and
cabinet was provided by the Office of the Prime Minister, the officials
of which included a senior minister of state, a political secretary, a
secretary to the prime minister, and a secretary to the cabinet. The
Office of the Prime Minister coordinated and monitored the activities of
all ministries and government bodies and also directly supervised the
Corrupt Practices Investigation Bureau and the Elections Department.
Each minister was assisted by two secretaries, one for parliamentary or
political affairs and the other for administrative affairs. The latter,
the permanent secretary, was the highest ranking career civil servant of
the ministry.
The constitutional head of state was the president, who occupied a
largely powerless and ceremonial role. The president was elected by the
Parliament for a four-year term. He could be reelected without limit and
removed from office by a two-thirds vote of Parliament. In turn, the
president formally appointed as prime minister the member of Parliament
who had the support of the majority of Parliament. On the advice of the
prime minister, the president then appointed the rest of the ministers
from the ranks of the members of Parliament. The president, acting on
the advice of the prime minister, also appointed a wide range of
government officials, including judges, and members of advisory boards
and councils.
In 1988 the government discussed amending the Constitution to
increase the power of the president. A white paper introduced in
Parliament in July 1988 recommended that the president be directly
elected by the people for a six-year term and have veto power over
government spending as well as over key appointments. It also proposed
an elected vice president with a six-year term of office. The proposed
changes originated as a device intended to permit Prime Minister Lee
Kuan Yew, who had been prime minister since 1959, to retain some power
should he retire, as he had hinted, and assume the presidency. No
specific dates for the proposed constitutional change were given in the
white paper. As of late 1989, no action had been taken.
Singapore
Singapore - Constitutional Framework
Singapore
Singapore became an autonomous state within Malaysia, with its own
constitution, on September 16, 1963. It separated from Malaysia on
August 9, 1965. On December 22, 1965, the Legislative Assembly passed a
Singapore Independence Bill and a Constitutional Amendment. The
Constitutional Amendment provided for a parliamentary system of
government, with a president, whose duties were largely ceremonial,
elected every four years by the Parliament.
The Constitution can be amended by a two-thirds vote of Parliament. A
1966 amendment allowed appeal from the Court of Appeal in Singapore to
the Judicial Committee of Her Majesty's Privy Council in Britain. In
1968 an amendment created the office of vice president and liberalized
the requirements of citizenship. A 1969 amendment established the
Supreme Court in place of the High Court and Court of Appeal as the
highest appeal tribunal. A 1972 amendment entitled "Protection of
the Sovereignty of the Republic of Singapore," introduced a measure
to ensure the sovereignty of the city-state. It prohibited any merger or
incorporation with another sovereign state, unless approved in a
national referendum by a two-thirds majority. Under the same terms, it
also prohibited the relinquishment of control over Singapore police
forces and armed forces. In 1978 the Fundamental Liberties section of
the Constitution (Part IV, Articles 9-16) was amended; the amendment
extended government powers by establishing that arrests to preserve
public safety and good order and laws on drug abuse would not be
inconsistent with liberties set forth in that section of the
Constitution.
Singapore
Singapore - Major Governmental Bodies
Singapore
The President
The Constitution states that the president shall be elected by
Parliament for a term of four years. In consultation with the prime
minister, the president appoints to his personal staff any public
officers from a list provided by the Public Service Commission. In the
exercise of his duties, the president acts in accordance with the advice
of the cabinet or of a minister acting under the authority of the
cabinet. The president may use his discretion in the appointment of the
prime minister and in withholding consent to a request for the
dissolution of Parliament.
The Executive
The Constitution stipulates that the executive authority of Singapore
is vested in the president and exercised by him or the cabinet or any
minister authorized by the cabinet, subject to the provisions of the
Constitution. The cabinet directs and controls the government and is
responsible to Parliament. The president appoints a member of Parliament
as prime minister and, in accordance with the advice of the prime
minister, appoints an attorney general. The attorney general advises the
government on legal matters and has the discretionary power to initiate,
conduct, or terminate any proceedings for any offense.
The Legislature
The legislature consists of the president and Parliament. Members
must be citizens of Singapore, twenty-one years of age or older, on the
current register of electors, able to communicate in either English,
Malay, Mandarin Chinese, or Tamil, and of sound mind. Membership ceases
with the dissolution of a Parliament, which takes place every five years
or at the initiative of the president. A general election must be held
within three months of the dissolution of Parliament. Parliament
convenes at least once a year, scheduling its meetings after the first
session is summoned by the president. Members may speak in English,
Malay, Mandarin Chinese or Tamil, and simultaneous translation is
provided. Parliamentary procedure follows the British pattern: all bills
are deliberated in three readings and passed by a simple majority. Only
the government may introduce money bills, those that allocate public
funds and so provide for the ongoing operations of the state. Once
passed, bills become laws with the assent of the president and
publication in the official Gazette.
The final step in the passage of laws is the examination of bills by
the Presidential Council for Minority Rights. The council, established
by the Constitution (Amendment) Act of 1969, must determine if bills or
other proposed legislation discriminate against any religious or ethnic
community or otherwise contravene the fundamental liberties guaranteed
by the Constitution. It also renders advisory opinions on matters
affecting ethnic and religious communities that are referred to the
council by the Parliament or government. The council is composed of ten
members appointed for life and ten members and a chairman appointed for
three-year terms by the president on the advice of the cabinet. Any bill
on which the council renders an adverse opinion may not become law
unless modified to its satisfaction or passed by two-thirds of the
Parliament. The council has no jurisdiction over money bills or over any
bill certified by the prime minister as affecting the defense or
security of Singapore or the country's "public safety, peace, or
good order." In addition, bills certified by the prime minister as
so urgent that it is not in the public interest to delay their enactment
are also exempted from review by the council.
Elections
The electoral system is based on single-member constituencies. The
law (amendments to the Constitution and to the Parliamentary Elections
Act) providing for group representation constituencies also stipulated
that the total number of members of Parliament from group representation
constituencies had to total less than half the total number of members.
Slightly more than half the constituencies would remain single-member
constituencies. The candidate receiving the largest number of votes wins
the election in that constituency. The consequence of this electoral
rule, common to most Britishstyle constitutions, is to eliminate
parliamentary representation for minority parties and to encourage the
organization of parties whose candidates can win pluralities in many
constituencies. In theory it is possible for a party to win every seat
in parliament by receiving a plurality in every constituency.
The Judiciary
Singapore's judicial power is vested in the Supreme Court, consisting
of a chief justice and an unspecified number of other judges. All are
appointed by the president, acting on the advice of the prime minister.
The judiciary functions as the chief guardian of the Constitution
through its judicial review of the constitutionality of laws. The
Supreme Court of Judicature Act of 1969, and various subsequent acts
ensured judicial independence and integrity by providing for the
inviolability of judges in the exercise of their duties and for
safeguards on their tenure.
The Constitution establishes two levels of courts--the Supreme Court
and the subordinate courts. The subordinate courts are the magistrates'
courts, trying civil and criminal offenses with maximum penalties of
three years' imprisonment or a fine of S$10,000; the district courts,
trying cases with maximum penalties of ten years' imprisonment or a fine
of S$50,000; the juvenile courts, for offenders below the age of
sixteen; the coroners' courts; and the small claims courts, which hear
civil and commercial claims for sums of less than S$2,000. The Supreme
Court consisted of the High Court, which has unlimited original
jurisdiction in all civil and criminal cases and which tries all cases
involving capital punishment; the Court of Appeal, which hears appeals
from any judgment of the High Court in civil matters; and the Court of
Criminal Appeal, which hears appeals from decisions of the High Court in
criminal cases. The final appellate court is the Judicial Committee of
Her Majesty's Privy Council in London. According to Article 100 of the
Constitution, the president may make arrangements for appeals from the
Supreme Court to be heard by the Judicial Committee of the Privy
Council. In May 1989, Parliament abolished the right to appeal to the
Privy Council except for criminal cases involving the death sentence and
civil cases in which the parties had agreed in writing to such an appeal
at the outset. The judicial system reflected British legal practice and
traditions, except for trial by jury. Singapore abolished jury trials
except for capital offenses in 1959; all jury trials were abolished by
the 1969 amendment of the code of criminal procedure.
The chief justice and other judges of the Supreme Court are appointed
by the president on the advice of the prime minister. The prime
minister, however, is required to consult the chief justice on his
recommendations for the Supreme Court. Judges of the subordinate courts
are appointed by the president on the advice of the chief justice.
Singapore's judges and superior courts repeatedly demonstrated their
independence from the government by ruling against the government in
cases involving political opponents or civil liberties. The government
response in such cases was to amend the law or to pass new laws, but it
did not attempt to remove or to intimidate judges. Although internal
political struggle in Singapore from the 1950s through the 1980s was
often intense, and the ruling government was quite willing to intimidate
and imprison its political opponents, it always followed legal forms and
procedures.
The attorney general is appointed by the president, on the advice of
the prime minister, from persons qualified to become judges of the
Supreme Court. A judge may be removed from office only for misbehavior
or incapacitation, which must be certified by an independent tribunal.
The attorney general, who is assisted by the solicitor general, is the
principal legal advisor to the government, serves as the public
prosecutor, and is responsible for drafting all legislation. The office
of the attorney general, the Attorney General's Chambers, is divided
into the legislation, civil, and criminal divisions.
Singapore
Singapore - The Public Service
Singapore
The public services included the Singapore Armed Forces, the
Singapore Civil Service, the Singapore Legal Service, and the Singapore
Police Force. A Public Service Commission (PSC), consisting of a
chairman and no less than five nor more than nine other members, was
appointed by the president, with the advice of the prime minister. The
PSC acted to appoint, confirm, promote, transfer, dismiss, pension, and
impose disciplinary control over public officers. A Public Service
Division, established within the Ministry of Finance in 1983, managed
civil service personnel. It was headed by a permanent secretary who was
responsible to the minister for finance.
A Legal Service Commission, with jurisdiction over all officers in
the Singapore Legal Service, was composed of the chief justice as
president, the attorney general, the chairman of the Public Service
Commission, a judge of the Supreme Court nominated by the chief justice,
and not more than two members of the Public Service Commission nominated
by that commission's chairman. The Legal Service Commission acted to
appoint, confirm, promote, transfer, dismiss, pension, and exercise
disciplinary control over officers in the Singapore Legal Service.
The investigation of corruption in both the public and private
sectors was under the sole authority of the Corrupt Practices
Investigation Bureau, part of the prime minister's office. The Auditor
General's Office, an independent agency functioning without interference
from any ministry or department, monitored Parliament to ensure its
compliance with laws and regulations and to identify irregularities in
its disbursement of government resources.
Singapore
Singapore - The Public Bureaucracy
Singapore
The government played an active role in managing the society and
developing the economy and was the country's largest single employer.
Government bodies and their employees fell into two distinct categories.
The regular ministries and their civil service employees concentrated on
recurrent and routine administrative tasks. The three ministries of
education, health, and home affairs (including police, fire, and
immigration) employed 62 percent (43,000) of the 69,700 civil servants
in 1988. Members of the civil service in the strict sense of the term
were those public employees who were appointed by the Public Service
Commission and managed by the Ministry of Finance's Public Service
Division. Active projects in economic development and social engineering
were carried out by a large number of special-purpose statutory boards
and public enterprises, which were free from bureaucratic procedures and
to which Parliament delegated sweeping powers. As of 1984, there were
eighty-three statutory boards employing 56,000 persons. About 125,000
members of the 1987 total work force were public employees.
The two branches of the public service served different functions in
the political system. The civil service proper represented institutional
continuity and performed such fundamental tasks as the collection of
revenue, the delivery of such goods as potable water, and the provision
of medical and educational services. The various quasigovernmental
bodies, such as statutory boards, public enterprises, commissions, and
councils represented adaptability, innovation, and responsiveness to
local conditions. The constitutional framework of Singapore's
government, with its Parliament, cabinet, courts, and functional
ministries, resembled that of its British model and its peers in other
countries of the British Commonwealth
of Nations. The particular collection of boards and
councils, which included everything from the Central Provident Fund to
the Sikh Advisory Board, reflected the successful adaptation of the
British model to its Southeast Asian environment.
Public service employment carried high prestige, and there was
considerable competition for positions with the civil service or the
statutory boards. Civil servants were appointed without regard to race
or religion, and selected primarily on their performance on competitive
written examinations. The civil service had four hierarchical divisions
and some highly ranked "supergrade" officials. On January 1,
1988, there were 493 supergrade officers, who included ministerial
permanent secretaries and departmental secretaries and constituted less
than 1 percent of the 69,700 civil servants. Division one consisted of
senior administrative and professional posts and contained 14 percent of
the civil servants. The mid-level divisions two and three contained
educated and specialized workers who performed most routine government
work and who made up the largest group of civil servants, 33 and 32
percent of all civil servants, respectively. Division four consisted of
manual and semiskilled workers who made up 20 percent of employees. In
1987, there were 3,153 appointments from the 9,249 applicants for
positions in divisions one through three; 2,200 (some 70 percent) of the
appointees were women.
The Singapore public service was regarded as almost entirely free
from corruption, a fact that in large part reflected the strong emphasis
the national leadership placed on probity and dedication to national
values. The Corrupt Practices Investigation Bureau enjoyed sweeping
powers of investigation and the unreserved support of the prime
minister. Official honesty was also promoted by the relatively high
salaries paid to public officials; the high salaries were justified by
the need to remove temptations for corruption. In a system with clear
echoes of the Chinese Confusian tradition, and the British
administrative civil service, which recruited the top graduates of the
elite universities, Singapore's public service attempted, generally
successfully, to recruit the most academically talented youth. The
Public Service Commission awarded scholarships to promising young people
for study both in Singapore and at foreign universities on the condition
that the recipients join the civil service after graduation. Young
recruits to the development-oriented statutory boards were often given
substantial responsibilities for ambitious projects in industrial
development or the construction of housing estates. Officials had
greater social prestige than their peers in business; power and official
title outranked money in the local scale of esteem.
Singapore
Singapore - Statutory Boards
Singapore
The eighty-three statutory boards were a distinctive feature of
Singapore's government. In law, a statutory board was an autonomous
government agency established by an act of Parliament that specified the
purpose, rights, and powers of the body. It was separate from the formal
government structure, not staffed by civil servants, and it did not
enjoy the legal privileges and immunities of government departments. It
had much greater autonomy and flexibility in its operations than regular
government departments. Its activities were overseen by a cabinet
minister who represented Parliament to the board and the board to
Parliament. Statutory boards were managed by a board of directors, whose
members typically included senior civil servants, businessmen,
professionals, and trade union officials. The chairman of the board of
directors, who was often a member of Parliament, a senior civil servant,
or a person distinguished in some relevant field, was appointed by the
cabinet minister who had jurisdiction over the board. The employees of
the board were not civil servants, as they were not appointed by the
Public Service Commission. The salary scales and terms of service of
employees differed from board to board. Statutory boards did not receive
regular allocations of funds from the public treasury, but were usually
expected to generate their own funds from their activities. Surplus
funds were invested or used as development capital, and boards could
borrow funds from the government or such bodies as the World
Bank. Statutory boards included the Housing and
Development Board, the Central Provident Fund, the Port of Singapore
Authority, the Industrial Training Board, the Family Planning and
Population Board, and the Singapore Muslim Religious Council (Majlis
Ugama Islam Singapura).
The statutory boards played the major role in the government's
postindependence development strategy, and their activities usually
served multiple economic and political goals. The Housing and
Development Board (HDB) provided a good example. The HDB was established
by the first People's Action Party (PAP) government on February 1, 1960,
to provide low-cost public housing. The Lands Acquisition Act of 1966
granted the board the power of compulsory purchase of any private land
required for housing development. The prices paid by the board were
about 20 percent of the estimated market value of the land, which was in
fact if not in form being nationalized. Between 1960 and 1979, the
percentage of land owned by the government rose from 44 to 67 percent,
increasing the government's control over that scarce resource and
benefiting lowincome voters, who supported the PAP, at the expense of
the much smaller number of private landowners. Rents for Housing and
Development Board apartments were subsidized, and selling prices for the
apartments were set below construction costs and did not include land
acquisition costs. Purchase prices for board apartments in the 1980s
were 50 to 70 percent below those of privately owned apartments. By 1988
Housing and Development Board apartment complexes were home to 86
percent of the population, and construction of new apartments continued.
The HDB succeeded in its primary goal of building large numbers of
high-quality apartments. Its success depended on several factors, among
them: access to large amounts of government capital; sweeping powers of
land acquisition; the ability to train its own construction workers and
engineers; the freedom to act as a building corporation and develop its
own quarries and brick factory; the opportunity to enter into
partnerships and contracts with suppliers of construction materials; and
the ability to prevent corruption in contracting and allocation of
apartments to the public. The government raised the capital for housing
construction from the Central Provident Fund, a compulsory savings plan
into which all Singapore workers contributed up to 25 percent of their
monthly incomes, and from low-interest, long-term loans from such
international development agencies as the World Bank.
By providing adequate housing at low cost to low-paid workers in the
1960s, the PAP delivered a highly visible and concrete political reward
to the electorate and laid the foundations for its unbroken electoral
success. In the 1960s and early 1970s, before the growth of
export-oriented industry, housing construction provided much employment
and an opportunity for workers to learn new skills. By controlling the
pace and scale of housing construction, the government was able to
better regulate the economy and smooth out cycles of economic activity.
The result of rehousing practically the entire population was to make
the government either the landlord or the mortgage holder for most
families and so bring them into closer contact with the state. The
government used resettlement to break up the ethnic enclaves and
communities that had characterized colonial Singapore. It put its policy
of multiracialism into practice by seeing that all apartment buildings
contained members of all ethnic groups in numbers that reflected their
proportion of the national population. The program kept the
cost of housing in Singapore relatively low and helped to avert pressure
to raise wages. Because access to subsidized housing was a benefit
extended only to citizens, it served to promote identification with the
new state. Providing most of the population with low-cost housing gave
the government and ruling party much favorable publicity, won public
support, and was used as evidence for the correctness of the
government's policies of centralized planning and social engineering
implemented by experts on behalf of a passive public.
In a similar fashion, the Central Provident Fund benefited the
citizens by providing them with secure savings for their old age and the
satisfaction of having their own account, which could be used as
security for the purchase of a Housing and Development Board apartment,
for such expenses as medical bills, for college tuition, or to finance a
pilgrimage to Mecca. The government benefited by gaining control of a
very large pool of capital that it could invest or spend as it would and
by removing enough purchasing power to limit inflationary tendencies.
Furthermore, the proportion of the wage contributed to the fund by both
workers and their employers could be adjusted at any time, enhancing the
government's ability to control the economy. In 1988 the Fund took 35
percent of all wages up to S$6,000 per month; 25 percent was paid by the
worker and 10 percent by the employer. Among its other functions, the
Central Provident Fund was one of the major instruments used by the
government to control wages.
Singapore
Singapore - Public Enterprises
Singapore
Apart from the statutory boards, which met general development and
infrastructure goals, the government owned or held equity in many
businesses that operated in the private sector. The government asserted
that such businesses received no special subsidies and would be
liquidated if they proved unprofitable. The wholly government-owned
Temasek Holdings (Private) Limited was the country's largest
corporation. Operated as an investment and holding corporation, its
offices were in the Ministry of Finance, which provided the corporation
with free accounting and secretarial services. Some government
enterprises included former government departments, such as the
Government Printing Office, which in 1973 became the Singapore National
Printers Limited and offered its services to the private sector at
market rates. Most government enterprises either provided key and
potentially monopolistic services, such as Singapore International
Airline or Neptune Orient Line, an ocean shipping firm, or they met
strategic and defense needs. The Ministry of Defence wholly owned or had
large equity shares in a range of companies engaged in weapons
production, electronics, computer software, and even food production. In
some cases the government banks, holding companies, or corporations were
partners or had shares in local operations of multinational
corporations. In such cases, the goal was both to attract the
corporations to Singapore by offering investment funds and the promise
of cooperation from government departments and to ensure that the
corporations transferred proprietary technology and training to
Singapore. The strategic nature of much government enterprise was
acknowledged by the January 20, 1984, passage of the Statutory Bodies
and Government Companies (Protection of Secrecy) Act. The law barred the
unauthorized disclosure of confidential information by anyone associated
with a statutory board or government enterprise and was considered
necessary because the Official Secrets Act did not cover those bodies.
Singapore
Singapore - Parapolitical Institutions
Singapore
After independence, Singapore's rulers perceived the population as
uncommitted to the new state and as lacking a common identity.
Accordingly, the government devoted much effort to fostering popular
identification with the nation and commitment to the government's goals.
In 1985 the Ministry of Community Development was formed by combining
the former Ministry of Social Affairs with activities previously
administered by the Office of the Prime Minister and by the Ministry of
Culture. The new ministry coordinated a network of grassroots agencies
intended to promote community spirit and social cohesion. These were the
People's Association, the Citizens' Consultative Committees, the
Residents' Committees, and the Community Center Management Committees.
The People's Association was a statutory board established in 1960 and
until 1985 a part of the Office of the Prime Minister. Its primary
activity was to manage a system of 128 community centers, which offered
recreational and cultural programs, along with such services as
kindergartens and a limited number of day-care centers for children of
working parents. The members of the various consultative and management
committees were volunteers who received prestige but no salary. Each
parliamentary constituency had a Citizens' Consultative Committee, whose
members were in frequent contact with their member of Parliament. All
Housing and Development Board apartment complexes had Residents'
Committees, headed by volunteers and intended to promote neighborliness
and community cohesion. The committees' activities included organization
of neighborhood watch programs and tree-planting campaigns, in which the
committees were assisted by the civil servants of the Residents'
Committees Group Secretariat. In 1986 the government began organizing
Town Councils in the larger housing estates. Though not official
government bodies, the councils' immediate purpose was to take over some
responsibilities for management of the complexes from the HDB. Their
larger purpose was to promote a greater sense of community and public
involvement in the residents of the clusters of high-rise apartment
buildings. In March 1985, the government inaugurated a Feedback Unit, a
body intended to collect public opinion on proposed government policies
and to encourage government departments to respond quickly to public
suggestions or complaints.
The various advisory committees and the Feedback Unit provided
functions that in many countries are provided by political parties. In
Singapore the parapolitical institutions, which had the clearly
political goal of generating public support for government policies,
were presented as apolitical, inclusive, communityoriented bodies,
headed by people motivated by a selfless desire for public service. Such
an approach reflected a decision made by the country's rulers in the
1960s to avoid trying to organize a mass political party, in part
because many less-educated citizens tended to shy away from partisan and
overtly political groups. Others habitually avoided government offices
and officers but would participate in community-oriented and attractive
programs. The ruling elite had had serious problems both with opposition
parties and with left-wing opposition factions within the PAP and
apparently found the controlled mobilization offered by the
parapolitical institutions more to its liking. Members of all the
advisory and consultative boards were appointed by the government and
were carefully checked by the security services before appointment. The
government closely watched the performance of the leaders of the
community organizations and considered the organizations a pool of
talent from which promising individuals could be identified, promoted to
more responsible positions, and perhaps recruited to the political
leadership.
Singapore
Singapore - Political Parties
Singapore
In 1989 the government of Singapore had been led since 1959 by one
political party, the PAP, and one man, Prime Minister Lee Kuan Yew. In
the 1988 parliamentary elections, opposition candidates challenged the
ruling party in an unprecedented seventy contests, but the PAP still won
eighty of the eighty-one seats in Parliament with 61.8 percent of the
popular vote, 1 percent less than in 1984, and 14 percent less than in
1980.
The PAP was founded in 1954, and in the 1950s acted as a leftwing
party of trade unionists, whose leadership consisted of English-educated
lawyers and journalists and Chinese-educated and pro-communist trade
union leaders and educators. It won control of the government in the
crucial 1959 election to the Legislative Assembly, which was the first
election with a mass electorate and for an administration that had
internal self-government (defense and foreign relations remained under
British control). The PAP mobilized mass support, ran candidates in all
fifty-one constituencies, and won control of the government with
forty-three of the fifty-one seats and 53 percent of the popular vote. After a bitter internal struggle the
English-educated, more pragmatic wing of the party triumphed over the
pro-communists in 1961 and went on to an unbroken string of electoral
victories, winning all the seats in Parliament in the 1968, 1972, 1976,
and 1980 general elections.
With a single party and set of leaders ruling the country for thirty
years, Singapore had what political scientists called a dominant party
system or a hegemonic party system, similar to that of Japan or Mexico.
There were regular elections and opposition parties and independent
candidates contested the elections, but after the early 1960s the
opposition had little chance of replacing the PAP, which regularly won
60 to 70 percent of the popular vote. The strongest opposition came from
the left, with union-based parties appealing to unskilled and factory
workers. In the early 1960s, the union movement split between the
leftist Singapore Association of Trade Unions and the National Trades
Union Congress (NTUC), which was associated with Lee Kuan Yew's
pragmatic wing of the PAP. In 1963 the Singapore Association of Trade
Unions was banned and its leaders arrested as pro-communist subversives.
The NTUC was controlled by the PAP and followed a government-sponsored
program of "modern unionism," under which strikes were unknown
and wages were, in practice, set by the government through the National
Wages Council.
The dominance of the PAP rested on popular support won by economic
growth and improved standards of living combined with unhesitating
repression of opposition leaders, who were regularly arrested on charges
of being communist agents or sympathizers. In the mid-1980s, eighteen
other political parties were registered, although many of them were
defunct, existed only on paper, or were the vehicles of single leaders.
Much of the electoral support for opposition parties represented protest
votes. Those voting for opposition candidates did not necessarily expect
them to win or even wish to replace the PAP government. They used their
votes to express displeasure with some or all PAP policies.
At the top of the PAP organization was the Central Executive
Committee (CEC). In 1954 the PAP constitution provided for a CEC of
twelve persons directly elected by party members at the annual general
meeting. The CEC then elected its own chairman, vice chairman,
secretary, assistant secretary, treasurer, and assistant treasurer. This
practice continued until August 1957, when six procommunist members of
the party succeeded in being elected. In 1958 the party revised its
constitution to avoid a recurrence. The document called for CEC members
to be elected at biennial party conferences by party cadre members, who
in turn were chosen by a majority vote of the committee. The CEC was the
most important party unit, with a membership overlapping the cabinet's.
The two bodies were practically indistinguishable. Chairmanship of the
CEC was a nominal post. Actual power rested in the hands of a secretary
general, a post held by Lee Kuan Yew since the party's founding. He was
assisted by a deputy secretary general who was charged with day-to-day
party administration.
Subordinate to the CEC were the branches, basic party units
established in all electoral constituencies. The branches were
controlled by individual executive committees, chaired in most cases by
the local delegate to Parliament. As a precaution against leftist
infiltration, the CEC approved all committee members before they assumed
their posts. One-half of the committee members were elected, and
one-half nominated by the local chairman. Branch activities were
monitored by the party's headquarters through monthly meetings between
members of the party cadre and the local executive committee. The
meetings provided a forum for party leaders to communicate policy to
branch members and a means to maintain surveillance over local
activities.
The party's cadre system was the key to maintaining discipline and
authority within the party. Individual cadres were selected by the CEC
on the basis of loyalty, anticommunist indoctrination, education, and
political performance. Cadre members were not easily identified but were
estimated to number no more than 2 percent of the party's membership of
1989, a list of cadres had never been published.
Although clearly the dominant party, the PAP differed from the ruling
parties of pure one-party states in two significant ways. Unlike the
leaders of communist parties, the leaders of the PAP made no effort to
draw the mass of the population into the party or party-led
organizations or to replace community organizations with party
structures. Singapore's leaders emphasized their government roles rather
than their party ones, and party organizations were largely dormant,
activated only for elections. Compulsory voting brought the electors to
the polls, and the record of the government and the fragmented state of
the opposition guaranteed victory to most if not all PAP candidates. In
many general elections, more than half of the seats were uncontested,
thus assuring the election of PAP candidates. The relatively weak party
organization was the result of the decision of the leaders to use
government structures and the network of ostensibly apolitical community
organizations to achieve their ends. By the 1970s and 1980s, the leaders
had confidence in the loyalty of the public service and had no need for
a separate party organization to act as watchdog over the bureaucracy.
The government was quite successful at co-opting traditional community
leaders into its system of advisory boards, committees, and councils,
and felt no need to build a distinct organization of party activists to
wrest power from community leaders. Second-echelon leaders were
recruited through appointment and co-optation and were preferentially
drawn from the bureaucracy, the professions, and private enterprises,
typically joining the PAP only when nominated for a Parliamentary seat.
The path to Parliament and the cabinet did not run through constituency
party branches or the PAP secretariat. In the view of the leadership,
political parties were instruments used to win elections and could be
dispensed with if there was little prospect of serious electoral
competition.
Singapore
Singapore - Political Dynamics
Singapore
Power Structure
In 1989 political power in Singapore was in the hands of a small
group of individuals who had been instrumental in Singapore's gaining
independence. The leadership core ruled through a second echelon of
potential successors, who tended to be technocrats, administrators, and
managers rather than politicians or power brokers. The PAP leaders,
convinced that a city-state without natural resources could not afford
the luxury of partisan politics, acted after 1965 to
"depoliticize" the power structure. Economic growth and
political stability would be maintained instead by the paternal guidance
of the PAP. Politics, as a result, was only exercised within very narrow
limits determined by the PAP. Singapore was thus administered by
bureaucrats, not politicians, in a meritocracy in which power was gained
through skill, performance, and demonstrated loyalty to the leaders and
their policies.
At the top of the hierarchy in 1989 were fifteen cabinet ministers,
who were concurrently members of Parliament and the CEC, the PAP's
highest policy-making body. Among these ministers was an inner core of
perhaps five members. Below this group was a tier of senior civil
servants who, in addition to their official duties, filled managerial
and supervisory roles as directors of public corporations and statutory
bodies. PAP members of Parliament without cabinet or government
portfolios also tended to function at this level of the power hierarchy,
providing links between the government and the populace.
Rifts within the leadership were rare. Although minor differences
over policy may have existed, the top leaders presented a united front
once decisions were made. The mode of decision making was consensus, and
the style of leadership was collective, but in 1989 Prime Minister Lee
Kuan Yew was by far the first among equals on both counts. The leaders
identified themselves with the nation, were convinced that they knew
what was best for the nation, and interpreted opposition to themselves
or their policies as a threat to the country's survival.
The overwhelming majority of the leadership were not propertied or
part of the entrepreneurial class. They did not appear particularly
motivated by profit, gained lawfully or through corruption (which was
almost nonexistent), or by the perquisites of their office (which
although increasing, remained less than could be achieved in the private
sector). Their reward, instead, derived from their access to power and
their conviction that they were working for the nation and its long-term
survival. Prime Minister Lee Kuan Yew and his close associates were
highly conscious of their roles as founding fathers of the new
city-state.
The power structure was extremely centralized. It was characterized
by a top-down style, featuring appointment rather than election to most
offices; the absence of institutional restraints on the power of the
prime minister and cabinet; and more effort devoted to communicating the
government's decisions and policies to the public than to soliciting the
public's opinion. The high degree of centralization was facilitated by
the country's relatively small size and population. Although members of
Parliament were elected by the public, they were selected by the core
leadership, often ran unopposed, and regarded their positions as due to
the favor of the prime minister rather than the will of the voters. At
the highest levels, the distinction between the bureaucracy and the
political offices of Parliament was only nominal, and many members of
Parliament were selected from the upper ranks of the civil service and
the public enterprises. Many high-level civil servants had direct access
to the prime minister, who consulted them without going through their
nominally superior cabinet minister.
Political Culture
Singapore possessed a distinct political culture, which fit into no
simple category formulated by political scientists. It was centralized,
authoritarian, and statist. It was also pragmatic, rational, and
legalistic. In spite of possessing the superficial trappings of British
institutions such as parliamentary procedure and bewigged judges,
Singapore was, as its leaders kept reiterating, not a Western country
with a Western political system. Although elections were held regularly,
elections had never led to a change of leadership, and citizens did not
expect that political parties would alternate in power. Nor was there a
tradition of civil liberties or of limits to state power. The rulers of
an excolony with a multiethnic population, and a country independent
only by default, assumed no popular consensus on the rules of or limits
to political action. Singapore was a city-state where a small group of
guardians used their superior knowledge to advance the prosperity of the
state and to bring benefits to what they considered a largely ignorant
and passive population.
Singapore's leaders were highly articulate and expressed their
principles and goals in speeches, books, and interviews. Their highest
goal was the survival and prosperity of their small nation. They saw
this as an extremely difficult and risk-filled endeavor. Conscious of
the vulnerability of their state and aware of many threats to its
survival, they justified their policy decisions on the grounds of
national survival. They viewed government as an instrument intended to
promote national ends and recognized no inherent limits on government
concerns or activities. They prized intellectual analysis and rational
decision making, and considered their own decisions the best and often
the only responses to problems. The senior leadership prided itself on
its ability to take the long view and to make hard, unpopular decisions
that either responded to immediate dangers or avoided problems that
would become apparent one or two decades into the future. They valued
activism and will, and tried to devise policies, programs, or campaigns
to deal with all problems. In a characteristic expression of Singapore's
political culture, the rising young leader Brigadier General (Reserve)
Lee Hsien Loong, when discussing the threat to national survival posed
by declining birth rates, said "I don't think we should ...
passively watch ourselves going extinct." Passivity and extinction
were linked and identified as trends the government's policies must
counter.
The leadership's conviction of the state's vulnerability to manifold
dangers and of the self-evident correctness of its analysis of those
dangers resulted in very limited tolerance for opposition and dissent.
According to Singapore's leaders, their opponents were either too
unintelligent to comprehend the problems, too selfish to sacrifice for
the common good, or maliciously intent on destroying the nation.
Although by the 1980s Singapore had the highest standard of living in
Southeast Asia, its leaders often compared it with generalized Third
World countries. They saw such countries suffering from widespread
corruption and demagogic politics, both reflecting concentration on
immediate payoffs at the expense of long-term prosperity and the common
good. For Singapore's leaders, politics connoted disruptive and
completely negative activities, characterized by demagoguery,
factionalism, and inflammatory appeals to communal, ethnic, or religious
passions. When they spoke of "depoliticizing" Singapore's
government, they had this view of politics in mind.
Singapore
Singapore - KEY POLITICAL ISSUES
Singapore
Succession
Prime Minister Lee Kuan Yew marked his sixty-fifth birthday in
October 1988 and celebrated thirty years as prime minister in May 1989,
and the question of political succession received increasing attention.
The prime minister and his long-time associates devoted a good deal of
their attention to the issue during the mid- and late 1980s. They
continued their efforts to identify promising younger leaders and bring
them into the cabinet. The process of selection was an elaborate one,
which began by identifying welleducated administrators from the public
service or private sector. Those people selected would be promoted to
managerial positions, often when in their thirties; those who succeeded
would be considered for appointment to a government position, often by
being designated a parliamentary candidate. In addition to identifying
good administrators, the older leaders tried to select persons of
integrity and good character who were able to work as members of a team.
Second-generation leaders were then tested by being given ministerial
portfolios and encouraged to go out and meet the common people. The
selection favored technocrats and administrators and rewarded those able
to defer to senior leaders and get along smoothly with their peers. The
senior leaders were aware that the process did not test the ability of
the second-generation leaders to cope with a severe political crisis,
but apparently could find no way to select for that skill.
The first-generation leaders were confident of their own rectitude
and ability to use their very extensive powers for the common good, but
they were not confident that their successors would be so
self-restrained. Throughout the 1980s, they considered various limits on
executive power that would minimize the possibility of arbitrary and
corrupt rule. These included constitutional changes such as a popularly
elected president with significant powers. The leaders claimed, perhaps
with hindsight, that their refusal to build up the PAP as a central
political institution and their efforts to bring a wide range of
low-level community leaders into the system of government advisory
bodies reflected a deliberate effort to disperse power and, in this
sense, to "depoliticize" the society. The effort to encourage
the circulation of elites between the government and the private sectors
and between the military and the civilian structures served the same
end. In so centralized a system, much depended on the decisions of the
prime minister and undisputed leader, who was reluctant to appoint a
designated heir or to approve any measure that would diminish his
authority. The expectation clearly was that a much more collective
leadership would replace the old guard.
The next generation of leaders, called the "new guard," was
led by Lee Kuan Yew's son, Lee Hsien Loong. A brigadier general in the
army, he first attained prominence in mid-1984 when he was cited as a
possible candidate for the December 1988 general election. His
prominence soared when, as minister of trade and industry and second
ministor of defence (services), he was appointed head in 1986 of the
critical Economic Committee assigned to redraft Singapore's economic
strategy.
Lee Hsien Loong's ascendancy and his consolidation of administrative
and political power assisted the political fortunes of bureaucrats who
formerly had served in the Ministry of Defence (known as the
"Min-def mafia") and ex-army officers who had served with Lee
when he was a brigadier general. The ascendancy of the socalled
"Min-def/ex-army officer group" under Lee initially was
suggested by some observers when the Singapore Armed Forces (SAF)
appeared to assume new importance in government policy decisions. In
March 1989, when the government announced a substantial pay raise for
the civil service, the military received an even larger raise with
guarantees that future raises would be consistently higher than those
allotted for the civil service. The government also announced that the
policy of assigning SAF officers to twoyear rotations in civil service
positions would continue. The policy ensured that the SAF would be
represented in all branches of the government and that the distinction
between the civilian and military bureaucracies would be less clear.
The younger Lee's ascendancy to positions of greater power both in
the PAP and the cabinet demonstrated his increased political stature. He
was elected second assistant secretary general of the party in 1989, a
post that had been vacant since 1984. This position placed him second in
line in the party hierarchy behind his father and Goh Chok Tong, who was
first assistant secretary general of the party and deputy prime minister
and minister for defence in the cabinet. Lee enhanced his position in
the cabinet when, as minister for trade and industry, he was named
chairman of a special economic policy review committee. In this
capacity, he gained the power to review the policies of all the
ministries for their economic impact on Singapore. Previously such
reviews were conducted only by the Ministry of Finance. Some Singapore
observers speculated in 1989 that Lee would one day be appointed
minister for finance and add control of Singapore's purse to his
influence over the armed forces.
Generational ties supplemented the institutional links. Lee Hsien
Loong and his associates were in their mid- to late thirties in 1989.
Lee's nearest rival for power was Goh, who was forty-seven years old.
Goh and his few allies in the cabinet, who were in their mid- to late
forties, appeared to be increasingly losing ground to the younger group,
however. For those with a military background, the military connection
remained important even though they had resigned from the military
before undertaking their civilian posts. The obligation of all males to
periodically undergo reservist training assured that the military
connection was not severed. If the army became a source of future
cabinet ministers, some political observers expected that ethnic Malays
and Indians would find it even more difficult to gain access to senior
government positions. Ironically, the army in pre-independent Singapore
was predominantly Malay and Indian. After independence, however, the
government changed this bias by increasing Chinese representation
through universal conscription.
Singapore
Singapore - Relations Between State and Society
Singapore
By the late 1980s, Singapore's leaders generally agreed that the
extensive economic and social transformation achieved after independence
required a changed pattern of relations between the government and
society. Government policies and practices devised to deal with the much
simpler economy and less educated and prosperous citizenry of the 1960s
were becoming increasingly ineffective in the 1980s. The major issues
were economic, involving debate over the optimal form of government
involvement in the economy, and political, centering around highly
contentious questions of the limits of government efforts to regulate
the lives of citizens and to suppress dissent and criticism.
The Government's Economic Role
Singapore had achieved economic success with an economy that was
heavily managed by the government. The state owned, controlled, or
regulated the allocation of capital, labor, and land. It controlled many
of the market prices on which investors based their investment decisions
and was the exclusive provider of social services and infrastructure.
The 1985- 86 recession, however, stimulated discussion of impediments to
economic performance and of dysfunctional aspects of the government's
role in the economy. A 1987 report by the governmentappointed Private
Sector Divestment Committee recommended that the state dispose of most
of its interest in private companies over a ten-year period. It
recommended privatizing forty-one of ninetynine government-controlled
companies and investing the proceeds in high-technology companies.
Throughout the 1970s and 1980s, the government controlled wages
through the annual wage guidelines set by the National Wages Council, a
body in which representatives of employers, trade unions (which were
controlled by the PAP), and the government reached a consensus on wage
levels for the coming year. The council's wage guidelines were in the
form of macroeconomic projections and were applied across the board in
all sectors of the economy. In December 1986, the cabinet approved a
National Wages Council report calling for a revised wage system that
permitted greater flexibility, (the flexi-wage policy) with more use of
bonuses and wage increases linked to increases in productivity. It was,
however, not clear how the productivity of white-collar workers and
civil servants, who constituted an increasing proportion of the work
force, was to be measured. The call for wages to reflect the
productivity and profitability of particular industries and firms
implied more bargaining between workers and employers and a diminished
role for the government, which could not impose a single rate on
hundreds of distinct firms.
Although there was general agreement on the need for changed economic
policies and modes of administration, significant tensions remained
between those who favored greater flexibility and liberalization and
those who wanted government direction of the economy. For Singapore's
leaders, the challenge was to devise more sophisticated means of
ensuring overall control while permitting greater autonomy and
flexibility at lower levels.
The Limits of Government Control
The highly ordered quality of life in Singapore itself became a
political issue. Many citizens felt that they were overregulated ,
governed by too many laws that were too easy to break. Singapore's
leaders attributed the cause of the assumed decline of Western societies
to the excessive individualism fostered by Western culture and warned
that Singapore would suffer a similar fate unless saved by a national
ideology.
The perceived need for an ideology was a phenomenon of the 1980s.
Previously, Singapore's leaders had been concerned with physical
survival more than cultural survival and had dismissed official
ideologies as contrary to Singapore's status as an open port unfettered
by conventional wisdom or fashionable orthodoxies. In the 1980s, as
peace prevailed in the region, the government shifted its focus to the
cultural sphere. Cultural preservation replaced physical survival as the
major concern of leaders who feared being overrun by foreign cultures.
Looking ahead, senior leaders identified two major dangers to the
nation: the failure of the nation to reproduce itself and the loss of
national identity. The first threat was manifested in steadily falling
birth rates, particularly among the nation's best educated citizens,
many of whom failed even to marry. The second threat, loss of identity,
it was feared, would lead to loss of cohesion and hence to the
destruction of the nation.
Singapore's leaders addressed these problems by proposing a series of
policies intended to encourage citizens to marry and reproduce and to
create a distinct Singaporean identity. The programs addressing the
population problem included extensive publicity and exhortation, along
with material incentives for giving birth to third and fourth children.
Women university graduates were singled out for special attention
because of their failure, in general, to marry and pass on their
supposedly superior genes. The efforts to foster a Singaporean identity
involved defending positive traditional Asian values against the
perceived threat from Western culture. Both the schools and the society
at large emphasized mastering Asian languages, such as Mandarin Chinese,
and promoting Confucianism. Such programs, which attempted to modify the
personal and intimate behavior of citizens but did not clearly reflect
the demands of economic development, aroused a good deal of opposition,
especially from younger and better educated citizens. The leadership's
paternalistic style and its intolerance of criticism became political
issues and were blamed by some observers for the increased vote for
opposition candidates in the 1984 and 1988 elections.
Opponents of programs relating to Singapore identity claimed that the
leaders' purpose was to shift support for a national ideology into
support for the government and the ruling PAP. Promoting Confucianism,
for example, was a convenient means of convincing individuals to
subordinate their interests to those of society. Others held that the
government's real fear was not that Singapore would lose its culture or
values but that continued Westernization of the society would mean more
pressure for real democracy, more opposition candidates, and the
possibility of a change in government.
The electoral vote for the PAP dropped considerably, going from 75.6
percent in 1980 to 62.9 percent in 1984 and by a lesser amount to 61.8
percent in 1988. In 1988 the PAP campaign slogan was "More Good
Years" and the opposition had no solid issues with which to attract
support. The election resulted in another landslide victory for the PAP
and the winning of eighty out of eighty-one parliamentary seats.
The PAP's style of leadership emphasized control by a strong
bureaucratic leadership intolerant of political opposition. The PAP
mind-set has been traced to its battle for political preeminence with
its communist rivals in the 1950s and 1960s. In the late 1980s,
Singapore had one of Asia's highest standards of living and was not
regarded as fertile ground for a communist insurrection. The PAP
maintained that Singapore was too small for a two-party system to work
effectively and did not anticipate sharing power. It stymied the
development of a legitimate opposition by a range of political tactics,
such as using the provision of public services to induce citizens to
vote for PAP candidates. Critics also charged that the party controlled
the press, preventing the free flow of ideas. Although there was no
direct censorship of the press, newspapers were closely monitored and
radio and television stations were owned by the government.
Singapore
Singapore - Political Opposition
Singapore
In the elections of September 1988, the only opposition member to win
election was Singapore Democratic Party candidate Chiam See Tong who
repeated his 1984 victory. However, in the contest over eight additional
seats--two representing single-seat constituencies, and six representing
two newly formed three-member group representation constituencies-- the
PAP received less than 55 percent of the vote. Furthermore, under a
constitutional amendment passed in 1984, the opposition was to be
allotted three parliamentary seats, whether it won them or not. Thus, as
a result of the 1988 election, in addition to Chiam, the opposition was
permitted to seat two additional, nonconstituency, nonvoting members of
Parliament in the new Parliament.
In the 1988 elections, Lee Siew Choh, a candidate of the Workers'
Party and one of the two opposition members chosen to sit in Parliament
as nonvoting members, was forced on the campaign's opening day to go to
court and pay damages for comments he made about PAP during the 1984
election. The other opposition member, Francis Seow, faced trial for
alleged tax evasion, and, if convicted, faced disqualification from
Parliament. Shortly afterwards, Prime Minister Lee threatened to bring a
defamation suit against Workers' Party leader J. B. Jeyaretnam. Another
Workers' Party candidate, Seow Khee Leng, was threatened by the
government with bankruptcy proceedings. All three had been successfully
sued by Lee for slander in earlier elections.
The state of the opposition was rooted in the PAP's drive, beginning
in 1963, to suppress all communist and leftist influence in Singapore.
The government discouraged opposition political activity through the use
of open-ended laws such as the Internal Security Act (ISA), which was
originally intended to deal with armed communist insurrection during the
Malayan Emergency of 1948- 60. This law permitted the indefinite
detention by executive order of any person suspected of leftist or
procommunist activity. Amnesty International frequently cited Singapore
for using the ISA to suppress legitimate, nonviolent political
opposition. That organization also cited Singapore's use of deprivation
of citizenship and banishment as means of repression. The government
often associated opposition with foreign manipulation, which compounded
its fear of dissent of any kind.
There were few issues on which the PAP could be challenged. Under PAP
rule, Singapore had achieved unprecedented economic prosperity as well
as marked social progress in racial harmony, education, health care,
housing, and employment. The PAP's achievements had created a popular
confidence in the party that was difficult to overcome. The opposition
parties themselves were divided along racial and ideological lines and
unable to compete with the PAP as a common front.
In May and June 1987, twenty-two people were detained without trial
under the ISA for alleged involvement in a communist conspiracy. All
detainees were released by the end of the year with the exception of
Chia Thye Poh, who was held for more than two years. A virulent critic
of the government and former member of Parliament representing the Barisan
Sosialis (The Socialist Front), he was finally
released in May 1989 after having been detained since October 1966.
Although Chia was never charged, the government alleged that he was a
member of the outlawed Communist Party of Malaya (
CPM), assigned to infiltrate the Barisan Sosialis in
order to destabilize the government. In 1987 amendments were made to the
Parliament Privilege, Immunities, and Powers Act of 1962, giving
Parliament the power to suspend any parliamentary member's immunity from
civil proceedings for statements made in Parliament and to imprison and
fine a member if he or she were found guilty of dishonorable conduct,
abuse of privilege, or contempt.
The Workers' Party, led by J.B. Jeyaretnam in 1989, was the principal
opposition party. The Workers' Party stood for a less regimented
society, constitutional reforms, less defense spending, and more
government social services. It was supported by lower income wage
earners, students, and intellectuals. Next was the United People's
Front, founded in December 1974 as a confederation of the Singapore
Chinese Party, the Singapore Islamic Party, and the Indian-supported
Justice Party. It campaigned for a more democratic political system. A
third party, ideologically to the left of both the United People's Front
and the Workers' Party, was the People's Front, established in 1971. In
1972 its campaign platform advocated a democratic socialist republic and
no foreign military ties. In 1973 the party's secretary general, Leong
Mun Kwai, received a six-month prison sentence for inciting the people
of Singapore to seize government leaders. Seventeen other opposition
parties were registered in 1989, including the Barisan Sosialis, once
the primary target of the government's political surveillance activities
because of its former role in antigovernment street demonstrations,
student protests, and industrial strikes. Lee Siew Choh, a nonvoting
member of Parliament in 1989, was the leader of the party's moderate
wing.
Singapore
Singapore - FOREIGN POLICY
Singapore
Governing Precepts and Goals
Minister for Foreign Affairs Suppiah Dhanabalan described the
governing precepts of the country's foreign policy in 1981 as a
willingness to be friends with all who sought friendship, to trade with
any state regardless of ideology, to remain nonaligned, and to continue
to cooperate closely with Association of Southeast Asian Nations (
ASEAN) members. These precepts, while consistent with the thrust of
foreign policy from the 1960s to the mid-1980s, failed to account for
the basic role that the survival of the nation played in determining
foreign policy goals. A primary foreign policy consideration until the
mid1980s , survival became an issue because of Singapore's size and
location and Indonesia's Confrontation ( Konfrontasi) campaign against
Malaysia in the 1960s. It was further linked to the concept of the
"global city" first proposed in 1972 by then Deputy Prime
Minister for Foreign Affairs Sinnathamby Rajaratnam. This concept
suggested that Singapore's survival depended on its ability to create a
continuing demand for its services in the world market. By implementing
a policy of international self-assertion, Singapore would shift from a
reliance on entrep�t trade and shipping to export-oriented industries.
The focus on survival was evidenced in Singapore's reaction to
Vietnam's invasion of Cambodia in 1978. Of the many issues surrounding
the event, one of particular interest to Singapore was Vietnam's blatant
disregard for the sovereignty of a small nation. Singapore's decision to
draw international attention to the situation was based, in part, on the
need for international recognition of its own sovereignty. Following the
invasion, Singapore heightened its international profile by expanding
diplomatic representation abroad and attending international forums.
Singapore was a member of ASEAN, the Nonaligned Movement ( NAM), the
Asian Development Bank ( ADB), the Group of 77, the International
Telecommunications Satellite Organization ( INTELSAT), and the United
Nations and its affiliated organizations.
With the passing of the first generation of leaders in the late
1980s, foreign policy was shaped less by the old fears produced by the
events of the 1960s and 1970s and more by the experience of regional
stability that prevailed during the formative years of the new guard or
second generation of leaders. The self-assertion of a decade earlier was
no longer required, and Singapore could afford to be less abrasive in
its foreign policy style. Foreign policy objectives in the late 1980s
were far more subtle than simple survival.
In March 1989, Singapore announced that it was charting a new course
of "economic diplomacy" to meet future international
challenges. It sought expanded economic ties with China, the Soviet
Union, several Eastern European nations, and the three nations of
Indochina: Laos, Cambodia, and Vietnam. In a speech to Parliament on
March 17, 1989, Minister of Foreign Affairs Wong Kan Seng announced that
Singapore was hoping to reverse its previous staunchly anticommunist
posture and normalize relations with several communist countries to
promote more compatible relationships based on mutual economic
interests.
Foreign policy also had to accommodate the views of predominantly
Islamic neighbors who were viewed by Singapore's leaders as possible
threats to its existence. As a gesture toward its neighbors and in
recognition of its own regional roots, Singapore maintained its
membership in the Nonaligned Movement, although it consistently rejected
neutrality as a foreign policy option. Singapore's leaders had reasoned
that avoiding entanglements with the great powers would leave Singapore
far too vulnerable to threats from regional neighbors, as Indonesia's
Confrontation campaign had demonstrated. Neutrality also was perceived
to be inconsistent with the Total Defence style of defensive vigilance
that the PAP attempted to instill in the citizenry following the Soviet
invasion of Afghanistan and the Vietnamese occupation of Cambodia. The
guiding concept of Total Defence was known as national integration and
was meant to unify a population made up of immigrants and a mix of
racial groups into a people with the "human will" to be
"unconquerable."
Foreign policy, therefore, stressed maintaining a balance of power in
the region. Singapore promoted the regional involvement of all great
powers because it feared aggravating a neighbor by relying on any one
power. Although it would have preferred relying upon the United States
to guarantee its security, such dependence would not have been tolerated
by the other ASEAN states. Singapore also remained suspicious of the
ability of the United States to pursue a consistent foreign policy
following its withdrawal from Vietnam.
Retaining its developing nation status was another foreign policy
goal of the 1980s. The 1989, however, Singapore lost the concession
enjoyed under the United States government's Generalized System of
Preferences ( GSP) on imports from developing countries and the ability
to borrow from the World Bank and the ADB at concessional rates.
<"57.htm">Regional
<"58.htm">The United States
<"59.htm">China
<"60.htm">The Soviet Union
Singapore
Singapore - FOREIGN POLICY - Regional
Singapore
Association of Southeast Asian Nations
Cooperation with ASEAN, which included Indonesia, Malaysia, Thailand,
the Philippines, and Brunei, was the center of Singapore's foreign
policy after 1975. Before 1975, Singapore's interests were global rather
than regional, and its policy toward ASEAN was characterized by
detachment. As the wealthiest country in Southeast Asia, it was
criticized for failing to help its neighbors. After 1975, however,
Singapore was criticized for being too ASEAN oriented, too active, and
too vocal in the organization for its size, particularly where matters
of regional security were concerned. The shift in Singapore's stance
toward ASEAN followed the communist victory in Vietnam in 1975, the
waning of a United States military presence in Asia, and new signs of
Soviet interest in the region. Furthermore, the other ASEAN states
permitted Singapore to assume a leading role in regard to the issue of
Vietnam's invasion of Cambodia in 1978. The situation in Cambodia in
fact, became the unifying force for the diverse countries belonging to
ASEAN. Singapore's Minister of Foreign Affairs Wong Kan Seng commented
in March 1989 that, if the situation were resolved, some other force
would be required to unite the member nations. The resolution of the
Cambodian conflict would also raise the possibility of Vietnam being
considered for membership, although in 1989 Singapore was not prepared
to support Vietnam's immediate entry.
ASEAN provided Singapore with a means of improving its bilateral
relations with Indonesia and Malaysia, two neighbors who were potential
threats to Singapore's security. Singapore's leaders never identified
the external enemy Singapore's armed forces were trained to deter. When
asked in 1984 who was Singapore's biggest threat, Prime Minister Lee
responded only that "the biggest threat...is that any threat will
come from someone bigger than us."
Malaysia
The acrimony that once characterized Singapore's relationship with
Malaysia began to change in the 1980s when the two countries adopted a
course of reconciliation. The improvement in relations began when
Mahathir Mohamad became prime minister of Malaysia. Lee Kuan Yew and
Mahathir achieved a personal rapport that established the tone for a
rapprochement, but Singapore's expulsion from Malaysia in August 1965
continued to color the relationship. Singapore's primary concern was
that Malaysia maintain a political system that tolerated multiracialism.
In Singapore's view, the undermining of this political principle in
Malaysia would have regional ramifications. Regional tolerance of
multiracialism, for example, might be reduced if an Islamic revival in
Malaysia led to the establishment of an Islamic state and the status of
Malaysia's Chinese population were subsequently endangered.
Singapore was linked with Malaysia militarily through the 1971
Five-Powers Defence Arrangement, an arrangement under which the security
of Singapore and Malaysia was guaranteed by Britain, Australia, and New
Zealand. Singapore cooperated with both Malaysia and Indonesia in
maintaining the security of the Malacca and Singapore straits. Another
link with Malaysia was the Inter-Governmental Committee (IGC), a forum
established in 1980 for the informal discussion of bilateral issues by
delegations headed by each country's minister for foreign affairs.
Indonesia
Singapore's relationship with Indonesia, like its relationship with
Malaysia, was built on a foundation of past discord, specifically
Indonesia's Confrontation campaign against Malaysia from 1963 to 1966.
After President Sukarno (1945-67) was deposed, relations were based to a
large degree on Lee Kuan Yew's personal relationship with President
Soeharto (1967---). Because bilateral relations lacked an institutional
foundation, they were vulnerable to the departure of either leader.
Indochina
Singapore's relationship with Indochina in mid-1989 permitted the
conduct of normal commercial transactions, but discouraged aid,
training, infrastructural development, and trade in strategic goods. In
April 1989, the Ministry of Home Affairs and the Ministry of Foreign
Affairs informed Singaporean companies that they could not invest in
Vietnam until the Vietnamese had withdrawn their troops from Cambodia.
The companies were allowed to conduct negotiations with Vietnam, but
could not commit any investments until the Vietnamese withdrawal was
complete. A few Singaporean companies had invested in Vietnam while
normal commercial transactions were still going on, before the
government had a clear policy concerning investments. Minister for
Foreign Affairs Wong Kan Seng indicated in 1989, however, that Singapore
was looking beyond the Cambodian problem to its future relations with
Indochina.
Singapore
Singapore - The United States
Singapore
Relations between Singapore and the United States became strained in
1988 after the United States was accused of meddling in Singapore's
internal affairs and a United States diplomat was expelled as a result
of the charge. The United States had objected to the government's policy
of restricting the circulation of several Hong Kong-based newspapers,
including the Asian Wall Street Journal and the Far Eastern
Economic Review, and to the use of the Internal Security Act to
detain indefinitely dissidents or those deemed a threat to the existing
order. The expelled diplomat was accused of instigating members of the
opposition to contest the 1988 elections. The essence of a speech on
United States-Singapore relations, given by Lee Hsien Loong before the
Asia Society in Washington, D.C., on May 16, 1989, was that the
relationship was strong but that the United States should refrain from
interfering in Singapore's internal affairs.
The United States was Singapore's largest trading partner in the
1980s. It also was viewed as a benevolent power whose military presence
in the region kept Soviet influence in check, balanced China's
increasing military strength, and obviated Japan's rearming. Singapore
was concerned, however, that the United States eventually would tire of
its role in the Asia-Pacific region. This concern was somewhat allayed
in 1989 when President George Bush, demonstrating his commitment to
maintain American interests in the area, both dispatched Vice President
Dan Quayle on an Asian tour and visited the region himself in the first
few months of his administration.
Singapore
Singapore - China
Singapore
In 1989 Singapore had not yet established diplomatic relations with
China, largely out of deference to Indonesia, the ASEAN state most
concerned about China's intentions in the region. Indonesia's move to
initiate diplomatic relations with Beijing in February 1989, however,
was expected to clear the way for Singapore to follow. Regarding
Indonesia's announced intentions, Singapore's First Deputy Prime
Minister Goh Chok Tong stated in February 1989 that it was
"logical" for Singapore "to follow suit"; however,
he saw no need to move hastily because Singapore already had a cordial
trading relationship with China. Singapore's trade with China in 1988
amounted to US$2.98 billion, a 27 percent increase over 1987. Reexports
to China were up by 108 percent over the same period.
The other side of improving relations with China was maintaining good
relations with Taiwan. Although Singapore lacked diplomatic ties with
Taiwan in 1989, the two enjoyed a flourishing economic exchange. Trade
with Taiwan in 1988 reached S$5.2 billion, exceeding that with China
($5.7 billion). Some analysts suspected, however, that once serious
negotiations to establish diplomatic ties began with Beijing, China was
likely to pressure Singapore to end its relationship with Taiwan,
particularly in matters of military cooperation such as the training in
Taiwan of Singaporean troops. Others speculated that the relationship
would not be affected. Lee Kuan Yew said in March 1989 that he did not
expect Singapore's relationship with Taiwan to change because both
countries had been aware for some time of Singapore's intention to
follow Indonesia in normalizing relations with China and both had taken
such a development into consideration. A visit by Taiwan's President Li
Teng-Hui shortly after Indonesia's diplomatic initiative, was
interpreted as a sign of continuing warm relations between Taiwan and
Singapore.
Singapore
Singapore - The Soviet Union
Singapore
In 1989 Singapore maintained both economic and diplomatic relations
with the Soviet Union. From the mid-1960s until the mid1970s ,
Singapore's leaders promoted trade relations with Moscow in the belief
that a Soviet role in Southeast Asia would ensure the permanent interest
of the United States in the region. The Soviet Union was viewed as a
major power and as a counterweight to China, and, therefore, as a
significant factor in maintaining the regional power balance. This view
changed when the Soviets established a military presence at Cam Ranh Bay
in Vietnam, following the signing of the Soviet-Vietnamese Treaty of
Friendship and Cooperation in November 1978, and actively supported the
Vietnamese invasion of Cambodia a month later. At that time, according
to Singapore, Moscow became a threat to regional stability.
Soviet diplomacy toward the region changed, however, in the mid-1980s
under the leadership of new geeneral secretary Mikhail Gorbachev.
Beginning with a milestone foreign policy address in Vladivostok in July
1986, he initiated extended ties with the ASEAN states and committed the
Soviet Union to playing a more constructive role in resolving the
Cambodian issue. His interest in improving ties with the region and his
new emphasis on Soviet economic development acted to modify regional
perceptions. Singapore, as well as many of its ASEAN partners, became
increasingly receptive to upgrading their bilateral relations with
Moscow.
Trade, banking, and shipping were the three critical areas of
Singapore's economic ties with Moscow. Singapore's exports were mainly
in the form of repairs to Soviet vessels in Singapore shipyards. Other
exports included rubber, coconut oil, and fuel oil. In return, the
Soviets exported fish and fish products, cast iron, light machinery, and
crude oil. Beginning in the mid-1980s, the Soviets encouraged
Singaporean firms to invest in joint ventures in the Soviet Union.
Singapore's shipyards were reported in 1988 to be interested in
reconstructing and developing the port of Nakhodka, the second largest
port in the Soviet Far East after Vladivostok.
Singapore
Singapore - THE MEDIA
Singapore
The government did not normally censor the press, but it owned the
radio and television stations and closely supervised the newspapers.
Under the Newspapers and Printing Presses Act (NPPA), passed in 1974 and
amended in 1986, the government could restrict-- without actually
banning--the circulation of any publication sold in the country,
including foreign periodicals, that it deemed guilty of distorted
reporting. These laws provided the legal justification for restrictions
placed on the circulation of such foreign publications as the Asian
Wall Street Journal and Time magazine's Asian edition in
1987. The government also restricted the circulation of Far Eastern
Economic Review and Asiaweek in 1987 for "engaging in
the domestic politics of Singapore."
Singapore had seven daily newspapers at the end of 1987: two in
English, The Straits Times and The Business Times;
three in Chinese, Lianhe Wanbao, Shin Min Daily News,
and Lianhe Zaobao; one in Malay, Berita Harian; and
one in Tamil, Tamil Murasu. With the exception of the Tamil
Murasu, all were published by Singapore Press Holdings Ltd, a group
that comprised Singapore News and Publications Ltd, the Straits Times
Press Ltd, and the Times Publishing Company. Daily newspaper circulation
in 1988 totaled 743,334 copies, with Chinese language newspapers
accounting for the highest number (354,840), followed by English
(340,401) and Malay (42,458) newspapers.
The Singapore Broadcasting Corporation (SBC) operated five radio
channels and three television channels. Established in 1980, it provided
programming in Singapore's four official languages-- Malay, Chinese,
Tamil, and English--and was supported by revenue from radio and
television licensing fees and commercial advertising. Each of four of
the five radio stations broadcast in one of the four official languages,
while the fifth alternated between English and Mandarin. The television
stations, which provided a total of about 163 hours of programming a
week, also broadcast in separate languages. Channel Five's programming
was in Malay and English, Channel Eight's in Mandarin and Tamil, and
Channel Twelve's in English. In many cases, programs also were subtitled
in several languages.
By 1989 Singapore's leadership had been in place for three decades,
during which it guided an extraordinarily successful program of economic
development and physical rebuilding. In the 1990s, a new generation of
leaders would take over, and the debate over the need to change the
political system that had been so successful in the past would grow.
Some elements of an increasingly prosperous and well-educated
population, who took Singapore's national viability and survival for
granted, questioned the elderly leaders' assertions that a host of
pressing dangers justified their authoritarian and paternalistic style
of governance. To the leaders, however, the country's prosperity and
their continued electoral victories demonstrated the correctness of
their policies and methods of rule. They envisioned a new generation of
leaders who would continue the proven practices established by the
country's founding fathers. The inherent tensions between generations
and between the advocates of change and those of continuity were likely
to mark the politics of the 1990s.
Singapore
Singapore - Bibliography
Singapore
Boyce, Peter. Malaysia and Singapore in International
Diplomacy. Sydney: Sydney University Press, 1968.
Brailey, Nigel J. Thailand and the Fall of Singapore.
Boulder, Colorado: Westview Press, 1986.
Callahan, Raymond. "The Illusion of Security: Singapore 1919-42,"
Journal of Contemporary History [London], 9, No. 2,
April 1974, 69-92.
Clutterbuck, Richard L. Conflict and Violence in Singapore and
Malaysia, 1945-1983. Singapore: Graham Brash, 1984.
------. Riot and Revolution in Singapore and Malaya,
1945-1963. London: Faber and Faber, 1973.
Colless, Brian E. "The Ancient History of Singapore," Journal
of Southeast Asian History [Singapore], 10, No. 1, March
1969, 1-11.
Crawfurd, John. Journal of an Embassy from the Governor-General
of India to the Courts of Siam and Cochin China, II.
London: H. Colburn, 1828.
Dartford, Gerald Percy. A Short History of Malaya. London:
Longmans, Green, 1957.
Drysdale, John. Singapore: Struggle for Success.
Singapore: Times Books International, 1984.
Esterline, John H., and Mae H. Esterline. "How the Dominoes
Fell": Southeast Asia in Perspective. Lanham, Maryland:
Hamilton Press, 1986.
George, F.J. The Singapore Saga. Singapore: Fernandez
Joseph George, 1985.
Hall, D.G.E. A History of South-East Asia. New York: St.
Martin's Press, 1968.
Jack-Hinton, Colin. A Sketch Map History of Malaya, Sarawak,
Sabah and Singapore. London: Hulton Educational, 1966.
Kennedy, J. A History of Malaya, A.D. 1400-1959. London:
MacMillan, 1962.
Lowe, Peter. Great Britain and the Origins of the Pacific
War. Oxford: Clarendon Press, 1977.
Mackie, J.A.C. Konfrontasi: The Indonesia-Malaysia Dispute,
1963-66. Kuala Lumpur: Oxford University Press, 1974.
Makepiece, Walter, Gilbert E. Brooke, and Roland St.J. Braddell
(eds.). One Hundred Years of Singapore. (2 vols.)
London: John Murray, 1921.
Marshall, David. "Singapore's Struggle for Nationhood, 1945-1959,"
Journal of Southeast Asian Studies [Singapore], 1,
No. 2, September 1970, 99-104.
Moore, Donald. The First 150 Years of Singapore.
Singapore: Donald Moore Press, 1969.
Ooi, Jin-bee, and Chiang Hai Ding (eds.). Modern
Singapore. Singapore: Singapore University Press, 1969.
Pluvier, Jan. South-East Asia from Colonialism to
Independence. Kuala Lumpur: Oxford University Press,
1974.
Rose, Saul. Britain and South-East Asia. London: Chatto
and Windus, 1962.
Ryan, N.J. A History of Malaysia and Singapore. Kuala
Lumpur: Oxford University Press, 1976.
------. The Making of Modern Malaysia and Singapore. Kuala
Lumpur: Oxford University Press, 1969.
Sheppard, Tan Sri Datuk Mubin (ed.). 150th Anniversary of the
Founding of Singapore. Singapore: Journal of the
Malaysian Branch of the Royal Asiatic Society, 1973.
Short, Anthony. The Communist Insurrection in Malaya,
1948-60. New York: Crane Russak, 1975.
Singapore. Ministry of Communications and Information.
Singapore, 1988. Singapore: 1988.
------. Ministry of Communications and Information. Singapore
Facts and Pictures, 1988. Singapore, 1988.
------. Ministry of Culture. Information Division. Singapore:
An Illustrated History, 1941-1984. Singapore: 1984.
Song Ong Siang. One Hundred Years' History of the Chinese in
Singapore. Singapore: Oxford University Press, 1984.
Tan Ding Eing. A Portrait of Malaysia and Singapore.
Singapore: Oxford University Press, 1975.
Tarling, N. Piracy and Politics in the Malay World.
Melbourne: F.W. Cheshire, 1963.
Tregonning, K.G. A History of Modern Malaysia and
Singapore. Singapore: Eastern Universities Press, 1972.
Trocki, Carl A. Prince of Pirates: The Temenggongs and the
Development of Johor and Singapore, 1784-1885. Singapore:
Singapore University Press, 1979.
Turnbull, Constance M. A History of Singapore: 1819-1975.
Kuala Lumpur: Oxford University Press, 1977.
------. A Short History of Malaysia, Singapore, and
Brunei. Stanmore, N.S.W.: Cassell Australia, 1979.
------. The Straits Settlements, 1826-67. London:
University of London, 1972.
Wurtzburg, C.E. Raffles of the Eastern Isles. London:
Hodder and Stoughton, 1954.
Yen Ching Hwang. The Overseas Chinese and the 1911 Revolution
with Special Reference to Malaya and Singapore. Kuala
Lumpur: Oxford University Press, 1976.
Singapore