Roberts A.D. The Cambridge History of Africa, Volume 7: from 1905 to 1940
Подождите немного. Документ загружается.
terrible droughts in
1930—
1 and 1932—3: tens of thousands of
livestock died. The 'poor white problem', a particular concern
of Afrikaner politicians and churchmen, thus became graver than
ever. The Carnegie Commission which surveyed it in 1928-32
reported that one in five Afrikaner men were poor whites, in the
sense of being unable to obtain
a
livelihood. However, it was black
rather than white workers who were first laid off by urban
employers and, unlike whites, blacks were under pressure to
return to drought-stricken reserves. Even on the narrowest
calculation of white self-interest, the physical condition of Africans
was alarming. In 1930 the government appointed a commission
to examine how employment among unskilled whites was threat-
ened by the very low wages accepted by black migrants. The
commission reported in 1932 that the root of the problem lay in
the frequently abysmal poverty of the reserves, where an average
annual income for
a
family, in both cash and kind, was worth £14.
Rural wages quite failed to provide African workers and their
families with a diet 'consistent with reasonable maintenance of
health', and one member of the commission (the chairman of the
Wage Board) considered that the condition of African farm labour
amounted to serfdom.
24
The depression, by aggravating racial conflict, intensified the
new government's determination to suppress opposition. Already,
in June 1929, it had condoned whites in Durban who had started
a riot at the ICU office in which eight people were killed, after
Africans had protested against the municipal beer-brewing
monopoly. The minister of justice, Oswald Pirow, sought to
terrorise Africans into subservience; in November 1930 he led
armed police in Durban who attacked compounds at dawn with
tear-gas under the pretext of rounding up tax-defaulters. In the
previous May, an amendment to the Riotous Assemblies Act gave
the government sweeping powers to deal with anyone whom it
claimed was promoting ' racial hostility': these powers were soon
used against farm-workers in the western Cape, Kadalie on the
Rand, Champion in Natal, and leaders of the CPSA. The CPSA,
moreover, was divided in 1931 by new directives to detach itself
from black
'
reformists' and instead mobilise both black and white
masses for revolution; several white trade unionists were expelled.
24
Report of the Native Economic Commission, ipjo-rpji (UG 22-32), 69; C. W. de
Kiewct, A history of South Africa: social and
economic
(London, 1941), 230.
583
Cambridge Histories Online © Cambridge University Press, 2008
SOUTHERN AFRICA
Meanwhile, there had been mounting pressure within Congress
to dissociate itself from the Communists, and in April 1930
Gumede was replaced as president by the deferential Pixley Seme.
But those in Congress who still had hopes of constitutional
advance received a new rebuff
in
1930-1,
when the voting power
of Africans in the Cape was devalued, first by the enfranchisement
of white women and then by the exemption of whites from the
' civilisation' franchise test. African rights were further eroded by
lawyers in 1934 when the Court of Appeal held that the
government did not need parliamentary authority to introduce
separate but (in theory) equal social services. It was also in 1934
that the Court of Appeal, in rejecting a challenge to the Riotous
Assemblies Amendment of
1930,
implicitly denied that judges had
any right to criticise legislation; this pronouncement was soon to
have a marked impact on judicial behaviour.
For all its severity against Africans, the Hertzog government's
handling of the economic crisis lost it a good deal of white
support. It was determined somehow to muster the two-thirds
majority needed to push through the Native Bills, but it seemed
most unlikely to achieve this on its own at the next election.
However, many Nationalists were now much readier than they
had been to contemplate alliance with Smuts's SAP. In the 1920s
the two parties were at odds over the issue of South African
autonomy. For the Nationalists, it was vital that the Union's
membership of the British Commonwealth should be the result
of choice, not constraint. At the Imperial Conference in 1926,
Hertzog asserted South Africa's right to secede, and in effect he
had his way: he influenced the drafting of Balfour's definition of
Dominion autonomy. In 1927 South Africa set up its own
department of external affairs. In 1931 the offices of governor-
general and high commissioner were separated while the Statute
of Westminster formally enacted the legislative sovereignty of
South Africa and the other Dominions. In March 1933 Hertzog
and Smuts took this accomplished fact as the basis for a wide-
ranging agreement on policy, and formed
a
coalition government;
Smuts became minister of justice and deputy prime minister. A
general election in May returned a massive majority for the
coalition and opened the way to a fusion of the two leading
parties. Early in
1934
parliament passed a Status of the Union Act
which used the Statute of Westminster to confirm South African
584
Cambridge Histories Online © Cambridge University Press, 2008
1924—194°
independence, and in particular its right to conduct its own
foreign policy. By September the Nationalist Party and the SAP
had merged in a new United Party. Republican-minded National-
ists feared fusion and 19 MPs followed D. F. Malan in forming
a Purified Nationalist Party; a few English-speakers, by contrast,
feared that fusion would encourage republicanism and, led by
Colonel Stallard, formed a Dominion Party.
The gold boom
During the middle and later 1930s the South African economy
expanded very considerably, chiefly as a result of the favourable
market for gold. Between the end of
1933
and 1939 the price rose
by
a
further
22
per cent. Goldmining became more important than
ever to the country's economy. Between 1932 and 1939 the annual
value of gold sales rose from £5om to £99m and usually accounted
for around three-quarters of total exports, as compared to about
a half in the previous decade. Since working costs remained
roughly stable, large profits were made. The rate of return on
goldmining capital increased and between 1933 and 1940 the
annual value of dividends was usually at least double the average
for the later 1920s.
25
These short-term gains made it possible to
secure the future of the industry in the longer term: from 1933
to 1938 it not only reinvested £nm but attracted new capital
worth £6 3 m. Due to the growth of local finance houses, notably
the Anglo American Corporation, a substantial proportion of this
new capital was contributed by South Africans, whose percentage
of mining shares rose from 14 in 1913 to about 40 in 1937.
26
This money was chiefly devoted to deepening existing mines and
developing new ones: work on the Far East Rand continued,
while from 1933 the Far West Rand was opened up. This work
did not begin to affect output significantly until the end of the
decade; indeed, output actually fell in 1933-4 and returned to the
1932 level only in 1937, since high prices rendered it economic
25
In
the decade before
the
revaluation
of
gold, the rate
of
return was around
11 per
cent, scarcely more than that
for
British equities; but
by the
mid-i93os
the
return
on
South African gold was
16
per cent while that
on
British equities fell
to 8 per
cent.
S. H. Frankel, Investment and the return to equity capital in the South African gold
mining
industry, 1SS7-19S) (Oxford, 1967),
8,
44,
124.
26
S. H.
Frankel and
H.
Herzfeld, 'An analysis
of
the growth
of
the national income
of the Union
in the
period
of
prosperity before
the
War', South African Journal
of
Economics, 1944,
12, 2,
112-15; Frankel, Capital investment,
93
n.2, IOJ.
585
Cambridge Histories Online © Cambridge University Press, 2008
SOUTHERN AFRICA
to mill lower grades of ore. But between 1932 and 1937 the
effective mining area of the Rand was increased fourfold. Mean-
while, the diamond industry began to expand once more; the
South African cartel was reorganised in 1933 and up to 1937 the
value of sales slowly rose, though throughout the 1930s annual
output remained lower than in almost any year since 1880.
The prosperity of the goldmining industry was promptly
exploited by the government. The state was already the residual
owner of the country's gold resources: since 1908 it had gained
revenue from the industry not only by levying income tax but by
issuing mining leases which yielded a share in profits on a sliding
scale. In 1925-7 government revenue from the goldmines was
about 23 per cent of their gross profits. In 1933 the revaluation
of gold encouraged the government to introduce an excess profits
tax. Between 1932-3 and 1933-4 total government revenue rose
by almost one-third to £37.6m, of which
33
per cent came directly
from goldmining (compared with 8 per cent in 1913). In 1934-6
total direct government income from the goldmines was 42 per
cent of taxable profits and around three-quarters of the value of
dividends (compared with just over one-third ten years earlier).
27
In 1935 the goldmining industry claimed that the burden of
government levies was far heavier than that on gold-producers
in the USA, Australia or Canada, and the tax system was further
revised to increase incentives to exploit lower-grade ores and thus
extend payable reserves. Thereafter, government receipts from
goldmining gradually declined, but between 1933-4 and 1938-9
they amounted to £i
2m
— more than four times their value in the
period between 1923-4 and 1928-9.
The expansion of goldmining, and the greatly increased share
of the state in its fortunes, had far-reaching effects. Between 1932
and 1940, national income rose by about three-quarters, and in
real as well as cash terms, for prices rose little. Swollen revenues
enlarged the government's capacity to shape the economy.
Increased spending on public works boosted the construction
industry, and better roads lowered transport costs. There were
also augmented resources available to finance the apparatus of
protection for South African producers. The state-owned iron and
17
The rate of actual tax (as distinct from tax income from mining leases and credits
to the government loan account) amounted to 26, 29 and 28.5 per cent for these years.
Cf. Frankel, Capital
investment,
86, 97-8, 113, 116.
586
Cambridge Histories Online © Cambridge University Press, 2008
1924-194°
steel industry came into production in 1933. The value of
manufacturing output doubled between 1929 and 1939, when it
represented 18 per cent of national income; this growth was due
to protection, to the investment of mining profits and to the
establishment of South African branches by British and US firms.
The structure of agricultural subsidies continued and in 1933—7
was extended to wool-producers. Between 1910 and 1936 the
government spent £112111 on agriculture, almost wholly for the
benefit of white farmers.
The huge investment in white farming produced very mixed
results. The total output of white farmers increased, but more
slowly in the 1930s than in the 1920s. Subsidies cushioned
inefficiency as well as innovation, and they encouraged the use of
marginal land to the point of exhaustion, so that on white as well
as black farms soil erosion reached alarming proportions. The real
significance of state aid to white farming, as to manufacturing,
was political rather than economic. Hertzog's government was
much concerned to stem the drift of Afrikaners from the land.
The industrial boom increased the number of white wage-earners
by 100,000 between 1932 and 1939, while the white rural
population steadily declined. Though the 'poor white problem'
was fast disappearing, controls were still needed to prevent
unskilled white workers from being undercut by blacks. It was
thus important that the demand for jobs by unskilled whites fresh
from the land should not outstrip the capacity of employers to
pay wages appropriate to whites. In 1935 the Labour Department
spent £800,000 on subsidising job-creation for poor whites. The
Marketing Act of 1937, which further extended government
powers of farm subsidy, was specifically intended to narrow the
gap between white incomes on towns and farms.
The increased output which was encouraged by farming
subsidies required more use of African labour. White farmers who
had retained African squatters owing labour-rent increasingly
sought to replace them with full-time wage labour. Supplies were
not hard to find. The black population was steadily growing
28
, yet
the carrying capacity of the reserves was declining, especially after
the droughts of the early 1930s. Between the wars there was an
28
See table 4. Average annual rates of population increase (percentages), 1921-56:
Coloured, 2.52; black, 2.29; Asian, 1.90; white, 1.86. White immigration reached a peak,
at over 10,000, in 1936; it was partly offset by emigration.
587
Cambridge Histories Online © Cambridge University Press, 2008
SOUTHERN AFRICA
Table
4.
South Africa, population
{percentages
in
parentheses).
'Natives'
Total
1
Reserves
1
White farms
2
Urban areas: total
4
''
women
5
'
6
under
14
s
*
6
Johannesburg: total''
9
women
10
women employees
10
•Whites'
Total'
Rural areas*
Urban areas'
Coloured
Total'
•Asiatic'
Total
1
Urban population
Total
5
White>
'
Native '«•»••
Coloured
5
'
6
'
Asiatic
>5
>
6
1911
4,019,006
508,142 (126)
101,971
4357
1,276,242
617,9)6
6)8,286
5*5.943
152,20)
1,477,868
658,286(44)
508,142
()4))
1921
4,697,81)
j 87,000(125)
'47.*93
87,000
118,652
12,160
5633
1,519,488
671,980
847,508
545.548
'65.73'
'.7)5.795
847.
J 08 (49)
587,000 ()))
249,968 (i4"4)
JI,2O9()O)
19)6
6,596,689
2,42O,)49
2,O5),44O
1,141,642(17))
356.874
226,944
2)1,889
60,992
24,781
2,OO),857
696,471
1,307,586
769,661
219,691
3.OO9.5
3'
1.307.386
(434)
1,141.64*
(380)
414.907 (>3'8)
I45.J96
(4'8)
1
Official yearbook
of
the Union
of South Africa,
no. 22
(1941),
984.
1
Ibid.,
398.
1
Ibid.,
99).
4
Union
yearbook:
statistics for
the period
1910-2) (1927),
86).
5
Ibid.,
864.
6
Report on
1936
Census,
vol.
IX
(UG 50,
19)8).
7
Report of the Native Lavs
Commission
(UG 28,
1948),
6.
8
J. P. R.
Maud, City
Government
(19)8),
584.
•
Report on
1946
Census,
vol. X (UG 51,
1949),
78ff.
10
Gaitskell,
as
cited
on
p. 590,
n. 52.
overall,
if
still very gradual, decline
in
agricultural productivity
in
the
reserves.
29
And
despite
the
great increase
in
government
revenue after 1934 there was no reduction
of
African taxation.
By
1937 there were 400,000 Africans working
on
white farms
(as
against 360,000
in
1930), and once there they were frequently tied
2
» Steep decline dates only from
the
1950s:
cf. C.
Simkins, 'Agricultural production
in
the
African reserves',
Journal
of
Southern
African Studies, 1981,
7, 2,
256-83.
588
Cambridge Histories Online © Cambridge University Press, 2008
down not only by pass-laws but by cash debts and much-prized
grazing rights. The latter helped workers to support families:
altogether, there were over two million Africans living on white
farms in 1936. But the combined effect of immiseration in the
reserves and artificial barriers to mobility was to depress real farm
wages even below their value in 1910.
Other fields of employment expanded fast. Between 1929 and
1939 the black workforce in manufacturing nearly doubled, while
that on the mines rose by one-third to 424,000, the first major
increase since 1907—10. During the worst of the depression, from
1929 to 1933, the Transvaal mines reduced their reliance on
contract labour from Mozambique
30
in favour of men from
within the Union, of whom only 40 per cent were contracted by
labour recruiters. In the same period, the proportion of mine-
workers from the Cape in the Transvaal rose from 30 to 43 per
cent, even though by 1935 one-third of job-seekers from the
Transkei were rejected by the mines on medical grounds. Mean-
while, the recent discovery of a sulphonamide drug provided a
cure for the pneumonia to which mineworkers from the tropics
were specially prone. In 1934, under pressure from the mine
companies, the government allowed a gradual resumption of
recruitment north of latitude 22°S, and by 1939 the Transvaal
mines employed some 20,000 men from central Africa, mostly
from Nyasaland. On the goldmines, increasing mechanisation
underground widened the scope for white workers: by 1939 there
were 43,000, almost twice as many as in the later 1920s. However,
there was no marked change in real wages for either black or white
mineworkers, which in 1941 were much the same as they had been
in
1911.
31
This was broadly true of workers in other industries
and in government or municipal employment. Annual wages for
black workers in manufacturing averaged £41 to £46 in 1924-37;
mineworkers earned two-thirds as much.
By and large, the boom in goldmining had the effect of
intensifying existing trends towards white proletarianisation and
black dependence on wage labour. Between 1921 and 1936 the
number of whites in urban areas increased by a
half,
yet that of
blacks in urban areas doubled. Moreover, there were clear signs
that Africans were tending increasingly to make their homes in
J0
A new labour convention had been made with Portugal in 1928; cf. p. 516.
Jl
F. Wilson, Labour in the South African gold
mines,
1911-1969 (Cambridge, 1972), 46-
589
Cambridge Histories Online © Cambridge University Press, 2008
SOUTHERN AFRICA
towns. Whereas in
1921
there were three black men for every black
woman in towns, in 1936 the ratio was only
2.2:1,
and a larger
proportion of urban Africans were children. African settlement
had gone much further in some towns than in others: in Port
Elizabeth, the sexes had long been evenly balanced. The patterns
of urbanisation were clearest on the Witwatersrand. By 19 j
7
the
total population of the area was over a million, and over half were
Africans. Most Africans in employment were housed by their
employers, usually as single men (as on the mines); they were
certainly transient. But the increase in African urban employment
meant more opportunities for self-employed Africans such as
retail traders, craftsmen and herbalists. Besides, there were said
to be over 90,000 Africans 'living by their wits', without
any obvious source of income. And the 60,000 who lived in
Johannesburg's municipal locations consisted mostly of families;
they were regarded by officials as permanent residents. Between
1921 and 1936 the black female population of Johannesburg
increased fivefold, and less than half were in white employment
(mostly as domestic servants). Over 500 Africans had bought land
in areas set aside by the municipality for 'non-white' owner-
occupiers, while others owned plots outside the city boundary.
32
Segregation
The growth of African urban settlement between the wars
sharpened racial antagonism. The more stabilised Africans threat-
ened white jobs; the
L-umpenproletariat
threatened social order.
There was increasing pressure to keep Africans out of towns,
except as temporary wage-earners. In 1930 there were 40,000
convictions under the pass-laws, and the government empowered
municipal authorities to exclude women who could not prove they
had accommodation. But there was also wider-ranging debate on
the place of Africans in South African society. Hertzog's Native
Bills became a focus for the discussion of segregation at all levels.
The 1926 bills had been referred to a parliamentary committee
which included the member for Zululand, G. Heaton Nicholls,
32
Ray E. Phillips, The
Bantu
in the city (Lovedale, 1958), xxiv, xxvi—xxvii; Deborah
Gaitskell, "Christian compounds for girls": Church hostels for African women in
Johannesburg, 1907-1970'',
journal
of
Southern
African
Studies,
1979, 6, 1, 48. Neither
the 1913 Land Act nor the 1936 Native Trust and Land Act applied to urban areas.
590
Cambridge Histories Online © Cambridge University Press, 2008
i924—1940
a leading sugar-planter. In 1931 Nicholls (echoing recent state-
ments by Smuts) argued that the only way to prevent African
proletarianisation and consequent class war was to strengthen the
economic resources of the reserves and foster African 'race pride'
by restoring the power of African chieftainship. In these objectives
he had the support not only of the Zulu royal family but also of
John Dube, the ANC leader. Dube, like other educated Zulus,
was related to a chiefly family and in the 1920s made common
cause with the richer African farmers who themselves employed
wage labour and shared white fears of the ICU. Such Africans also
had an interest in playing down class-conflict, and to this end
sought both to relieve mounting pressure on land and to
strengthen attachment to traditional rulers. In 1931 Dube told
Nicholls that Africans wanted land, not votes, and he toured the
Union to gain African support for Nicholls's plan to couple the
disfranchisement of black voters in the Cape with a large
extension of the reserves and special funds for their development.
Hopes that segregation could be given some such positive
content received little encouragement from the bills presented to
parliament in 1935. The Native Trust and Land Bill finally
adopted the recommendations of the Beaumont Commission in
1916 to increase the extent of'scheduled areas' in African owner-
ship by about 75 per cent.
33
But this reform was hedged about
with qualifications. Africans might continue to buy land within
the enlarged scheduled areas, but out of their own pockets; all
the remaining additional land would eventually be bought for
communal ownership through
a
government fund, and occupation
would be conditional on satisfying official standards of land use.
Thus chiefs, who distributed communal land,-were favoured at
the expense of prospective African freeholders. Besides, much of
the additional land was already occupied by Africans living on
white property, so that the bill's effect on land scarcity was slight.
Furthermore, Africans in the Cape were finally prohibited from
buying land from whites outside scheduled areas. At the same
time,
the bill obstructed African farming on white land; it
outlawed cash tenancies and extended to 180 days the period of
service which could be exacted from labour tenants. The political
11
This figure takes account of some 2.5 million acres bought by Africans since 1913
with the governor-general's permission: cf. T. R. H. Davenport,
South
Africa: a
modern
history
(London, 1977), 336; the real gain was thus only 63 per cent.
59
1
Cambridge Histories Online © Cambridge University Press, 2008
SOUTHERN AFRICA
segregation of Africans was adumbrated by the Natives Repre-
sentation Bill. This proposed to disfranchise African voters in the
Cape and outlined a separate structure of African representation.
For each province, one white senator would be chosen
by an
electoral college consisting of chiefs, headmen and other African
local authority officials; these colleges would also elect twelve
members of an advisory Natives Representative Council, which
would also include four nominated members and six officials.
The bills provoked concerted African criticism. In December
193 5
an All-African Convention at Bloemfontein brought together
delegates from numerous organisations. Congress members of all
shades of opinion were prominent; already Dube was disenchanted
with segregation and declared: 'The Government
is
trying
to
replace what
it
has already destroyed — our tribal system. '
34
Communists also took part; since the advent of Hitler, their party
line had swung round again to favour a broad popular front.
In
February 1936 Hertzog received
a
deputation from the A AC
headed by D. D.
T.
Jabavu, leader of the Cape Native Voters'
Convention. More persuasive, however, were
a
group
of
Cape
MPs whose support was needed to pass the bills. They induced
Hertzog to allow Africans removed from the Cape common roll
to elect instead three whites
to the
lower house. With this
alteration,
the
Native Bills obtained
the
necessary two-thirds
majority in April and became law.
Within parliament, the most powerful attack on the Natives
Representation Bill
in its
final form had come,
not
from
the
Nationalist opposition,
but
from
a
Cabinet minister,
J. H.
Hofmeyr (then
in
charge
of
education, the interior and public
health). More consistently than Smuts, Hofmeyr believed that
relations between black and white
in
South Africa should
be
guided
by
the principles
of
trusteeship proclaimed
by
colonial
regimes further north. Trusteeship was, of course,
a
cant word
freely used
for
political advantage;
it
was Hofmeyr's merit
to
point this out:
I have always regarded trusteeship as implying that at some stage or another,
the trustee is prepared to hand over the trust to his ward.
I
have yet to learn
that the European trustee in South Africa contemplates any such possibility.
And that being so,
I
find it very difficult
to
reconcile the use of the word in
34
S.
Marks, 'The ambiguities of dependence: John L. Dube
of
Natal',
Journal of
Southern African Studies, 1975,
1, 2,
179.
59
2
Cambridge Histories Online © Cambridge University Press, 2008