Denver Journal of International Law & Policy

Volume 5
Number 1 Spring Article 20
Symposium - International Business
Transactions- Tax and Non-Tax Aspects

January 1986

Judges in an Unjust Society: The Case of South Africa

Joe W. Pitts II

Follow this and additional works at: https://digitalcommons.du.edu/djilp

Recommended Citation
Joe W. Pitts, Judges in an Unjust Society: The Case of South Africa, 15 Denv. J. Int'l L. & Pol'y 49 (1986).

This Article is brought to you for free and open access by the University of Denver Sturm College of Law at Digital
Commons @ DU. It has been accepted for inclusion in Denver Journal of International Law & Policy by an
authorized editor of Digital Commons @ DU. For more information, please contact [email protected],dig-
[email protected].
 Judges in an Unjust Society: The Case of
South Africa
JOE W. ("CHIP") PITTS II*

INTRODUCTION

What should a judge do when confronted with the problem of apply-

ing unjust law? Grant Gilmore noted some of the options that the judge
has:
He can resign his judgeship. Or he can offer himself as a candi-
date for impeachment by saying: I regard this law as immoral and
refuse to enforce it in my court. Or he can evade the issue by seizing
on minor technical lapses (usually procedural) and dismissing the
case. Or he can enforce the law, with death in his heart-because it is
the law, duly established by the constituted authorities, and because,
as a judge, he has no other choice.'
The recurring dilemma is not merely an abstract problem of jurispru-
dence, though it is occasionally broached as such.2 It is also a practical
and immediate problem for judges, lawyers, legal officials, and (to some
degree) everyone else, particularly in legal systems like that of South Af-
rica. As Dante observed, "the charge of complicity does not discriminate:
the hottest places in hell are reserved for those who, in a period of moral
crisis, do nothing."
3
Although it has been called "the most unjust society in the world,"
one could debate whether South Africa's legal system is unjust. The
widely perceived tension between law and morality in that country leads

* Associate, Carrington, Coleman, Sloman & Blumenthal, Dallas, Texas; B.A. 1982 Tu-
lane University; J.D. 1985 Stanford University. The author would like to thank Bryan Gar-
ner, of Carrington, Coleman, and Professor Thomas Grey of the Stanford Law School for
their generous comments on this article.
1. G. GILMORE, THE AGES OF AMERICAN LAW 37-38 (1977).
2. See, e.g.., the Hart-Fuller exchange on law and morality- Hart, Positivism and the
Separation of Law and Morals, 71 H~Av. L. Rzv. 593, at 614-629 (1958); Fuller, Positivism
and Fidelity to Law - A Reply to Professor Hart, 71 HAiv. L. REv. 630, 659 (1958). See
also, the Hand-Frank exchange on the role of courts in wicked societies: L. Hand, The
Contribution of an Independent Judiciaryto Civilization, in THE SPIRIT oF LIEmwry, 164
(1953) (a society so riven that the spirit of moderation isgone, no court can save); Frank,
Some Reflections on Judge Learned Hand, 24 U. CH. L. REV. 667, 697-8 (1957) (courts
should do what they can). A recent formulaic description of the Critical Legal Studies move-
ment has noted its tendency to characterize U.S. law as fundamentally unjust and an instru-
ment of repression. TIME, Nov. 18, 1985, at 87. See, e.g., Freeman, Legitimizing Racial Dis-
crimination Through Antidiscrimination Law: A Critical Review of Supreme Court
Doctrine, 62 MINN. L. REv. 1049 (1978).
3. Barry, Injustice as Reciprocity, in JUSTcE at 75 (E. Kamenka & A. Soon Tay eds.
1979).
 DEN. J. INT'L L. & POL'Y VOL. 15:1

one to ask the larger question of whether an unjust law is, in a sense, no
law at all. Anti-egalitarian elements in some formulaic descriptions of jus-
tice might actually accord with South Africa's system: Anaximander's
stress on the importance of reconciling opposites in strife' does not pre-
clude the possibility of one opposite dominating the reconciliation. Aris-
totle's "distributive justice" based on social status may tolerate insensi-
tivity to those at the bottom of the social ladder. Justinian's concept of
"giving everyone his due" may be perverted if rights "due" are narrowly
defined. And Plato's vision of each person performing an appropriate so-
cial function can easily lead to complacency about inherent inequality.
Not surprisingly then, current thinkers tend to emphasize the concept of
"equality" as an essential component of justice.6 No effort is made in this
article to define justice or morality. Although undefined, they will serve
for the purposes of this article as somewhat utopian standards by which
man-made, often unjust laws, may be judged. In keeping with jurispru-
dential tradition on this subject, I merely assume that the man-made laws
of South Africa are largely unjust.'
Part I of this article contains an overview of South Africa's legal sys-
tem: its common law, legislation, adherence to parliamentary sovereignty,
and lack of judicial review. A comment on the nature and role of South
Africa's courts concludes this section. In Part II, the positivist and natu-
ral-law models of law and adjudication are considered for the (ultimately
inadequate) light they shed on the theoretical problem of the judge in an
unjust system. Part III describes how these theoretical inadequacies af-
fected the judicial responses of America under slavery and twentieth-cen-
tury South Africa under apartheid. The conclusion ventures some
thoughts toward a new theory of law and adjudication as interpretation.
This new approach could frame a more appropriate role for judges who
encounter unjust law, whether in South Africa or elsewhere.

I. THE SOUTH AFRICAN LEGAL SYSTEM

South Africa is a fascinating, if appalling society. The country

presents an incomparable laboratory for study of jurisprudential issues
because of the extremity of its social (largely racial) problems, and the
extremely divergent perspectives on those problems. The sociol order
both reflects and is shaped by the legal order: apartheid is appalling be-
cause it is the only institutionalized system of legal racial discrimination
in the world. To the extent that revolution is a real possibility, the health
and legitimacy of the legal system are implicated at the core. It has often

4. 1 PHILOSOPHIC CLASSICS 8 (W. Kaufmann 2d ed. 1968).

5. See, e.g., Dworkin, Why Liberals Should Believe in Equality, N.Y. REv. BOOKS, Feb.
3, 1983, at 32; What Liberalism Isn't, N.Y. REV. BOOKS; June 20, 1983, at 47; What is
Equality? (Pts. 1 & 2), 10 PHIL. & PUB. AFF. 185, 283, 314 (1981); J. RAWLS, A THEORY OF
JUSTICE 73-74 (1971).
6. E.g., R. DWORKIN, TAKING RIGHTS SERIOUSLY 326-7, 343 (1977) (though South Africa
and Nazi Germany are examples, the idea of a wicked legal system is assumed).
1986 JUDGES IN AN UNJUST SOCIETY

been often said that the common law of South Africa recognizes most
basic individual liberties and the concept of equality before the law. Yet
the plethora of repressive and racially discriminatory statutes in South
Africa calls into question the viability of liberal precedent and tradition.
In practice, courts are extremely circumscribed by parliamentary sover-
eignty and judicial review. These features have resulted in a legal system
with a very poor image, especially among South African blacks. The fol-
lowing is a brief overview of the South African legal system.

A. Common Law
The common law of South Africa is predominantly Roman-Dutch law
transplanted from Holland in 1652 when the settlers of the Dutch East
India Company arrived at the Cape of Good Hope. British rule (especially
from 1806) resulted in substantial influence, if not direct imposition, if
English common law. Not only were the courts and the English law of
criminal evidence and procedure introduced in 1827-1834, but also the
mandatory training of advocates in England combined with the appoint-
ment of English judges to yield constant recourse to English precedent
and modes of legal thought. Thus, Hahlo and Kahn have written that the
old "two layer cake" of Roman-Dutch law has now acquired a "third
layer," English law. 7 While English common law has a tremendous his-
toric "status" component, the tradition of formal (especially procedural)
equality before the law has always been potent. In theory, therefore,
South African judges could appeal to a precedent like Somerset's Cases
as compelling persuasive authority. The thrust of Lord Mansfield's 1722
holding, which freed the Virginia slave Somerset who deserted his master
in England, was that slavery was odious and against natural law. After
the Nationalist Party came to power in 1948, legal sentiment for a purifi-
cation of South African law gradually built up, culminating in Appellate
Division Chief Justice L.C. Steyn's repudiation of "unjustified reliance"
on English sources.' Unfortunately, the largely Afrikaans "purists" were
notoriously insensitive to civil liberties. 0 Greater sensitivity has normally
been shown by a group of largely English lawyers who see a veritable
charter of human rights at the core of Roman-Dutch law. Lawyers and
academics like Anthony Mathews, John Dugard, and Sydney Kentridge,
for example, clearly see liberal rules, principles, and canons of construc-
tion in the Roman-Dutch common law. At times, perhaps, their enthusi-
asm has approached overstatement. 1' Yet, these distinguished and pas-

7. H.R. HAHLO & E. KAHN, THE SOUTH AFRICAN LEGAL SYSTEM AND ITS BACKGROUND,
584 (1968).
8. Lofft, 1 Lofft's Rep. 1; 20 Howell's State Trials 1; 98 Eng. Rep. 499 (1722).
9. See, e.g., Regal v. African Superslate (Pty.) Ltd., 1963 (1) S.A. 102 (A); Trust Bank
van Afrika Bpk. v. Eksteen, 1964 (3) S.A. 402 (A).
10. Blerk, The Irony of Labels, 99 S. AFR. L.J. 365, 377 (1982).
11. E.g., "The South African common law is color blind... [c]onsequently courts have
not discriminated between white and black in the allocation of civil rights." J. DUGARD,
HUMAN RIGHTS AND THE SOUTH AFRICAN LEGAL ORDER, 71-2, 382-3, 393 (1978) (hereinafter
 DEN. J. INT'L L. & POL'Y VOL. 15:1

sionately committed lawyers are under no illusions about the repressive

character of South Africa's legal system; they merely place the blame on
statutes in derogation of the common law. They have thus adopted the
"liberal" model of South African common law for clearly reformist
reasons.
Nevertheless, an issue-partly empirical, partly theoretical-remains
as to the nature of South African common law. Can it be optimistically
characterized as containing liberal principles? Or does the tincture of re-
alism force one to see the heart of darkness in the tradition? As will be-
come evident, the issue is crucial both to the characterization of the legal
order as just or unjust and to consideration of the appropriate judicial
response. Even after the empirical question is answered by recognition
that South African common law is a mixed bag of "progressive" and "re-
gressive" decisions, the judge is faced with the theoretical question of the
scope for creativity and law-making in interpretation of the common law.

B. Legislation

In contemporary South Africa, the significance of the debate regard-

ing the liberality of the common law fades in light of the many repressive
and racially discriminatory statutes in derogation of the common law. To-
day, there are so many discriminatory statutes on the books that it would
be misleading to ignore the fact that South African law is primarily di-
rected against blacks. It has not always been quite so bad: reforms under
British rule elevated the ideal of "the rule of law" and the notion of at
least formal equality before the law. "Coloreds" were recognized as for-
mally equal in 1828, and all slaves were freed in 1834. These liberaliza-
tions were largely responsible for the "Great Trek" of Afrikaaners inland
from the Cape in 1835-1837. Yet black Africans have never really been
seen as persons with full civil and political rights, and other blacks (in-
cluding coloreds and Indians) 12 have only been seen thus far as second-
class citizens at best. The following account of the highlights of apartheid
legislation does not aim at the comprehensiveness and specificity at-
tempted elsewhere,'8 but briefly introduces some of the main statutes re-
lied on by the government.

cited as DUGARD, HUMAN RIGHTS); see also, A.S. MATHEWS, LAW, ORDER AND LIBERTY IN
SOUTH AFRICA (1972); Mathews & Albino, The Permanenceof the Temporary - An Exami-
nation of the 90- and 180- Day Detention Laws, 83 S. AFR.L.J. 16, 37-8 (1966); Kentridge,
The Theories and Realities of the Protection of Human Rights Under South African Law,
56 TUL. L. REv. 227, 229-31 (1981); Hahlo & Maisels, The Rule of Law in South Africa, 52
VA. L. REv. 1, 13 (1966); Centlivres, The South African Constitution and the Rule of Law,
1956 BUrrERWORTH'S S. AFR. L. REV. 3, at 12.
12. "Blacks" throughout most of this paper will be used in this expansive sense urged
by those fighting for equal civil rights in South Africa. When a more restrictive reference is
needed, black Africans will be referred to as such.
13. Most admirably in DUGARD, HUMAN RIGHTS, supra note 11, at 53-201; Landis, South
African Apartheid Legislation, 71 YALE L.J. 437 (1961).
1986 JUDGES IN AN UNJUST SOCIETY

1. Before 1948

Before 1948, when the Nationalists came to power, a number of stat-

utes reflected the racist attitudes that fed the stepped-up campaign of the
Nationalists to separate the races in 1948.' 4 Among these early segrega-
tion laws was the Immorality Act.' 5 This statute allowed the government
to intervene in the most personal areas of life by prohibiting interracial
intercourse. Section 16 of the expanded 1957 version of this Act was re-
pealed by the Nationalists in 1985, along with the Prohibition of Mixed
Marriages Act of 1949.16 The repeal of these Acts had been expected for
some time, both because of the difficult and sordid investigation needed
for enforcement and the recognition that black women are much more in
need of protection from white men than are white women from black
men. Thus, the Badenhorst Select Committee of 1984 decided that repeal
of the anachronistic laws could allow the South African government some
relief from international pressure without threatening the remaining leg-
islative structure of apartheid.
Other pre-1948 restrictive laws were the "pass laws" which governed
the movement of Africans within South Africa, and laws reserving jobs
for whites and prohibiting blacks from joining recognized trade unions.
These two areas, influx control and labor law, are two chief areas of legis-
lative activity by the Nationalists. The Bantu Land Act 27 of 1913 and
the Bantu Trust and Land Act 18 of 1936 set aside approximately 87
percent of the country's most desirable territory for whites, leaving 13
percent of the most arid and undesirable land for Africans. This grossly
inequitable distribution is exacerbated by the fact that the population
ratio of whites and blacks is almost inversely proportional to the amounts
of land received through these early Acts: the 23 million blacks make up
about 75 percent of the country's population.

2. Post-1948

Since 1948, the process of racial discrimination has intensified and

been systematized and institutionalized by the Nationalists. The corner-
stone Population Registration Act 30 of 1950 requires classification and
registration of the entire South African population on the basis of race.
Every person must fit into one of four racial categories: White, Indian,
Colored, or African. The "separate but equal" approach that was rejected
in the United States by Brown v. Board of Education17 was ironically

14. These laws are fully described in Suzman, Race Classificationand Definition in the
Legislation of the Unions of South Africa, 1960 ACTA JURIDICA 339.
15. Act 5 of 1927.
16. The Immorality Act of 1927 was expanded by Act 23 of 1957; the Prohibition of
Mixed Marriages Act was Act 55 of 1949. Repeal of these two Acts was announced by South
African Home Affairs Minister F.W. de Klerk on April 15, 1985, effective as of June 1, 1985.
See also, New York Times, Tues., April 16, 1985, at 1, col. 6.
17. 347 U.S. 483 (1954).
 DEN. J. INT'L L. & POL'Y VOL. 15:1

taken a step further by the South African government at the same time
Brown was being argued. Under the Reservation of Separate Amenities
Act of 19538 the implicit contemplation of blacks as separate but une-
qual was explicitly recognized in the realm of public facilities. Though
much petty apartheid has been eliminated, much persists in the provision
of separate restrooms, train cars, entrances, and so on. Separate and in-
ferior educational facilities have existed for blacks since the Extension of
University Education Act.' 9
Influx control was tightened up under the Group Areas Act 20 and the
Urban Areas Act,21 which together classified geographic and living areas
as white or non-white. These Acts allowed ownership and occupation in
certain areas only by designated racial groups. The pass laws governing
movement between racial areas were consolidated and expanded under
the inaptly named Abolition of Passes and Coordination of Documents
Act. 2 Despite the recent repeal of the pass laws, the legislative require-
ment that Africans apply for new identity documents means that the vast
majority of Africans must continue to carry the much-hated reference
books that identify them by name and race, where they have worked,
what taxes they have paid, and whether they have permission to be in the
white areas.
Influx control legislation also formed the basis for the government's
denationalization strategy, which involves giving the "homelands" set
aside by the early Land Acts mentioned above gradual "independence" as
separate sovereign states. This Kafkaesque political fantasy of the South
African government has not been shared by the other nations of the
world, none of whom have extended recognition to the fictionally "inde-
pendent" states thus created. But the real impact of the government's
fantasy is severe. Because the government has been striving until recently
toward creation of a "white" South Africa without a single black African
citizen, 8 the security of Africans (who have been considered aliens in the
land of their birth) has been fragile indeed. The result has been the trag-
edy of state-ordered forced removals of blacks from white areas to the
homelands, where jobs and housing are usually unavailable and starvation
24
is rampant.

18. Act 49 of 1953. On October 1, 1986, President P.W. Botha confirmed the long-
awaited intent to repeal this anachronistic legislation. See, Wall St. J., Oct. 8, 1986, at 32,
col. 3.
19. Act 45 of 1959.
20. Act 41 of 1950, consolidated through Act 77 of 1957 and Act 36 of 1966.
21. Act 25 of 1945, § 10 inserted by § 27 of the Black Laws Amendment Act 54 of 1952.
The government announced repeal of the Urban Areas Act, the pass laws and related influx
control legislation in April, 1986, effective as of July 1, 1986. The Abolition of Influx Control
Act 73 of 1986.
22. Act 67 of 1952.
23. Dr. C.P. Mulder, then Minister of Information, stated this government goal in
House of Assembly Debates, col. 579 (Feb. 7, 1978).
24. For an overview of the whole process, see Dugard, South Africa's "Independent"
1986 JUDGES IN AN UNJUST SOCIETY

Labor law was not immune from Nationalist legislative activity, and
job reservation for whites was expanded as a total prohibition on the
right of Africans to strike was imposed.25 A tension between labor law and
influx control has existed historically, primarily because whites want
blacks in white areas when cheap labor is needed, but not when manage-
ment purposes are not being served.2 6Those conflicting demends persist
today, despite the formal repeal of much of the influx control legislation.
Security legislation is so arbitrary and repressive that it is frequently
cited as illustrating the erosion or collapse of the rule of law in South
Africa. 7 As discrimination intensified after 1948, so did black resistance
and then further repression. This cycle of repression was initially justified
by cold war rhetoric invoking fears of a communist onslaught, and the
early statutes in conception and application associated black liberation
movements with communism. The Suppression of Communism Act,2 now
known as the Internal Security Act, was originally directed at communist
subversion. Because the Act included within the definition of communism
any doctrine "which aims at bringing about any political, industrial, so-
cial or economic change, ' 29 the Act was vigorously enforced against
noncommunist black and white activists.
In 1976, the legitimacy of proceedings against even anticommunist
subversives was recognized by amending the name of the Act and ex-
tending its coverage to any organizations or individuals who "endanger
the security of the state. '3 0 The Internal Security Act allows severe intru-
sions on freedoms of speech, movement, assembly, and association, usu-
ally without any recourse to judicial review. Moreover, the recent trend is
for the sphere of unchecked executive discretion in the security area to
increase rather than decrease. The current version of the Act allows de-
tention without trial and "banning" of persons deemed to be security
risks.8"
Following the Sharpville massacre of 1960, security legislation was
introduced that banned the chief black liberation organizations (the Afri-
can National Congress and the Pan-Africanist Congress), and declared a

Homelands: An Exercise in Denationalization,10 DEN. J. INT'L. L. & POL'Y 11 (1980).

25. E.g., Black Building Workers Act 27 of 1951; Industrial Conciliation Act 28 of 1956,
77 (job reservation); Black Labor (settlement of disputes) Act 48 of 1953 (prohibition on
strikes).
26. See Budlender, Incorporationand Exclusion: Recent Development in Labor Law
and Influx Control, S. AFR. J. HuM. RTs. 3 (1985).
27. Suzman, South Africa and the Rule of Law, 85 S. APR. L.J. 261 (1968); Mathews,
The South African Judiciary and the Security System, 1 S. AFR. J. HUM. RTs. 199, 200
(1985). For an excellent overview of security legislation in general, see DUGARD, HUMAN
RIGHTS, supra note 11, at 146-365.
28. Act 44 of 1950.
29. Id. at § 1.
30. Internal Security Amendment Act 79 of 1976, inserting § 2(2)(A) of Act 44 of 1950.
31. The Internal Security Act 74 of 1982, §§ 18-29.
 DEN. J. INT'L L. & POL'Y VOL. 15:1

state of emergency. 2 The South African government's reaction to the

most recent spate of unrest has also included declarations of states of
emergency."3 After attempts to amend other security laws, the govern-
ment finally introduced the sweeping detention-without-trial provisions
of The Terrorism Act.3 4 This Act allows a person to be indefinitely de-
tained in solitary confinement, without trial, for purposes of interroga-
tion. Courts are denied ability to review the validity of the detention.35

3. Recent Reforms

Much attention has been given to recent "reforms" of the South Afri-
can government. These include the 1979 amendments to South African
labor law resulting from the recommendations of the Wiehan Commis-
sion,3 6 the recent pronouncements accepting the permanency of urban
blacks and the "restoration of South Africian citizenship", the new con-
stitution adopted in 1984, and the repeal of the Prohibition of Political
Interference Act,37 the Mixed Marriage Act, s and section sixteen of the
Immorality Act and the pass laws." The government also has committees
"examining" other discriminatory legislation to determine whether repeal
or "adaptation" is possible or necessary.
The 1979 amendments to the Industrial Conciliation Act 40 did open
a new era in South African labor law by allowing the registration of black
unions and an industrial court with an equitable jurisdiction to consider
unfair labor practices. Yet strikes are still "extremely circumscribed,"
meetings and picket lines are outlawed, and union leaders are often de-
tained or banned.4 1 Recent legislation, in fact, has expanded the ability of
the state to proceed against the emergent unions by prohibiting their
ability to affiliate or assist political parties, or to "influence" members
regarding political positions. 4' The recently formed Congress of South Af-

32. Unlawful Organizations Act 34 of 1960.

33. E.g., N. Y. Times, July 21, 1985, at 1, col. 6; N. Y. Times, Oct. 26, 1985, at 1, col. 3;
The latest state of emergency was imposed on June 12, 1986, by Proclamation R. 109 on
President Botha.
34. Section 6 of Act 83 of 1967.
35. Id. at 6 (5).
36. Commission of Inquiry into Labor Legislation, RP 47 (1979).
37. Act of 1968; repeal announced May 25, 1985. N. Y. Times, May 26, 1985, § 1, pt. 1,
at 3, col. 4.
38. Act 55 of 1949; repeal announced May 25, 1985. N. Y. Times, May 26, 1985, § 1, pt.
1, at 3, col. 4.
39. Act 5 of 1927, expanded by Act 23 of 1957; repeal of 16 announced April 15, 1985.
N. Y. Times, Apr. 16, 1985, at 1, col. 6. On repeal of the pass laws, see supra note 21.
40. Act 94 of 1979.
41. Gould, Black Unions in South Africa: Labor Law Reform and Apartheid, 17 STAN.
J. INT'L L. 99, 147 (1981).
42. Labor Relations Amendment Act 57 of 1981, § 8(6) and § (7), read with §
4(5)(a)(iii). See also, The Labor Relations Amendment Act 81 of 1984, § 3, inserting § 31(a)
in the principal act, and providing for non-enforcement of union-employer agreements
which do not follow the requirements of the old legislation.
1986 JUDGES IN AN UNJUST SOCIETY

rican Trade Unions (COSATU) is thus risking prosecution for its strong
political stands. The decisions of the industrial court, though initially
greeted with enthusiasm, took a more restrictive turn in 1984-85. 4 1 In
light of the recent unrest in South Africa, which is inevitably tied to labor
and economic grievances of urban blacks, one can only conclude that the
Wiehan strategy of transferring industrial unrest from the streets and
shop floors to the formal statutory dispute-resolution machinery has
failed.
Reform of influx control laws was considered in the Riekert report of 4
1979, 44 and the Act itself was repealed with the pass laws in 1986. 5
Amendments to the Urban Areas Act were made in July 1985. Yet the
new hope for reform which emerged in late 1985 and early 1986 when
President Botha announced that the policy of stripping black Africans of
South African citizenship would be discontinued has been dashed by the
form of the legislation providing for issuance of new identity documents
and "restoring" South African citizenship. 4' The new "identity docu-
ments" will serve many of the same purposes as the old "reference
books," and will not be issued unless an application is made. Similiarly,
citizenship will be restored only to those who go through the cumbersome
process of administrative application, a process effectively unavailable to
those who are ignorant of their right to apply. Although difficult to un-
derstand, the Restoration of Citizenship Act seems to apply only to "ur-
ban" blacks - those "lawfully and permanently resident" in the "white"
South Africa that remains after certain homelands were granted indepen-
dent state status. The meaning of "permanent residence" in the Act is
unclear, but seems to refer to the control concept of the new-repealed
Urban Areas Act. This indicates that the government has simply insti-
tuted a new sort of influx control, with passports taking the place of
passes.' 7 Significantly, the legislative structure of homelands as "separate
states," which developed from influx control concepts of "separate areas,"
remains a core principle of apartheid.48 Riekerts recommendations have

43. See, e.g., Nat'l Union of Mineworkers & Others v. Driefontein Consolidated, Ltd., 5
INDUS. L.J. 107 (1984) (worker-protective test governing status quo orders rejected); Vetsak
v. Peter Ngobeni and Others, 5 INDus. L.J. 205 (1984) (contractual theory allows mass dis-
missal and discriminatory selective reinstatement).
44. P. J. Riekert, Commission of Inquiry into Legislation Affecting the Utilization of
Manpower, RP 47 (1979) (hereinafter cited as Riekert Report).
45. Section 10 of Act 25 of 1945 (amendments decreasing the time period that foreign
workers must remain in urban areas to qualify for residence rights, from 15 years to 10
years, and allowing residence in multiple towns to meet the requirements); effective July 3,
1985; Acat repealed in its entirety on July 1, 1986. See supra note 21.
46. See Identification Act of 1986, and the Restoration of South Africian Citizenship
Act of 1986.
47. Budlender, Influx Contorl in the Western Cape: From Pass Laws to Passports
(Aug. 21, 1986) (unpublished manuscript on file at the Denver Journalof InternationalLaw
& Policy.)
48. N. Y. Times, Oct. 1, 1985, at A6, col. 3. Four homelands: the Transkei; Bophuthat-
swana; Venda; and the Ciskei have taken independence, with Kwa Ndebele was expected to
 DEN. J. INT'L L. & POL'Y VOL. 15:1

led to restrictions on migrant workers, including those from the suppos-

edly "independent" homelands. Increased fines on employers found to
employ illegal migrants have resulted in thousands of newly unemployed
migrants, with repercussions extending to the migrant laborers' depen-
dents in the homelands.
The new Constitution adopted by South Africa in 1984 is in keeping
with the shift in ideology from baaskap (white domination) to separate
development. The new Constitution accepts Indians and coloreds as wor-
thy of representation in racially segregated houses of Parliament, but Af-
rican blacks are still wholly excluded from national political participation.
They do not possess the right to vote or any other form of representation.
Even the coloreds and Indians have only illusory power, for they are a
minority in "white" South Africa, and their consideration of matters is
subject to the discretion of the Executive State President. Today the Ex-
ecutive State President has vastly expanded powers, and the white house
of Parliament retains a veto over legislation passed by the other two
houses. Cosmetic reforms and anti-discriminatory rhetoric may be toler-
ated in "white" South Africa, because Indians and coloreds are seen as a
new, co-opted black middle class. In short, because the government's
terms of reference have changed, the labor reforms and legislative repeals
are utterly meaningless for the black majority of South Africa. Such "re-
forms" merely give the Nationalists an opportunity to gain a public rela-
tions advantage while bringing some of the most overtly discriminating
legislation in line with current realities."9 The essential legislative struc-
ture of apartheid, justified on grounds of separate development, remains
intact.

C. ParliamentarySovereignty

The repressive legislation of apartheid exists because the South Afri-

can Parliament, as in England, is supreme. Also as in England, party dis-
cipline is strong, so the Nationalists have little trouble implementing
their legislative program. Parliamentary sovereignty on the Blackstone-
Dicey English model dates back to British occupation of the Cape in
1806. It was the form of government in the Republic of South Africa (the
Transvaal), and has been rather uncritically accepted since South Africa
became a Union in 1910. Federalism, which might have resulted in in-
creased protection for human rights, was rejected. When South Africa be-
came a formal "Republic" in 1961, the Westminster style of parliamen-
tary sovereignty continued to give great play to the powers of a strong
cabinet. At the time the American Constitution was drafted, the idea of a
fundamental law binding Parliament was potent. But Dugard has written

do so in 1986; but popular protest led the legislative assembly to vote in late August, 1986,
to delay independence indefinitely.
49. As Rep. Stephen Solarz wrote in a recent New York Times editorial, whites see the
changes as cosmic, but blacks feel they are merely cosmetic. N. Y. Times, Oct. 8, 1985, at
A31, col. 1.
1986 JUDGES IN AN UNJUST SOCIETY

that by the time "of English influence on South African law, the concept
of parliamentary sovereignty was unchallenged wherever the Union Jack
flew. South African lawyers and politicians grew up under a totally
50
differ-
ent English tradition from that of their American counterparts.
The South African version of parliamentary sovereignty differs not
only from the American form of government; it also drastically differs
from its British parent. Unlike the British emphasis on the notion of the
rule of law, with its values of equality before the law and freedom from
arbitrary government action, any idea in South Africa of a Parliament
bound by law has evaporated. 5' The basic premise of representation in a
democratically elected legislature is denied. There is no effective legal op-
position in South Africa. Additionally, the concept of Parliament as ex-
isting to protect common-law rights against encroachment by the execu-
tive is foreign to South Africa. The result is "a poor imitation of
Westminster," a sovereign Parliament stripped "of all conventional re-
straints founded upon the historically based destiny and representative
composition of its English counterpart." 2 Dugard concludes that loyalty
to the principle of parliamentary sovereignty, "more than any other legal
factor, brought about the debasement of the South Africa legal system.""
The new tricameral Parliament does not change matters. One party,
the Nationalist party, continues to monopolize effective power. The Exec-
utive State president has vastly expanded powers, including the power to
determine which matters Parliament's houses will consider. The State
President also has final say over whether he wants the advice of the Pres-
ident's Council, made up from Parliament's members, at all.
I Naturally, a sovereign Parliament that considers itself above the law
has little patience for an institution like judicial review. Challenges to the
restrictive legislation sketched above become difficult at best, as Judge
Didcott of Durban has remarked: "Parliament has the power to pass the
statutes it likes, and there is nothing the courts can do about that. The
result is law. But that is not always the same thing as justice. The only
way that Parliament can ever make legislation just is by making just
legislation.""

D. The TraditionalLack of Judicial Review

fcp]Parliamentary (and increasingly, executive) sovereignty thus implies
a very limited scope for judicial review in South Africa. Again, it was not
always this way; South African judges were overwhelmingly selected from

50. DUGARD, HUMAN RIGHTS, supra note 11, at 16.

51. Van der Vyver, ParliamentarySovereignty, Fundamental Freedoms and a Bill of
Rights, 99 S. AFR. L.J. 557, 572 (1982) (citing the constitutional crisis over the colored vote
in the 1950's as evidence that Parliament assumed it was above the law).
52. Id. at 558.
53. DUGARD, HUMAN RIGHTS, supra note 11, at 36.
54. In re Dube, 1979(3) S. A. 820, 821 (N); See also, S. v. Adams, 1979(4) S.A. 793, 801
(T)(King, J.).
 DEN. J. INT'L L. & POL'Y VOL. 15:1

advocates who had been trained in England, so they were at least aware
8
of Bonham's Case,5 containing the dictum in which Chief Justice Ed-
ward Coke asserted the right to judicial review. But Bonham's Case had
even less success in South Africa than it did in England.
In the constitutional crisis of 1897, Chief Justice Kotze in the Orange
Free State asserted a right of judicial review. President Kruger (of
Krugeraand and Kruger Park fame) reacted by pushing through legisla-
tion denying any such right. Kotze was dismissed from office. Swearing in
the successor Chief Justice (Gregorowski), Kruger said that "the testing
right is a principle of the Devil." 6 Kruger's statement reverberates
throughout contemporary South Africa; a relatively impotent judiciary
can be granted some independence at no risk to the supremacy of a pow-
erful and unrepresentative legislature.
Yet South African judges did not give up. A second constitutional
crisis in the 1950's was tied to the most substantial object of protection of
judicial review: the constitutionally entrenched 5 7 colored right to vote. In
the original compromise that resulted in the Union of South Africa's four
provinces (the Cape, Natal, Transvaal, and the Orange Free State), the
Cape was allowed to retain its qualified franchise for coloreds. The Na-
tionalists passed the Separate Representation of Voters Act in 1951 to
remove the colored right to vote. But the entrenched section 35 of the
Constitution required the procedure of a two-thirds vote of both houses
sitting together to remove the entrenched right. Exercising a form of judi-
cial review, the Appellate Division in Harris v. Minister of the Interior"
struck down the Act because the constitutional procedure had not been
followed. The court thus overruled Ndlawana v. Hofmeyr, which had up-
held parliamentary sovereignty.59 Parliament responded with the "High
Court of Parliament" Act which provided for legislative review of any
judicial decision purporting to invalidate an act of Parliament. That Act
was invalidated in Minister of the Interior v. Harris,60 on the grounds
that the high court was really no court at all, but a sham for Parliament
itself.
The battle was not over yet. Parliament responded with a court
packing plan that would have been the envy of Roosevelt. Under the Ap-
pellate Division Quorum Act,61 eleven judges were substituted for the pre-
viously existing five when an act of Parliament was in issue. The Senate
Act,62 which followed, almost doubled the size of the Senate and changed

55. Bonham v. Atkins, 8 Co. Rep. 1136, 118a, 77 Eng. Rep. 646, 652 (1610).
56. DUGARD, HumAN RIGrrs, supra note 11, at 24.
57. An "entrenched" provision of the South African Constitution is one that requires a
two-thirds vote of Parliament to be disregarded.
58. 1952 (2) S.A. 428 (A.D.).
59. 1937 A.D. 229.
60. 1952 (4) S.A. 769 (A.D.).
61. Act 27 of 1955.
62. Act 53 of 1955.
1986 JUDGES IN AN UNJUST SOCIETY

the mode of electing senators to favor the Nationalist party. This legisla-
tive plan was upheld by the newly renovated Appellate Division on the
grounds that neither Act, considered alone, was invalid. Mr. Justice 0. D.
Schreiner was the sole dissenter, noting the improper and unconstitu-
tional purpose and effect of the legislation. The South Africa Act Amend-
ment Act6 s was then passed by a two-thirds majority, to give retrospective
validity to legislative denial of the vote to Cape coloreds.This Act also
expressly denied the power of judicial review over the validity of acts of
Parliament (except over the entrenched provisions guaranteeing equality
of the two official languages, English and Afrikaans), a provision that was
repeated in the constitution of 1961.4 The new Constitution of 1983 has a
comparable provision which provides for review, "subject to" an intricate
set of cross- and cross-cross-referenced provisions that give only vitiated
review. The Constitution further provides that, except for that limited
review, "no court of law shall be competent to inquire into or pronounce
upon the validity of an Act of Parliament." 5 The limited review allowed
divisions of the Supreme Court relates to whether the state President has
complied with certain procedures in deciding whether a specific matter is
a general affair or an "own" affair of one of the represented population
groups.66 This is a weak review of what is a weak requirement of formal
consultation with Parliament before certifying matters. Moreover, the
legislative history of the new Constitution indicates that expanded pow-
ers of judicial review were not intended by the government. 67 Thus, the
lack of an explicit reference to any entrenched clauses of the new Consti-
tution may be read by analogy to the 1961 Constitution as limiting review
to the entrenched amendment procedures (two-thirds of each house) re-
quired for derogating the equality of the two official languages," or to the
less entrenched and less stringent amendment procedures (a simple ma-
69
jority vote of each house) for many other provisions.
A strong argument clearly exists, however, for an expansive reading
of the new Constitution to allow review even of nonentrenched provisions.
The omission of the previous reference to entrenched sections may be
read with the general grant of review to trump any restrictive reading
suggested by legislative history. The issue would then arise whether the
power of review would be limited (as before) to procedural sections or
would go to substantive provisions. And, unfortunately, the tough issue of
the proper external staridards for reviewing the constitutionality of legis-
lation would also inevitably arise.

63. Act 9 of 1956.

64. REPUBLIC OF SOUTH AFRICA CONST. Act 32 of 1961, § 59 (2) (no judicial review);
§ 108 (language rights).
65. REPUBLIC OF SOUTH AFRICA CONST. Act 110 of 1983, 34 § 34(2).
66. Id. at § 18, referring to subsections 17(2) and 16(1).
67. House of Assembly Debates, Aug. 29 - Sept. 2, 1983, cols. 12891, 12892.
68. Act 110 of 1983, § 99(2), referring to § 89.
69. Id. at § 99 (3). These provisions notably include many clauses affecting the powers
of the State President and the Houses of Parliament, as well as the franchise.
 DEN. J. INT'L L. & POL'Y VOL. 15:1

The problem of standards is particularly acute in a country with no

bill of rights, a Constitution viewed primarily as a mere tool for structur-
ing government, and a history of repressive legislation. Positive law has
granted few rights to those most in need of rights, and natural-law stan-
dards of justice, morality, or general reasonableness have never been suc-
cessfully invoked for substantive, as opposed to procedural, protection.
Other standards for quasi-natural-law review, such as humanitarian val-
ues, consensus, policy, "the rule of law," utilitarianism, or international
law values are possibilities, but only possibilities. These will be discussed
in more detail in Part II below. Suffice it to say that all sides agree on the
characterization of South African courts as positivistic.
The materials for asserting a new and expanded right of judicial re-
view thus exist, and should be used by the South African courts. Other-
wise, the judiciary will be limited to the weak power of inherent common-
law review of subordinate legislation and administrative action. This type
of review does not extend to consideration of the substantive merits of
legislation. Subordinate legislation includes the ordinances, regulations
and the like issued by local government organizations with delegated au-
thority from Parliament. The limited review on certain grounds over ad-
ministrative actions is primarily to determine whether regular procedural
requirements, such as impartiality, proper attention applied to the mat-
ter, and at least a good-faith effort at fairness, have been met. Yet even
unreasonableness per se is not regarded as an irregularity. Only "grossly
unreasonable" decisions amounting to a "recognized irregularity" warrant
court intervention. Until the recent decision in Everett v. Minister of
the Interior,7 it was assumed that an opportunity to be heard had to be
granted. Limited use of the statutory presumption in favor of liberty has,
72
however, been employed to good effect, if only rarely.
Thus, the scope for judicial review had traditionally been circum-
scribed by rigid parliamentary sovereignty. The situation is not improved
by jurisdiction-curtailing measures" such as that under the Prohibition of
Interdicts Act,74 which prohibits interdicts against banishment from a
white area pending resolution of an attack on such an order; the Reserva-
tion of Separate Amenities Act," which ousts court jurisdiction to review
the validity of reservations; and the Terrorism Act,76 which effectively ex-
cludes court review of any matter other than the rare "illegal" police in-

70. Nat'l Transport Comm. v. Chetty's Motor Transport, 1972 (3) S.A. 727, 735, 727
(A.D.); Goldberg and Others v. Minister of Persons, 1979 (1) S.A. 14 (A.D.) at 38.
71. 1981 (2) C.P.D. B 1 453.
72. See, Dadoo Ltd. v. Krugersdorp Mun. Council, 1920 A.D. 530, 552.
73. On the American experience with jurisdiction stripping measures, see Gunther,
Congressional Power to CurtailFederal Court Jurisdiction:An Opinionated Guide to the
Ongoing Debate, 26 STAN. L. REv. 895 (1984).
74. Act 64 of 1956 (prohibiting issuance of interdicts in the influx control area).
75. Act 49 of 1953, expected to be repealed in 1987.; See also supra note 18.
76. Act 83 of 1967, § 6(5). See also the exclusion of court review of bannings and deten-
tions under the Internal Security Act 74 of 1982, § 29.
1986 JUDGES IN AN UNJUST SOCIETY

terrogation that comes to light. The limited common-law power of review-

ing delegated authority to determine whether it is unreasonable per se or
ultra vires is not very helpful. And even a court that successfully grasps
the opportunity under the new Constitution to assert an expanded power
of judicial review will not be able to reach injustices perpetrated against
the majority of African blacks, who have not been legally considered citi-
zens of South Africa subject to South African legislation. The deportation
of these denationalized aliens, is, of course, excluded from judicial
review.7 7

E. The Courts in South Africa

The lack of a bill of rights, coupled with the relative lack of judicial
review, makes it all the more important that the South African judiciary
be independent from the other branches of government. South African
courts have been glowingly applauded on all sides for their unassailable
independence. 8 Yet increasing criticism of the judiciary has cautiously
called that independence into question. The caution springs from the
strategical consideration that the myth of independence has a liberal ef-
fect: the courts in South Africa are, after all, the only legal institution in
which black Africans have anything approaching effective political power.
Thus, memories of the courageous stand of the Appellate Division in the
constitutional crisis of the 1950's linger and the fiction of political neu-
trality is maintained, to preserve the "best chance" that blacks have. Un-
derlying much of the new criticism is the perception that the myth of an
independent judiciary legitimates the unjust legal order by indicating
that the legal system provides free and equal individuals access to
courts. 79 Respect for the system is misplaced, for only whites are really
considered free and equal under South African law. The illusion of a flex-
ible and just procedural system that occasionally does justice obscures the
substantive lack of justice in most cases.
Increasingly, the South African press notes the grossly disparate sen-
tencing in cases involving blacks and whites. Lawyers and academics like
Sydney Kentridge and John Dugard write of the occasional sensitivity to
human rights, but conclude that the courts have at best a mixed record in
race and security cases. 80 Aloofness in race relations, says Dugard, is "but
one of the many judicial myths of South Africa."8 "

77. Admission of Persons to the Republic Regulation Act 59 of 1972, § 45.

78. See, e.g., Hahlo & Maisels, supra note 11, at 10-12; Final Rep't of the Constit.
Comm. of the Pres. Council on the Adaptation of Constitutional Structures in South Africa,
P.C. 4 (1984), at 115-117 (hereinafter Final Rep't).
79. See generally, J. HABERMAs, LEGITIMATION CRISIS, Pt. 3 (1973).
80. See Kentridge, supra note 11 at 231; Kentridge, The Pathology of a Legal System:
CriminalJustice in South Africa 128 U. PA. L. Rav. 603, 615 (1980); Kentridge, Telling the
Truth About Law, 99 S. AFR. L.J. 648, 649 (1982); DUGARD, HUMAN RIGHTS, supra note 11 at
387.
81. DUGARD, HUMAN RIGHTS, supra note 11, at 324.
 DEN. J. INT'L L. & POL'Y VOL. 15:1

Why is this so? The primary reason must be that the judges are ex-
clusively and uniformly white, and thus inevitably subject to all of the
contradictory fears and prejudices of white South Africa. In addition,
though, judges have been (and will continue to be under the new Consti-
tution)8s appointed by the Executive President. Proposals for a multira-
cial appointments commission were considered, but rejected by the con-
stitutional committee. Appointments are traditionally from the senior
members of the bar (advocates as opposed to attorneys). Political consid-
erations nonetheless play a huge role, and the security of tenure granted
the judges, who need not retire until age 70, and none of whom has ever
been impeached by Parliamentary request under the applicable proce-
dure,8 3 yields only a tenuous form of independence. Under the new Con-
stitution as under the old, judicial security can be repealed by ordinary
legislative process. Most judges support the Nationalist Party. The gov-
ernment can, moreover, afford to appoint more liberal judges because of
the jurisdiction-stripping measures and limited judicial review that have
governed since the 1950's. Judges politically opposed to the8 Nationalists
4
also tend in practice to get politically uncontroversial cases.
The Supreme Court of South Africa is made up of the Appellate Di-
vision, which is the highest court in the land, and seven provincial divi-
sions ranging from three to twenty-nine judges.85 There are also magis-
trates courts, which deal with the vast majority of apartheid-related civil
cases, but especially with criminal cases. Magistrates are often civil ser-
vants without a legal education who also exercise administrative func-
tions. The somewhat unrepresentative decisions of judges are the focus of
this article, however, because a right of appeal exists to the Supreme
Court, and the reported decisions of judges are much more reflective of
both the tougher social problems and the official attitudes of the legal
system than unreported magistrates' decisions.
For most judges in South Africa, then, the fact that they are part of
the system gives rise to no moral dilemma. The few judges who do feel a
moral dilemma in applying the unjust laws of apartheid arguably experi-
ence a feeling that should not have arisen, precisely because they are part
of the system and could not expect much real independence when they
were appointed. The fact that the judge accepted an obligation under
oath to apply the law and customs of South Africa compounds the sense
of obligation." The acceptance of such an obligation, however, does not
preclude the possibility that a person accepting a judicial appointment

82. REPUBLIC OF SOUTH AFRICA CONST. Act 110 of 1983, § 10(1).

83. Id. at § 10(7).
84. E.g., Kentridge, Telling the Truth About Law, supra note 80 at 649, 653 .
85. Republic of South Africa Constitution Act 110 of 1983, § 68.
86. The Supreme Court Act 59 of 1959 § 10(2)(a) (Judge's oath "to administer justice
to all persons alike without fear, favour or prejudice, and, as the circumstances of any par-
ticular case may require, in accordance with the law and customs of the Republic of South
Africa").
1986 JUDGES IN AN UNJUST SOCIETY

has a genuine desire to do justice to all persons alike - also part of the
oath - in accordance with law. Although it is almost inconceivable that
anyone would submerge strong reformist desires enough to become the
sort of establishment figure considered for a judicial appointment, it is
conceivable that a judge who becomes part of the judicial system will see
more, perhaps be repulsed at some tasks, and grasp the need for con-
fronting the problem of unjust law. Not all the judges who awaken to the
moral dilemma will have undergone a moral conversion, but such conver-
sions may be prompted by increased awareness of government policy, or a
particularly heinous example of forceful repression, or a piece of legisla-
tion. Whether the moral dilemma arises gradually through increased sen-
sitivity or suddenly through a reaction to a jurisdiction-stripping measure
or other specific law that drives home the sense of injustice, the judge will
seek ways to reconcile his perceived moral obligation with his judicial ob-
ligations to those who appointed him and those who seek justice in his
court. 87 The following Part discusses the dominant theories of law and
adjudication available in South Africa and elsewhere for that purpose.
II. THEORIES OF LAW AND ADJUDICATION

Various theories of the nature of law have attempted to clarify the

relationship between law and morality and the implications of the partic-
ular theory of law for a theory of adjudication. Jurisprudence has only
tangentially raised the problem of judges in unjust societies in the course
of discussing how to handle hard cases, in which no clear law applies, as
opposed to easy cases, in which legal answers are readily available. This is
in part because legal theory concerns itself with the normal legal system,
in which most laws are considered essentially just. But in either category,
hard or easy cases, the dilemma of the moral judge may assert itself.
Neither of the traditional models of law and judicial obligation -
positivist and natural law - seems to enthusiastically embrace the reality
of the judicial function: the law-making, creative role that judges inevita-
bly play. The ambivalence of both the positive and natural-law schools
about the law-making function of judges reflects an understandable fear-
of nonaccountable judges making law in free-wheeling fashion. But con-
straints such as precedent, legal theory, self-respect and popular tolerance
always exist in any society, and we have seen that in South Africa the
constraints are particularly severe. The South African government wants
to avoid "political" judicial review such as that in the United States. It is
almost absurd to speak of the central problem of judicial review (non-
accountable judges) when the society is characterized by drastically atten-
uated judicial review and a repressive sovereign Parliament totally unac-

87. The apparently sexist use of "his" in this and the following discussion is adopted
not only for facility of style but because there are no female judges in South Africa. The
society is almost as sexist as it is racist. What Paul Brest has said of American judges is
especially true of South African judges: they are "mostly white, male, professional, and rela-
tively wealthy." Brest, Interpretation and Interest, 34 STAN. L. REv. 765, 771 (1982).
 DEN. J. INT'L L. & POL'Y VOL. 15:1

countable to the majority of the population.

The positivist model of law has come to be associated with a mechan-
istic view of the judicial process, as and even the natural-law model may
lead to a mechanistic notion of the judicial role as merely 'declaring' the
law (although the law so declared usually hovers rather mysteriously
above or deep within the fabric of positive law). On the other hand, many
positivists have recognized the role of judicial legislation in "hard cases,"
and the association of natural law with divine law, morality, or justice
leads to the conclusion that a judicial appeal to natural law is more crea-
tive than simple application of positive law.

A. Positivism

The dominant legal theory in most of the world today, and especially
in South Africa, is positivism. 9 The foremost contemporary positivist is
H.L.A. Hart. 90 Hart has isolated three distinctive strands of positivism
originally set out by the utilitarians Bentham and Austin. First is the
insistence on the separation of law and morals. Second is the need for a
purely analytical study of legal concepts. Third is the imperative theory
of law as rules (Hart) or a command (Austin). 1
Human fallibility and the limited purview of law make it counterin-
tuitive today to assume that there is a necessary connection between law
and justice. Hart first clarifies the misconception that the positivists en-
dorse complacency about existing law. Positivists need not deny the rele-
vance of morality to the development of the legal system and specific
laws, nor deny the impact of law in shaping moral attitudes. Bentham, for
example, was a zealous reformer.92 Clearer thought is promoted, Hart
says, by remembering Austin's formula that "[t]he existence of law is one
thing; its merit or demerit is another." ' He concludes that there is no
necessary moral minimum content that a rule of law must satisfy in order
to be a law, beyond certain "fundamental" rules required for there to be a
point in having other rules (for example, those against free violence, and
for minimum guarantees of property). 94 Hart believes with Bentham that
a law may be a bad law and thus raise a dilemma of personal obedience,
95
but it is still a law.

88. D. LLOYD, THE IDEA OF LAW, 261-62 (1970).

89. South Africa has repeatedly been described as possessing an extremely positivistic
legal order. See, e.g., Final Rep't, supra note 78 at 125; DUGARD, HUMAN RIGHTS, supra note
11, at 378; Dugard, The Judicial Process, Positivism & Civil Liberty, 88 S. APR. L.J. 181
(1971); McWhinney, Race Relations and the Courts in the Union of South Africa, 32 CAN.
B. REv. 44, 45 (1954); Mathews, supra note 27, at 207.
90. See, Hart, supra note 2; H.L.A. HART, THE CONCEPt OF LAW (1961). Another influ-
ential modern positivist is Dennis Lloyd, supra note 88.
91. Hart, supra note 2, at 601.
92. Id. at 595.
93. Id. at 596.
94. Id. at 623.
95. Id. at 620. Dennis Lloyd has noted that in anomolous situations where power super-
1986 JUDGES IN AN UNJUST SOCIETY

The positivist theory of adjudication that follows from the idea of

law as rules distinct from morality would not be characterized by Hart as
a "mechanical" process of "finding" law.96 The judicial process is seen as
somewhat scientific research of the rules laid down by the legislature or
common law precedent. (The secular empirical metaphysics underlying
positivism come into play here.) Judges apply these rules even if harsh.
This ideal of "applying rules" is what has led to a vulgarization of positiv-
ism as mechanical. But because there can never be rules comprehensive
enough to cover every fact situation, judges inevitably engage in law-mak-
ing. Judges fill in the gaps and interstitially "legislate" where there is no
clear rule or where the "open texture" inherent in vague and ambiguous
general language leaves room for discretionary interpretation. 97 Moral
conceptions of what law ought to be do play a part when the judge makes
choices in the penumbras of hard cases. The Legal Realists' insights that
courts often turn to their own values and predispositions are relevant and
accurate on this point.98 Those values, policies or predispositions, how-
ever, are not "law." Justice Holmes, a positivist who adhered to the law/
morality distinction99 and significantly influenced Realism, summarized
the position: "I recognize without hesitation that judges do and must leg-
islate, but they can do so only interstitially; they are confined from molar
00
to molecular motions.'
One problem with advising the moral judge in South Africa to seek
guidance in the positivist model is that the judges' motions are, at best,
molecular. Considering the special potency of positivism in a system of
strong parliamentary sovereignty, the motions might be practically con-
fined to the sub-molecular realm. Judges in South Africa cling to the vul-
gar version of positivism that deprecates making law. Even if judges are
granted limited ability to effect interstitial relief, much of the vast legisla-
tive apparatus of apartheid is invulnerable from attack. And with South
Africa's form of parliamentary sovereignty, there are fewer gaps to be
closed and a greater risk that parliament will overturn an exercise of lib-
eral discretion between gaps.
The nature and role of the discretion to be exercised by the judge in
hard cases is also important. Dugard, Mathews, Kentridge, and other lib-
eral South African lawyers stress an expansive concept of strong discre-
tion in which "the judicial function is essentially an exercise in choice."' 1

sedes law, the realities of power must be considered in determining legal validity. LLOYD,
supra note 88, at 182-3.
96. Hart explicitly rejects the mechanical view and agrees with Austin's characteriza-
tion of this view as a "childish fiction." See supra note 2, at 609, n. 34, 610.
97. Id. at 608-615; see also HART, supra note 90. Dworkin describes the model in much
the same way. See DWORKIN, supra note 6, at 81-82.
98. Hart, supra note 2, at 605-608.
99. Holmes, The Path of the Law, 10 HARV. L. REv. 457, 459 (1897).
100. Southern Pacific Co. v. Jensen, 244 U.S. 205, 221 (1916)(dictum).
101. DUGARD, HUMAN RIGHTS, supra note 11, at 303; see also Mathews, supra note 27,
at 208.
 DEN. J. INT'L L. & POL'Y VOL. 15:1

They do not deny the severe constraints that exist in the South African
context, but urge judges to interpret those constraints in light of liberal
legal principles, which oppose these constraints. This expansive power of
interpretation will enable the judge to see the spaces between the rules as
quite large indeed. To put it another way, one might imagine that this
group of liberal lawyers would, perhaps properly, see fewer cases governed
by "settled meaning" than most positivists, including Hart.
Another problem with positivism in South Africa is the problematic
quest for legislative intent. This quest is problematic because, as Realist
Max Radin noted, legislative intent is frequently elusive or non-exis-
0
tent.0' Of course, the real problem with the quest for legislative intent in
South Africa is that the intent is likely to be malevolent. For this reason,
Dugard and others are understandably critical of the positivist approach,
which treats legislative intent as closer to the core sources of law than to
the penumbras. South African judges have indeed been (perhaps too) so-
licitous in the search for "true legislative intent," and have steadfastly
purported to "declare" the law illuminated by legislative intent. '10
Dugard calls for greater reliance on rules of interpretation "to moderate
the law's inequities," acceptance of "legal values" and "policy considera-
tions" as well as positive legal rules and recognition and denial of con-
cealed and inarticulate prejudices, such as the premise of loyalty to the
white status quo.'" Although he calls his new approach to law a "realist-
cum-value oriented approach"' 1 5 which is close to natural law and rejects
positivism, Dugard is trying to modify rather than eliminate the influen-
tial positivist model. The idea of the judge legislating between statutory
gaps is preserved. At the same time, Dugard urges greater creativity in
statutory interpretation by liberating the judge's discretion from legisla-
tive intent to more humanistic, value-oriented standards.
The flaw with even this version of positivism is that there is no
method of guaranteeing that the "legal values" or "policies" interpreted
by South African judges will advance human rights, particularly in light
of Dugard's recognition that background prejudices are difficult to eradi-
cate, even once they are laid bare. Dugard presumes that there is a tradi-
tion of accepted liberal legal values at the heart of Roman-Dutch common
law,'10 but it is doubtful that consensus on such values runs high in South
Africa. Values such as "racial purity" and "separate development" would

102. Dugard, The Judicial Process, Positivism and Civil Liberty, 88 S. AFR. L.J. 181,
183 (1971), citing Radin, Statutory Interpretation,43 HARV. L. REV. 863, 871-2 (1930).
103. See, e.g., Remarks of Chief Justice Ogilvie Thompson, Centenary Celebrations of
the Northern Cape Division, 89 S. APR. L.J. 30, 33 (1972) (a judge is bound by legislative
intent, and must "administer the law, not as he perhaps would like it to be, or as he might
consider it ought to be, but as set out in the relevant statutory provisions as interpreted.").
See also the judicial invocations of positivism in R. v. Koenig, 1917 C.P.D. 225, 242; Byers v.
Chinn, 1928 A.D. 433, 329; R. v. Christian, 1924 A.D. 101, 124.
104. DUGARD, HUMAN RIGHTS, supra note 11, at 366-388.
105. Id. at 400.
106. See supra note 11.
1986 JUDGES IN AN UNJUST SOCIETY

probably have more persuasive force. Hence, if it is a "fiction" for judges

to seek to discover legislative intent, it is also a fiction for them to ignore
a clearly evil legislative intent and an overwhelmingly repressive statutory
context.
Nevertheless, Dugard has brilliantly sketched the inadequacies of the
positivist model. In answer to Hart's question about whether the practical
consequences of stressing the distinction between law and morality have
been bad,10 7 Dugard resoundingly answers "yes!" The formulaic distinc-
tion has made it easier for the South African government to manipulate
the concept of law for unjust ends, and to enlist the support of judges in
perpetuating injustice. Judges are able to evade responsibility for the laws
they apply by deferring to Parliament and precedent and avoiding the
moral dilemma implicit in such activity. The situation in Dugard's eyes is
not unlike that in Nazi Germany, where Hitler exploited the distinction
so as to completely debase the German legal system.' Hart might an-
swer that, in theory, there is no reason why a judge might not take some
sort of moral action in response to his moral predicament, such as refus-
ing to apply the unjust law, or resigning. But the objection still stands
that the practical consequences of the distinction have been bad.

B. Natural Law

Natural law theory is the most potent historical alternative to posi-

tivism, because it manifests our constant striving for objective and uni-
versal values natural to all men. Natural law is a popular vessel for our
aspirations because it is such a flexible concept. From early on, the Greek
notion of natural law exploited the ambiguity of the term "nature." Na-
ture could at once refer to the natural order decreed by divine law, and to
the natural capabilities of man to live a virtuous and just life. Aristotle's
conception of natural law ambiguously involved both the idea of natural
rules universally binding men (though unequally, since some men he con-
sidered natural slaves), and the idea of man as a naturally social being
who could fulfill his potential in society. Cicero and the Stoics further
developed the dichotomy so that man could live a just life by ascertaining
the universal laws of nature through reason. In the Middle Ages, the
scholastics stressed the transcendent version of natural law, only to be
followed by humanist concepts of virtue in the Renaissance. Locke's ver-
sion of natural rights strongly influenced the leaders of the American
Revolution. After being submerged by positivism in the nineteenth cen-
tury, natural law experienced a revival after World War II as the world
sought standards by which to condemn the brutal and widespread fascist
violations of human rights. Although the legal philosophy of Ronald
Dworkin may be distinguished from the Aristotelian-Thomistic tradition
of natural law, it is his continuity with this tradition that makes him the

107. Hart, supra note 2, at 595.

108. DUGARD, HUMAN RIGHTS, supra note 11, at 395.
 DEN. J. INT'L L. & POL'O VOL. 15:1

chief representative of this school today.

Dworkin's theories of law and adjudication emphasize the essential
link between law and moral principles. Yet, despite stressing the relation-
ship between law and morality, Dworkin would disagree with the strong
or subtractive tradition of Catholic "higher" natural law, which holds that
"an unjust law is no law at all." He recognizes that law has moral dimen-
sion, because sometimes
the answer to the question of what the law is may depend on (though
it is never identical with) the question of what background morality
requires . . .This is so not only in cases in which some legislative
source deliberately embeds moral tests in legal rules, but also in cases
where it is controversial what the law requires because no legislative
source has said anything decisive at all. It is so not only when legal
principles embodying moral concepts are concededly decisive of legal
arguments, but also when the question in play is just the question of
what principles are to be taken as decisive. 10 9
Dworkin thus accepts the role of "moral reasoning" in adjudication, but
rejects the "absurd view"11 0 that an immoral or unjust law is not a law,
because laws exist that are "in fact unjust."1'11
Dworkin's natural justice theory of adjudication differs drastically
from the positivist model of rules sketched by Hart. Emphasizing the pro-
tection of individual rights, Dworkin identifies the distinctive materials of
judicial reasoning - moral principles - with rights. Principles are de-
fined early on as standards observed because they are requirements of
"justice or fairness or some other dimension of morality.' 1 12 Later the
focus is narrowed to arguments of principle, which "justify a political de-
cision by showing that the decision respects or secures some group
right." 13 Adjudication, which deals with principles, is distinguished from
legislation, which is characteristically about policies (instrumental goals).
Dworkin's "rights thesis" is that judicial decisions are and should be gov-
erned by principle, not policy, even in "hard cases.""' Additionally,
Dworkin attempts to satisfy the yearning for objectivity by providing that
"right answers" exist even in hard cases, although right answers cannot
be proved and may be controversial."'
This view of the nature and role of judicial discretion differs from
that of traditional positivists and the liberal South African supporters of
strong discretion.6 Dworkin believes that judges have no discretion in any
"strong" sense." Rather than an exercise of free choice, the interpreta-

109. DWORKIN, supra note 6, at 342.

110. Id. at 341.
111. See, e.g., id. at 122-3.
112. Id. at 22.
113. Id. at 82.
114. Id. at 84.
115. Id. at 81.
116. Id. at 31-39, 68-71.
1986 JUDGES IN AN UNJUST SOCIETY

tive function is viewed as bounded by objective legal texts. Background

considerations from the judge's life will inevitably play some role, but
they have no independent force and should be resisted. What results is
thus not a "choice" between competing texts and principles, but "judg-
ment" leading to a decision required by law. In short, Dworkin ties adju-
dication to the judges' job of seeking justice, and law is expanded (and
elevated above politics) so that judges 17
may "discover" just law and not
legislate, like the positivists' judges.'
In many ways, Dworkin's model sounds ideal for South Africa. The
thrust of positivism is order, whereas the thrust of natural law is toward
justice. Natural law was once strong in the country's jurisprudence. The
Dutch scholar Voet said that laws "must be just and reasonable." Grotius,
of course, was also in the natural law tradition. But positivism replaced
natural law in nineteenth-century South Africa. 1 ' Dworkin's model
avoids the tendency of positivism to subordinate the judge to the legisla-
ture even when the judge is making law, as if he were a deputy legislator
who enacts law that the legislature "would enact" if confronting the prob-
lem. Where political power is denied to blacks, an envelope of protection
provided by just law declared by judges sounds attractive. The fact that
judges are seen as having no strong discretion - they only "discover"
rather than "make" law - would make liberal decisions more politically
palatable in a country like South Africa. One of Dworkin's normative
goals seems to be enhancement of judicial power; that is important in a
system where blacks have access to power only in the courts. And the
injection of justice and morality into the distinctive activist judicial role
contemplated by Dworkin could not fail to help blacks gain their individ-
ual and group rights. Or could it?
When the model is applied to South Africa, several worrisome
problems arise. The first is that Dworkin assumes a liberal democratic
system with a majoritarian representative legislature and shared egalita-
rian values, an assumption woefully inapposite for South Africa. The fear
of unconstrained anti-democratic discretion that prompts Dworkin seems
at best irrelevant and at worst defeatist in South Africa. In his Introduc-
tion to Taking Rights Seriously, Dworkin sets out to "define and defend
a liberal theory of law."" 9 Most of his theory is descriptive, but the nor-
mative aspects conservatively seek a "political justification" for the "pre-
sent structure."' 2 0 Dworkin's focus is presumably on the Anglo-American
tradition, but especially on the United States with its Constitution and
powerful Supreme Court. He certainly would not try to justify an authori-
tarian and anti-egalitarian political structure like that of South Africa,
which denies political rights to the majority of its people.

117. Id. at 82.

118. See DUGARD, HUMAN RIGHTS, supra note 11, at 393; H. HAHLO & E. KAHN, SOUTH
AFRICA: THE DEVELOPMENT OF ITS LAWS AND CONSTITUTION 205-06 (1960).
119. DWORKIN, supra note 6, at vii.
120. Id. at 123.
 DEN. J. INT'L L. & POL'Y VOL. 15:1

The distortions following from this initial assumption are clearly evi-
dent at the adjudicative level, where Dworkin assumes that judgments
about political rights merely reflect "political decisions of the past.""'
The sources of rights (and, thus, of "principles") are identified variously
as institutional history such as cases and statutes, as well as personal and
community morality.'" Judges are therefore bound to be affected by un-
just precedent by the overriding requirement of consistency. Yet Dworkin
is ambivalent about morality. It sometimes seems to be akin to positive
morality (the specific customs and mores of a given community), and at
other times to critical morality (by which community morality is judged
and criticized). For example, "background" rights are defined as those
that provide a justification for political decisions of "society in the ab-
stract. 1 2 3 But with society left undefined in size, this could be either the
positive morality of white South Africa or the critical morality of world
society. Confusion is compounded, but then dispelled, when Dworkin il-
lustrates the ideal judicial process in hard cases by having his superhu-
man "Hercules" appeal to the "community's moral traditions," or "popu-
lar morality.' 2 4 These formulations indicate that Dworkin is referring to
positive community morality. Of course, the fact that the white commu-
nity has a monopoly over legal and political power means that it will
probably be defined as the relevant "community." And appeal to white
community morality in South Africa is far more likely to restrict blacks'
rights than protect or expand those rights.
General difficulties with Dworkin's judicial principle/legislative policy
dichotomy are also exaggerated in the South African context. His position
that judges do not deal with policy arguments forces him into a strained
account that allows the substitution of principle for policy arguments.' 5
Particularly in the extreme situation of South Africa, arguments of prin-
ciple about the rights of the white minority or the black majority tend to
blur with arguments of policy that segregation reduces social tension or
increases long-term social unrest. Apartheid, or separate development, is
usually seen as a collective community goal and thus as a policy, although
the black community would not share this goal. South African judges who
invoke apartheid policy must presumably be understood to refer to the
competing rights of the white minority. Dworkin's theory is unconcerned
with this, because the principled argument - here, the competing right
of the white minority - will not always be as strong when balanced
against other rights as the policy argument here, apartheid. In the ex-
treme circumstances of South Africa, however, the competing rights of

121. Id. at 87.

122. Id.
123. Id. at 93.
124. Id. at 123-4.
125. "If a judge appeals to public safety or the scarcity of some vital resource, for exam-
ple, as a ground for limiting some abstract right, then his appeal might be understood as an
appeal to the competing rights of those whose security will be sacrificed, or whose just share
of that resource will be threatened if the abstract right is made concrete." Id. at 100.
1986 JUDGES IN AN UNJUST SOCIETY

the white minority are overwhelmingly strong both in legal theory and in
power politics. Dworkin's answer to the substitution problem is thus not
responsive to the specific problems presented by a system like South Af-
rica's, or to the general problems of how to phrase, set the level of gener-
alization, or weigh competing rights or principles. As apartheid is seen by
many South Africans as part of the moral or religious order, it is hard to
say that racial discrimination is not a moral principle. 2 6
A final aspect of Dworkin's theory that would bode ill for South Afri-
can blacks is his approach to statutory interpretation. As we have seen,
many positivists view narrow legislative "intent" as close to the core con-
cept of law, but follow Hart in tending to exclude "purposes" attributed
to statutory rules from "law. 1 2 7 Dworkin, on the other hand, tends to
view the "purposes" of rules as law. 2 ' The same tendency on the part of
many South African judges is criticized perceptively by Dugard as posing
a threat to recognition of rights for blacks.

C. Positivism v. Natural Law on the Dilemma of the Moral Judge

Neither the positivist nor the natural-law model provides clear guid-
ance for the moral judge in an immoral system. But each suggests a tenta-
tive approach. Recall that the positivist distinction between law and mo-
rality suggests that existing laws may be unjust, but are still law. It is
after one notes that law exists and is unjust that the moral dilemma of
personal obedience arises."' The task of the judge is complicated by the
addition of judicial obligation to legal and moral obligations. Positivism
thus may result in a judge's either applying the law because it is his obli-
gation to do so, or not applying the law because personal moral obligation
defeats judicial obligation and prohibits obeying or applying an unjust
law. Hart merely wants to make certain that the problem is not concealed
in an illusion that laws always accord with morality.
Yet this apparently prudential focus on confronting the moral di-
lemma - without discussion of judicial obligation - may, in effect, fore-
close alternatives between the extremes of applying or not applying the
law. Evading the issue by avoiding it, for example, is opposed to con-
fronting the moral dilemma. Significantly, Hart's attack on the doctrine
that unjust laws are not laws begins only after noting that the "doctrine
meant that every lawyer and judge should denounce statutes that trans-

126. Dworkin admits that there "is no persuasive analysis . . .that insures that the
principle that blacks are less worthy than whites can be rejected as not a principle at all."
Id., at 343. Dworkin's original definition of principles, which requires that they be require-
ments of "morality" leaves ambiguous the distinction between what is good and what is
accepted as good. See id. at 83.
127. Hart, supra note 2, at 614-15.
128. DWORKIN, supra note 6, at 105-108.
129. "Surely the truly liberal answer to any sinister use of the slogan 'law is law' or of
the distinction between law and morals is, 'Very well, but that does not conclude the ques-
tion. Law is not morality. Do not let it supplant morality.'" Hart, supra note 2, at 618.
 DEN. J. INT'L L. & POL'Y VOL. 15:1

gressed the fundamental principles not as merely immoral or wrong but

as having no legal character." Whether or not the option of in-court or
out-of-court protest on other grounds remains open, Hart clearly disfa-
vors such methods even as rhetorical tools. Similarly, one would think
that lying about the law would be condemned for not confronting the
moral issue, for evading the imperative aspect of legal positivism, and for
distorting the analytical rigor of the law's vocabulary.
Moreover, the practical consequences of the positivist's distinction
between law and morality may lead to concealment rather than to expo-
sure of the moral dilemma, because judges, especially in easy cases, can
take refuge in the concept of "merely applying the law." Whether or not
positivism is the chief factor in debasement of legal systems like that of
Nazi Germany or South Africa, it seems fair to say that it is at least a
significant factor. '30 If judges are able to use positivism to evade responsi-
bility in easy cases, think how much more tempting it will be to do so in
hard cases, despite the theoretical existence of strong law-making
discretion.
The natural law approach to the problem of the judge in the wicked
legal system remains undeveloped, and points in contradictory directions.
On one hand, some natural law theorists suggest that an unjust law is no
law at all. On the other hand, theorists such as Dworkin would incorpo-
rate community moral principles as part of law. The consequent emphasis
on the judge's political responsibility, lack of discretion, the duty to be
consistent, and the "right answer" thesis pushes the judge in the direc-
tion of applying unjust law, even if it is in fact unjust or inconsistent with
the judge's personal moral views.' Judges cannot decide hard cases us-
ing background rights, but are limited to institutional rights. "Institu-
tional autonomy insulates an official's institutional duty from the greater
part of background political morality."" 2 Although the judge's own moral
convictions and beliefs will inevitably enter into his decision to a de-
gree,' 3 the judge is not free to give them independent force in his judge-
ment. For example, if two coherent justifications can be given for earlier
Supreme Court decisions enforcing the due process clause, with one con-
taining an extremely liberal principle that cannot be reconciled with the
criminal law of most states, and the other containing no such principle,
"Hercules cannot seize upon the former justification as license for decid-
ing the abortion cases in favor of abortion, even if he is himself an ex-
treme liberal."'3 4
Two other elements of Dworkin's theory may loosen this rigid con-

130. As several writers would imply. See Fuller, Positivism and Fidelity to Law, supra
note 2, at 657; DUGARD, HUMAN RIGHTS, supra note 11, at 374; R. COVER, JUSTICE ACCUSED:
ANTISLAVERY AND THE JUDICIAL PROCESS 1 (1975) [hereinafter cited as COVER].
131. DwORKIN, supra note 6, at 87-89.
132. Id. at 101.
133. Id. at 123-124.
134. Id. at 126 (emphasis added).
1986 JUDGES IN AN UNJUST SOCIETY

ception of judicial duty. The first is the largely semantic distinction made
between the "choice" or "discretion" recognized by the positivists and the
"judgment" stressed by Dworkin. While denying that judges have discre-
tion, Dworkin may let choices or discretion in through the back door by
recognizing that interpretation and judgment are required in determining
what the relevant moral principles are. As Dworkin's famous example of
Tal's smile in the chess game between the Russian grandmaster and
Fischer illustrates, the referee (like the judge) must interpret contested
conceptions such as "the character of the game" (or the character of the
legal system). 35 Despite Dworkin's protestations, this function could pro-
vide a loophole for judicial originality.
Dworkin never adopts the extreme natural law position that unjust
law may not be applied because it is not "law." But a second element of
Dworkin's theory that may increase the flexibility of the moral judge in
an immoral system links Dworkin to more traditional natural law
thought: the passion for justice, for infusing law with a moral dimension.
This passion is what drives Dworkin to defend the present American lib-
eral structure as essentially just. His initial approach thus tended to bind
the judge to apply these essentially just legal principles, certainly in easy
cases, but also in hard cases. Little treatment is given to the problem of
his "fairness" requirement of judicial consistency yielding unfair results:
the primary answer Dworkin gives is that only the community's concep-
tion of fairness may override the argument for consistency so as to result
in disposing of unjust precedent, for example, as a "mistake.""13 Yet even
in easy cases, legal rights may conflict with background moral rights. And
in hard cases, background moral rights of the particular community may
not be very "moral" (in a critical sense).
In his Reply to Critics, Dworkin's response is close to the positivists'
distinction between legal and judicial duty as opposed to moral duty.1"'
He reiterates his idea that certain allegedly "moral" principles must meet
a threshold adequacy of "fit" with institutional history before they can
count as a justification, and amongst those principles . . . the morally
soundest must be preferred." ' Morality here is clearly used in a new
way, as a critical morality external to the wicked system. Dworkin contin-
ues: "it may be that no principle we would find acceptable on grounds of
morality could pass the threshold test," in which case "the general theory
must endorse some unattractive principle as providing the best justifica-
tion of institutional history, presenting the judge with a legal decision and
also, perhaps, a moral problem."13 9 This is the situation in South Africa,

135. Id. at 102-3.

136. Id. at 122-23.
137. See id. at 327 ("We might want to describe the moral conflict a judge faces in such
a case by distinguishing his 'legal' duty which is to find for legal rights, from his 'moral' or
'overall' or 'final' duty, which may require him to ignore legal rights.")
138. Id., at 342.
139. Id.
 DEN. J. INT'L L. & POL'Y VOL. 15:1

where liberal principles would not provide an adequate fit because of the
high number of decisions that would have to be thrown out as mistakes.
The key issue then would be "whether the principle is so unjust that it
would be wrong for the judge to enforce any legal right it supplies, and
right to him to lie in order to avoid doing so. '' 4" Dworkin concludes that
"it may be that he must lie, because he cannot be of any help unless he is
understood as saying, in his official role, that the legal rights are different
from what he believes they are."'" The alternatives of resigning or pro-
testing on moral grounds are discounted as probably ineffective.
It is interesting that Dworkin chose to endorse the judicial lie as an
alternative, because it seems so inconsistent with his right-answer thesis,
which carries normative "gravity" calling for its application because it is
"right", and with his ideal of judicial consistency, which he has come to
call "integrity" in later works. Perhaps the solution is not surprising,
however, in light of Dworkin's constant identification of legal rhetoric
with reality. The right-answer thesis is defended as reflecting the way in
which lawyers and judges talk, and the moral judge's "lie" can similarly
be defended on normative grounds as encouraging "the beneficial and
unifying assumption that justice is always relevant."' 4 2 With Hart, how-
ever, Dworkin agrees that it would be misleading and3 therefore "unwise
to make this lie a matter of jurisprudential theory.'
Both the positivist and the natural-law models give ambivalent guid-
ance to the moral judge in a wicked system. Sophisticated positivists want
to expose the moral problem, which leaves open the possibility that the
judge will choose to follow moral duty rather than legal duty. Positivism
also tends to emphasize legal duty, and the rules binding judges to apply
the law are positive legal as well as moral rules. Natural law similarly
contains contradictory elements. Competing interpretations of natural
law may result in the application of immorality as morality in the wicked
legal system, and this fact compels even sophisticated natural-law schol-
ars to lean toward the idea that judges may not apply unjust law, or must
lie in order to avoid doing so. The following section examines the actual
response of two wicked legal systems to the dilemma of the moral judge.

III. THE DILEMMA OF THE MORAL JUDGE

A. The American Experience

The dilemma of the antislavery judge in America under slavery

presented many of the same problems faced by South African judges to-
day. Laws treating slaves as property, and a Constitution that in part
sanctioned slavery confronted judges with a system of unjust laws on a
daily basis. The passage of the Fugitive Slave Act in 1850 created new

140. Id. at 343.

141. Id. at 327.
142. Id. at 338.
143. Id. at 327.
1986 JUDGES IN AN UNJUST SOCIETY

opportunities to question the constitutionality of slavery and align the

law with growing antislavery sentiment, but these opportunities were for
the most part missed. Much scholarship has treated the issue of how con-
firmed antislavery judges like Joseph Story, Lemuel Shaw, and John Mc-
Lean adhered to their roles as judges and applied slavery laws, but none
has been as lucid, comprehensive, and thoughtful as that of Professor
Robert Cover in Justice Accused.""'
Cover begins by sketching the background traditions of natural law
and positivism in eighteenth-century America. The sources of the natu-
ral-law tradition were many (ranging from the Bible and philosophy to
Grotius and Somerset's Case),'45 and some of the sources justified slavery
as natural. But most of the more recent and authoritative natural-law
sources (including Somerset's Case, which confirmed Blackstone's natu-
ral-law condemnation of slavery)"" supplied a framework for arguments
attacking slavery laws. The natural-rights rhetoric of the American
revolution strengthened the arguments. Natural law persisted in conflict-
of-laws, international law, and state constitutional contexts. Yet, at the
time the nation was founded, constitutional positivism, which included
natural law phrases within a written constitution, asserted itself. What
Cover calls a "pattern of moral-formal conflict" thus emerged, and influ-
47
enced the judiciary.1
The first instance of judicial construction Cover examines involves
cases falling under the natural-law "free and equal" clauses of positive
law state constitutions. South Africa, of course, has no such documents.
These were adopted as part of the "first emancipation" of slaves in the
northern states in the years following the Revolution. In Vermont, on the
one hand, construction was easy, because the 1777 constitution specifi-
cally called for the prohibition of slavery. 48 In Virginia, on the other
hand, no one thought that the clause adopted in 1776 freed the state's
dominant labor system: its quarter-million slaves. Judicial construction
depended on the extra-judicial climate, including the presence of anti-
slavery pressure groups, and prior trends toward freedom. What is inter-
esting is that often the most forthright opponents of slavery, like St.
George Tucker of Virginia, could write opinions sanctioning slavery. A
prominent opponent of slavery, and Tucker's teacher and predecessor at
William and Mary, George Wythe had written the opinion of the Rich-
mond District Court of Chancery in Hudgins v. Wright 140 holding that
an Indian family that showed no negroid features was presumptively free
on the basis of appearance and the free and equal clause. Tucker felt
obligated to affirm the case merely on the racial ground with explicit dis-

144. COVER, supra note 130.

145. See supra note 8.
146. COVER, supra note 130, at 15-17.
147. Id. at 29.
148. Id. at 43.
149. 11 Va. (1 Hen. & M.) 133 (Va. 1806), cited in COVER, id. at 51.
 DEN. J. INT'L L. & POL'Y VOL. 15:1

approval of Wythe's expansive presumption of freedom. Tucker's opinion

must have been painful to write, because Wythe had just been murdered
for providing that Michael Brown, a former slave protege, be named ben-
eficiary of Wythe's estate.
Other judges at later times, in states where abolitionist rhetoric was
fiery and slavery was no longer a fact of life, were nonetheless constrained
by a "thoroughgoing judicial positivism."'"5 Instead of capitalizing on the
opportunity presented by the antislavery climate, Judges Randolph and
Nevius of the New Jersey Supreme Court relegated the free and equal
clause to a mere abstract, legally inoperative "preamble." Their chief jus-
tification was that an existing gradual scheme of emancipation rendered
the clause moot. 151 Thus they refused to free slaves petitioning their court
in the Post and Van Buren cases' 52 in the absence of a clearly expressed
legislative intent to do so. Cover attributes the harsh tone and intolerant
approach of Randolph and Nevius to their reaction to the demands made
on them by the opportunity presented for emancipation through judicial
law-making, the merely imitative aspects of New Jersey's adoption of the
clause in 1844, the perceived need to downgrade the constitutional uto-
pian rhetoric of the ideological antislavery advocate in the case, and the
perception that the gap between law and morality was closing on its own
53
accord.1
The second instance of judicial construction that Cover examines in-
volves statutory interpretation of emancipation and private manumission
acts. Not surprisingly, the same positivism nd role fidelity he discovered
in the context of broad and vague state constitutional language is also
present in interpretation of the narrow language of such statutes. Statu-
tory interpretation in favorem libertatis was a possible approach in
America under slavery, as it is in present-day South Africa. A statutory
presumption in favor of liberty can be applied in any of several ways:
(1) always; (2) only if the statute itself is held to have a liberal pur-
pose; (3) only if competing counterlibertarian interests do not outweigh
the liberal purpose; (4) only in areas of doubt not covered by express stat-
utory language; or (5) never.

The presumption in favorem libertatis was not always applied in the

United States under slavery, but it was applied occasionally. Judge Wil-
liam Tilghman of Pennsylvania applied it in the second way mentioned
above: only when the statute itself had a liberal purpose. In Miller v.
Dwilling 154 Tilghman noted the silence of the Pennsylvania Emancipa-
tion Act of 1780 on the issue of whether a child born to a servant inden-

150. COVER, supra note 130, at 60.

151. Id. at 58. Compare the arguments of some South Afircan judges that the govern-
ment's program of gradual reform renders liberal interpretations unnecessary.
152. State v. Post, State v. Van Buren, 20 N.J.L. 368 (1845), cited in COVER, at 55.
153. COVER, supra note 130, at 59-61.
154. 14 Serg. & Rawle 442, 443 (Pa. 1826), cited in COVER supra note 130, at 66.
1986 JUDGES IN AN UNJUST SOCIETY

tured for a term of years (as opposed to a slave for life) would be free.
Finding no "clear expression" to this effect in the statute, Tilghman
found such bondage too "contrary to the general spirit of the act."' 55
Such a tepid application of a presumption in favor of liberty is related to
the third approach mentioned above, for even if a statute has liberal pur-
pose, it may be seen to balance competing counterlibertarian interests,
such as property or security. The judge may rationalize that it is "unfair"
to favor either side in the balance. In any event, Cover notes that the
presumption in favor of liberty was never applied as a presumption that
operated of its own force, without regard to legislative intent.'5 6
By far the most common approach to statutory interpretation, and
one for which South African judges have been rightly criticized, is a posi-
tivistic inquiry into what the law is. As the civil war grew closer, Virginia
courts interpreted a private manumission statute in this way, so as not to
grant freedom to the child born to the manumitted slave. In Maria v.
Surbaugh 157 and Gregory v. Baugh 158 the interpretative process was de-
scribed as a simple positivistic task of ascertaining and applying the law.
Because it is the legislature's task to resolve moral and political issues
prior to judicial treatment, previous cases applying the presumption in
favor of liberty were said to have relaxed the rules of law "too much." 59
Cover summarizes the sporadic use of libertarian presumptions:
Because formal principles do not decide all cases, because there re-
mained areas of doubt as to what the legislature intended, how to
characterize the broader spirit of legislation, or which of the uses of
context were more significant, there remained a fair ground for differ-
ence of opinion between men of equally good faith but unequal liber-
tarian convictions. 60
The final cases Cover stresses arose under the fugitive slave clause of
the Constitution, and after 1850, the Fugitive Slave Act. The American
experience under the fugitive slave clause is now well known. Positivism
bound the Supreme Court and even confirmed opponents of slavery like
Justice Story and Judge Lemuel Shaw. Deferring to the importance of the
constitutional compromise, the judges almost uniformly felt duty-bound
to adhere to the clause and not to consider natural law.' 6' From 1850
until the outbreak of the Civil War, new constitutional issues regarding
the Fugitive Slave Act were raised and resolved. The statutory structure
established arguably violated due process by excluding the fugitive's testi-
mony and providing pecuniary incentives for the commission's resolution
of cases against the fugitives. Yet Chief Justice Shaw of the Massachu-

155. Id.
156. Id. at 80.
157. 23 Va. 287 (2 Rand.) (1824), cited in COVER, supra note 130, at 74-75.
158. 29 Va. (2) (Leigh) 665, 680 (1831), cited in COVR, supra note 130, at 74-75.
159. Id.
160. Id. at 81.
161. Id. at 171.
 DEN. J. INT'L L. & POL'Y VOL. 15: 1

setts Supreme Court held the act free from constitutional doubt in Sims
Case. 6 2 The West soon followed suit when Justice John McLean of Ohio
6
also upheld the constitutionality of the Act.6' Finally, the Supreme Court
of the United States, in Abelman v. Booth, dealt "the last and greatest
blow of the Taney court against antislavery."164 The Supreme Court de-
nied states the habeas corpus power to reach the restraint of one held
pursuant to federal process. The constitutionality of the 1850 Act was
also, incidentally, upheld.
Applying cognitive dissonance theory to the decisions of the United
States under slavery, Cover isolates several typical judicial responses: (1)
elevating the stakes of formal adherence to law, (2) emphasizing a mech-
anistic view of the obligation to apply law and only "law," and (3) ascrib-
165
ing responsibility for the decision to other groups like the legislature.
One might expect these characteristic responses from a moral judge in
any immoral legal system. As we shall see in the next section, they show
up in twentieth-century South Africa as well.

B. The South African Experience

The most thoughtful examination of the judicial process in South Af-

rica is that of John Dugard. In Human Rights and the South African
Legal Order 16 he discusses the judicial techniques used in some of the
more famous cases. As in the United States, liberal judges are easier to
identify than liberal decisions. Like Cover, Dugard concludes that positiv-
ism contributed to abdication of judicial duty to protect human rights.
Some early decisions, however, showed that judges were trying to err
on the side of human rights. In 1882, Judge Kotze decided In re
Marechane 167 in which he held that all persons are regarded as equals
under common law in the absence of an express statute to the contrary.
This is equivalent to applying a presumption in favor of liberty, like that
discussed by Cover. Marechane is Dugard's Brown, because its abstract
rhetoric suggests that there are liberal principles inherent in the hybrid
common law of South Africa. Unfortunately, other pre-Nationalist deci-
sions of South Africa after the Union of 1910 gave the imprimatur of
judges to racial segregation of the sort condemned in Brown. In the infa-
mous case of Moller v. Kemioes School Committee, "0 a Cape statute
that did not require school segregation was applied to the action of the
Kemioes School Committee requiring such segregation. Sir James Rose

162. 61 Mass. (7 Cush.) 285 (1851), cited in COVER, supra note 130, at 175-78.
163. Miller v. McQuerry, 17 F.Cas. 332 (C.C.D. Ohio, 1853) (No. 9, 583), cited in COVR
supra note 130, at 183.
164. CoVER, supra note 130, at 187, citing Abelman v. Booth, 62 U.S. (21 How.) 506
(1859).
165. COVER, supra note 130, at 229-238.
166. See note 11, supra.
167. 1 S.A.R. 27, 31 (1882).
168. 1911 A.D. 635.
1986 JUDGES IN AN UNJUST SOCIETY

Innes of the Cape Provincial Division read the statute restrictively and
would not assume discriminatory legislative intent in the absence of a
clear expression to the contrary. Some of the most distinguished judges of
South Africa, including Chief Justice De Villiers and Justice Koetze of
the Appellate Division, reversed. Some of the justices suggested that per-
haps the separate facilities were unequal; others suggested that the ine-
quality did not matter. The thrust of the decision was that the probable
intent of the legislators - who, being from Holland, regarded natives as
inferior - and the political morality of the community made it difficult
to ignore the malevolent legislative intent, even if the court could not
"from a philosophical or humanitarian view be able to approve of this
' 169
prevalent sentiment.
The same approach of recognizing or reading discriminatory legisla-
17 0
tive intent into a neutral statute guided the majority in R. v. Padsha.
There, a three-judge majority, over two dissenters including Sir James
Rose Innes, reversed a lower court holding that an immigration statute
authorizing exclusion on economic grounds could not be used as a blanket
racial prohibition against all Asians.
Nevertheless, a relatively liberal undercurrent of decisions persisted
until 1934. In Dadoo Ltd. and Others v. Krugersdorp Municipal Council
171 a statute denying Asians the right to hold property was
strictly con-
strued to allow their non-Asian corporation to do so. R. v. Dedtody 172
involved the very strict construction of a statute providing that "any na-
tive" without a pass was guilty of a crime. In an opinion very sensitive to
liberty, the court looked at the history of the Transvaal and determined
that the harm the legislation was aimed at was male violation of the pass
laws. The court refused to "extend" the pass laws to African women.
Other pre-1934 decisions involving "neutral" statutes that did not author-
ize discrimination tended more toward Brown than toward Plessy v. Fer-
17 4
guson.1 7 3 In R v. Plaatjies, for example, the court held that a city could
not set aside an "exclusively white" section of a stream for swimming.
Similarly, Williams & Adendorff v. Johannesburg Municipality 175 de-
feated a municipal attempt to provide separate trams for "coloreds."
In 1934, the basic legal structure changed when South Africa decided
its Plessy: Minister of Posts & Telegraphs v.Rasool. 176 The statute at
issue in Rasool was silent on the existence or validity of racially segre-
gated post-office counters. In a three-to-one decision of the Appellate Di-
vision, a majority held not only that "separate but equal" regulatory dis-

169. Id., at 643-4.

170. Cited in DUGARD, HUMAN RIGHTS, supra note 11, at 309.
171. 1920 A.D. 530.
172. 1926 A.D. 198.
173. 103 U.S. 537 (1896) (separate but equal facilities are constitutional).
174. 1910 E.D.L. 63.
175. 1915 F.P.D. 106.
176. 1934 A.D. 167.
 DEN. J. INT'L L. & POL'Y VOL. 15:1

crimination is legally "reasonable," but that separate post offices

accorded with the history of the Transvaal, "accepted principles and
good sense.' 17 Dugard is quick to point out that the court in Rasool was
actually making law, rather than merely declaring and applying it, 17 be-
cause the court overruled prior precedent (e.g. Williams & AdendorfJ)
and the trend of the law toward at least formal equality. The case went
even further than Moller in that it sanctioned racial segregation as rea-
sonable and respectable.
Even after the Nationalists came to office in 1948 and "separate but
equal" doctrine was clearly established in South Africa through Rasool,
some courts struck down regulations where substantially unequal facili-
ties were provided. "Mere technical inequality," however, was insufficient
grounds to do so. In R. v. Abdurahman,17' and R. v. Lusu,180 situations in
which no first class railway coaches were reserved for non- Europeans, or
were reserved but not substantially equal, were struck down by the Ap-
pellate Division. Unfortunately, these decisions resulted in legislative cod-
ification of the separate but unequal doctrine in the Reservation of Sepa-
rate Amenities Act."8 ' The fact that judges remained on the bench who
were unsympathetic with the early "white domination" form of apartheid
precipitated not only this legislative rebuke, but the constitutional crisis
of the 1950's described above. 8 ' Unlike the situation in the United
States, therefore, where Congress took over the enlightened libertarian
role played by the Supreme Court, the South African legislature took a
reactionary and regressive approach.
With the success of its court-packing plan, the government joined in
the paeans of praise about judicial independence as a spate of executive-
minded race and security decisions was handed down. In the field of race
relations, a number of decisions failed to apply Abdurahman, even
though no statute directly addressed the factual situation at issue. Mus-
tapha and Another v. Receiver of Revenue, Lichtenburg, and Others 188
involved the termination of Indians' trading permits on racial grounds.
The court upheld the termination, although the statute did not expressly
authorize unequal administrative treatment. A black lawyer held in con-
tempt for refusing to sit at a separate courtroom table got no relief in R.
v. Pitje,1 " although courtroom tables were not reserved for whites only
under the Reservation of Separate Amenities Act or any other authority.
In Minister of Interior v. Lockhat and Others,'" the removal of a group

177. Id., at 175-177.

178. DUGARD, HUMAN RIGHTS, supra note 11, at 314.
179. 1950 (3) S.A. 136 (A.D.).
180. 1953 (2) S.A. 484 (A.D.). See also, 1943 C.P.D. 242, 251 (test is manifestly oppres-
sive or unjust "equality").
181. Act 49 of 1953; see, n. 18, supra.
182. See notes 57-64 and accompanying text supra.
183. 1958 (3) S.A. 343 (A.D.).
184. 1960 (4) S.A. 709 (A.D.).
185. 1961 (2) S.A. 587, 602 (A.D.).
 1986 JUDGES IN AN UNJUST SOCIETY

of Durban Indians under the authority of the Group Areas Act was chal-
lenged. Though the Group Areas Act did not authorize the unequal treat-
ment that resulted, the court held that Parliament "envisioned" such un-
equal treatment, and it was not for the court to decide whether such
treatment was ultimately for the state's good. A similar trilogy of 1960's
cases in the security area illustrates the refusal of South African courts to
carefully scrutinize challenges to the ideal of individual liberty. In Loza v.
Police Station Commander, Durvanville'86 the court could have strictly
construed the new security laws to limit police powers. But instead the
Appellate Division held that a detainee can be re-arrested and detained
again after the expiration of the 190-day period provided for by statute.
In Schermbrucker v. Klindt, N.O., 1 87 the wife of a detainee got news of
the inhumane torture her husband was experiencing. She urgently ap-
plied for an injunction to stop the police brutality. The police denied the
charges. The court read the purposes of the detention statute broadly,
refused to apply a presumption that the legislature does not intend to
interfere with the judiciary and held, over two strong dissents, that it had
no power to allow a detainee's testimony (i.e., a detainee had no power to
testify) in an application to prevent wrongful interrogation. Finally, one
of the most widely criticized South African opinions, Rossouw v. Sachs, 88
found an intent on the part of the legislature to deny reading and writing
matter to detainees, although the security statute involved did not ex-
pressly remove such rights.
Dugard rightly concludes that the response of the South African judi-
ciary to race and security cases such as those sketched above has been
inadequate, and (with Cover) he places significant blame on the vulgar
positivism that declares the law "as is" and takes the search for "true"
legislative intent as a definitive guide. "This enables judges," Dugard
writes, "to apply the harshest of laws obediently with an easy conscience
and may result in a failure to realize the extent to which technical rules of
interpretation may be involved to moderate the laws' inequities.' ' 89
Cover's other two observations are also relevant to the many South Afri-
can decisions discussed here: the stakes of formal adherence to the law
were elevated, which is not too difficult to do in volatile South Africa, and
responsibility for the decision was usually ascribed elsewhere, to the legis-
lature. Dugard is correct about the importance of the search for intent in
these and more recent cases, but he may be somewhat artificial or unreal-
istic in playing down the fact that an "evil" intent in many apartheid
laws is often discernible, and in assuming that most South African judges
would want to invoke rules of interpretation to moderate the laws' inequi-
ties.. While Cover focused only on the antislavery judge, Dugard looked at

186. 1964 (2) S.A. 545 (A.D.).

187. 1965 (4) S.A. 606 (A.D.); See Dugard, The Courts and Section 6 of the Terrorism
Act, 87 S. APR. L.J. 289 (1970).
188. 1964 (2) S.A. 551 (A.D.).
189. DUGARD, HUMAN RIGHTS, supra note 11, at 374.
 DEN. J. INT'L L. & POL'Y VOL. 15:1

all judges ("moral" and "immoral"). Dugard's study shows us that subjec-
tive values do play a greater role in immoral systems than we might
think. Some judges - at least until the court packing plan of the 50's
succeeded - were able to ameliorate the evils of apartheid somewhat.
Since Dugard wrote Human Rights and the South African Legal Or-
der, several cases continue to exhibit the flaws of excessive deference to
the legislature and executive to which he refers.190 Other cases continue
the minority position of ameliorist sensitivity to liberty wherever possi-
ble. 191 The decisions and out-of-court statements of Judges Didcott,
Shearer,"s Milne,' and Corbett 94 are particularly noteworthy in this re-
gard. All in all, the room for creative statutory interpretation is limited in
South Africa because of malevolent legislative intent. Before the
problems of such malevolent intent can be remedied, they must be fully
recognized. Recognizing the harsh realities underlying South Africa's re-
pressive laws should not be an impediment to change; it should be the
first step to change.

IV. WHAT'S A JUDGE To Do?

Judges in South Africa are unquestionably part of the system. On

one level, everyone within a system of unjust laws is tainted. This may be
especially true of judges, for one may safely assume that they would not
be judges if they saw the whole system as utterly unjust. But such a blan-
ket condemnation is useful only as abstract rhetoric. On a more realistic

190. See, e.g., Goldberg v. Minister of Prisons, 1979 (1) S.A. 14 (A) (no enforceable
legal right of prisoners to read news; courts cannot question commissioner's decisions); S. v.
Adams, S. v. Werner, 1981 (1) S.A. 187 (A) (offense to live in white areas even though no
accommodation in Indian/colored areas and substantial inequality, unauthorized by statute,
would result). These cases are discussed in Dugard, Some Realism About the Judicial Pro-
cess and Positivism-A Reply, 98 S. AFR. L.J. 372, 384-387 (1981). See also Gumede v.
Minister of Justice, 1985 (2) S.A. 529 (N) (accepting Minister's casual "reasons" required to
be given by detention statute). Mathews says "[i]t is time for the South Africian appeal
court to review the earlier cases and modify, if not abandon, the principle that 'reason to
believe' and like clauses preclude it from any investigation whatever of the grounds for offi-
cial decisions." Supra note 27, at 205.
191. See, e.g., Komani N.O. v. Bantu Affairs Administration Board, Peninsula Area,
1980 (4) S.A. 448 (A); Ndabeni v. Minister of Law and Order, 1984 (3) S.A. 500 (D)
(Didcott, J.).
192. See, e.g., S. v. Meer, 1981 (1) S.A. 739 (N) (Shearer and Didcott, JJ.) (Orders
banning meetings under Internal Security Act held "incorrigibly obscure" and "void for un-
certainty"); In Re Dube, 1979 (3) S.A. 820 (N) (Didcott, J.); In Re Duma, 1983 (4) S.A. 469
(N) (Didcott, J.). See also Christian Science Monitor, Aug. 12, 1986, at 2, col. 3. (Natal
Supreme Court invalidates key clause in state of emergency regulations on grouds that the
government must prove that detention would help and the state of emergency).
193. E.g., S. v. Gibson, N.O., 1979 (4) S.A. 115 (N) (Milne, J.) (contempt of court and
defamation charges regarding press article critical of security police dismissed). See also,
Milne, Equal Access to Free & Independent Courts, 100 S. AFR. L.J. 681 (1983).
194. E.g., Goldberg v. Minister of Prisons, 1979 (1) S.A. 14 (A) (Corbett, J., dissenting)
(common law rights retained unless taken away by statute). See also Corbett, Human
Rights: the Road Ahead, 96 S. AFR. L.J. 192 (1979).
1986 JUDGES IN AN UNJUST SOCIETY

level, not everyone can be tainted. Distinctions that cut both ways can be
drawn. The judge's role is harder to justify than that of other actors in
the system, because the dilemma arises not only in obedience but in ap-
plication and enforcement of the laws as well. As Cover says, "the attor-
ney's role within a system of law assumed to be immoral is much easier to
justify than that of the judge."195
'
Yet if the judge hopes to work within
the system to change it, his role is potentially easier to justify. Many
judges and lawyers in South Africa have opted for such a role. The gen-
eral philosophical problem is familiar; many of us decide that participat-
ing in an organization or activity we disapprove of is worthwhile because
of the possibility of change. Is it better to do nothing at all? Sydney Ken-
tridge has put it well: "if one participates in a system that distorts justice,
truisms about the limited functions of a judge will not necessarily save
one's soul." 19 The dilemma is not just a legal dilemma; it is a moral di-
lemma. 197 As such, the choice will involve compromises between compet-
ing conceptions of what is good, or bad.

A. Apply the Law

Assuming a moral dilemma exists in South Africa, this is one resolu-

tion of the problem, but hardly a moral one. Those most affected by the
unjust laws have had no say in enacting the laws. Most judges in South
Africa, however, like most judges in the United States under slavery, will
resolve the dilemma in this way. Judges, after all, have not only been
trained in the law; they have come to live the law. It is understandably
hard for a judge to deviate from the dictates even of unjust law. An out-
sider is thus a bit presumptuous to criticize South African judges. In ad-
dition to the potent pressures pushing the judge toward role-fidelity we
have already encountered, such as positivism, parliamentary sovereignty,
and a bias toward the white status quo, rationalizations that allow the
judge to apply a law he has moral doubts about could come from various
other sources.
The political climate of South Africa is hot indeed, and the judge
may feel that the very security of the state or the order of being itself is
threatened by the slightest crack in the dike of apartheid. "Order" was a
justification for the maintenance of slavery, and it is a powerful justifica-
tion for "separate development" in South Africa. But liberty as well as
order is supposed to be guaranteed by law.
A different rationalization occasionally offered in South Africa takes
the empty antidiscrimination rhetoric of the government seriously. Ac-
cording to this view, the gap between law and morality is temporary and

195. CovER, supra note 130, at 159.

196. Kentridge, Criminal Justice in South Africa, supra note 80, at 619.
197. As Cover said of the antislavery judges, "[m]ore and more, it appeared the ques-
tion ought not to be put, 'How should a judge of integrity decide these cases?' but rather
'How can a man of integrity decide these cases?'" COVER, supra note 130 at 178.
 DEN. J. INT'L L. & POL'Y VOL. 15:1

closing. The pressure on the judge to effect change is concomitantly re-

duced. Several assumptions within this gradualist approach are subject to
question, such as whether real change is coming and whether the overall
system is thus truly benevolent. Also, gradualists tend to forget that they
are part of the system, and that an antilibertarian ruling in one case may
reverberate to retard progress in other cases. An even more tenuous ra-
tionalization available is that conservative application of repressive laws
will yield a revolutionary reaction ultimately serving justice for all. 98 '

This approach would not be seriously considered by most South African

judges actually sitting. It also suffers from a failure to recognize the posi-
tive benefits that successful litigation can achieve, quite apart from any
considerations of whether violent revolution is a moral or wise alternative
or whether applying repressive laws could realistically further that alter-
native. In any event, the conclusion that those who participate in the sys-
tem are morally responsible for that participation is hard to resist.
Evading the issue through docket control or procedural niceties only
delays the ultimate application of unjust law. Gilmore says that both
evading the issue and applying the law formalistically exalts procedure
over substance, and a "sort of Platonic" idea of law over reality. 19 9 The
reality that Gilmore refers to is the real need for tempering law with
mercy in systems like South Africa's. In the final analysis, the alternative
of applying unjust law cannot be a satisfactory solution to the moral di-
lemma, because applying the law ignores the dilemma itself.

B. Resignation

Once a judge experiences a moral crisis, the alternative of resignation

must loom large. A lively debate on the morality and utility of judicial
resignation has been carried on in South Africa for some time.2 0 0 Some
judges have actually said that resignation is the proper response to truly
unjust legislation.2 1 The basic argument for resignation is that the sys-
tem has become so pervasively unjust that remaining in it, especially in a
judicial capacity, confers legitimacy upon it. Resignation is the only hon-
orable alternative, and may have some impact.
Resignation has impeccable theoretical credentials: Gandhi called on
the judge who was condemning him to obey the law and apply the maxi-

198. This is the position of the "Celestial Curia" in the recent Foreword to the Harvard
Law Review's discussion of the United States Supreme Court's 1984 term. Bell, Foreword:
The Civil Rights Chronicles, 99 HARV. L. REv. 4, 17-20 (1985).
199. GILMORE, supra note 1, at 38.
200. See, e.g., Wacks, Judges and Injustice, 101 S. AFR. L.J. 266 (1984) (South African
legal system is unjust; judges can do little about it; judges should resign); Dugard, Should
Judges Resign ?-A Reply to Professor Wacks, 101 S. AFR. L.J. 286 (1984) (Room and
materials for liberal judgments still exist; judges cannot help by resigning, so should stay
on); Wacks, Judging Judges: A Brief Rejoinder to Professor Dugard, 101 S. AFR. L.J. 295
(1984) (Legal system repressive; resignation may have some impact).
201. E.g., S. v. van Niekerk, (Fannin, J.), quoted in Dugard, Judges, Academics and
Unjust Laws: The Van Niekerk Contempt Case, 89 S. AFR. L.J. 271, 283 (1972).
1986 JUDGES IN AN UNJUST SOCIETY

mum possible sentence, or resign. 02 Gandhi, of course, joined Socrates in

accepting the penalty bestowed by positive law. But both adopted that
position for a reason: to expose the injustice of the system. Calls for resig-
nation serve primarily to expose the moral dilemma facing the judge. If
judges are duty-bound to apply law, and the law is unjust or immoral,
then it follows that the moral thing for the judge to do is resign. This
syllogism was put to antislavery judges in America, 0 3 and it has been put
in the same form to South African judges. As Cover notes, the call for
resignation pushes judges "beyond the stage of reiteration of role defini-
tion. ' 2 0 4 When someone says "you must resign," it does not do to answer,
"but I am a judge."
The practical problem with resignation is that it is probably going to
be an ineffective means of helping those victimized by unjust laws. On the
contrary, unless massive resignations could be coordinated-a wholly un-
realistic scenario in South Africa, given the composition of the judici-
ary-moral judges' resignations would eliminate any current hope of get-
ting a moral judge at trial. Calls for judicial activism have had some
effect; calls for resignation have not. Dugard describes the situation:
All the one hundred or so judges who presently sit on the South Afri-
can Supreme Court accepted office in the knowledge that they would
be required to apply the laws of apartheid. Any isolated resignation or
handful of resignations would, therefore, inevitably be discounted and
ascribed to some ulterior motive. Only a.substantial number of resig-
nations would have the impact desired by Wacks, but such a possibil-
ity is so fanciful as to be completely eliminated. The sad truth is that
judges have missed the boat. Had they resigned in large numbers in
(say) 1955, when the Appellate Division was packed, or in 1967, when
the principle of indefinite detention without trial was first introduced,
there is little doubt that such resignations would have had considera-
ble effect. But today, absent widespread resignations in protest over
some new horrendous legislative measure, resignations would fall into
the category of 'too little and too late.'205

If the virtue of resignation is that it avoids both the violation of the

judicial oath and the application of oppressive law, the vice is that it
probably will be ineffective, because other judges will apply that law. Res-
ignation is not morally pure, either.

202. Gandhi, A Plea for the Severest Penalty Upon His Conviction for Sedition, in
THE LAW AS LITERATURE 465 (E. London ed. 1960).
203. COVER, supra note 130, at 151, citing W. PHILLIPS, A REVIEW OF LYSANDER
SPOONER'S UNCONSTITUTIONALITY OF SLAVERY (1847).
204. COVER, supra note 130, at 215.
205. Dugard, Should Judges Resign? A Reply to Professor Wacks, supra note 200, at
 DEN. J. INT'L L. & POL'Y VOL. 15:1

C. Protest or Civil Disobedience

Common to both protest and outright civil disobedience is the factor

of risk. Whether one protests in court or out of court, or engages in more
radical activity such as civil disobedience, sanctions can range from rever-
sal on appeal to criminal charges on contempt or more serious offenses.
Protest can be in court or out of court, and can be mere rationaliza-
tion designed to avoid confronting the moral dilemma or a genuine at-
tempt to spur change. Occasionally, as we saw in Moller 206 judges may
make a perfunctory homage to justice in an otherwise unjust decision.
Sometimes, as in S. v. Adams, 20 7 stern disapproval of a racially discrimi-
natory statute accompanies a decision upholding the statute. There, the
judge said "speaking for myself and if I were sitting as a court of equity, I
would have come to the assistance of appellant. Unfortunately, and on an
intellectually honest approach, I am compelled to conclude that the ap-
peal must fail. ' 20 8 Such disclaimers and hints of disapproval do little
more than highlight the injustice of the system, and not with much hope
of yielding substantial change.
The risks of protest are well-illustrated by the van Niekerk cases.210
Durban Professor Barend van Niekerk was dismayed at the diminution of
the rule of law in South Africa, and made several speeches in which he
criticized specific laws and judicial decisions in South Africa. The criti-
cism was trenchant but accurate. His first prosecution arose from an em-
pirical study of discrimination in the imposition of the death penalty. Af-
ter he was prosecuted unsuccessfully for contempt of court, he made a
speech in which he called on judges to protest the Terrorism Act, or at
least exercise their discretion in excluding evidence obtained through in-
definite detention in solitary confinement.2 10 Underlying his criticism was
a strong natural-law approach in which he asked whether there was not a
point at which unjust laws cease to be laws at all. At his second trial, van
Niekerk was convicted for attempting to obstruct the course of justice
and improperly attempting to influence judicial decisions; 211 this despite
the temperate nature of his criticism and the omission of specific refer-
ence to any pending cases.
Although van Niekerk was an academician, not a judge, his prosecu-
tion has undoubtedly stifled criticism of the judiciary from within as well
as without. In-court criticism of the system by judges may result in sanc-
tions if it is powerful and effective, and is meaningless if it is weak or
limited. In-court protests are all too often rationalizations that sidestep

206. See note 168 and accompanying text supra.

207. 1979 (4) S.A. 793 (T).
208. Id. at 801.
209. S. v. van Niekerk, 1970 (3) S.A. 655 (T); S. v. van Niekerk, 1972 (3) S.A. 711 (A).
210. Address by Prof. van Niekerk at Durban City Hall (Nov. 9, 1971), quoted in S. v.
van Niekerk, 1972 (3) S.A. 711, 716 (A.D.).
211. Id. at 711.
1986 JUDGES IN AN UNJUST SOCIETY

the moral-formal dilemma and attempt to explain the judges' failure to

do justice. Out-of-court protests are even more risky,' 1 2 but most would
agree that some point can be posited where they are necessary." 3 As Gus-
tav Radbruch said, the failure of German lawyers and judges to speak out
against the Nazi regime was a dereliction of duty.2"4 Protest is strategi-
cally limited, however. Explicit characterization of the whole system as
unjust might limit the protestor's ability to change it. Protest is also sub-
stantively limited. Unless coupled with a just result, protest achieves
little.

D. Applying Conscience Through the Judicial Lie

Applying the law, resigning, and protesting are all flawed morally in
that all may be analogized to "honor among thieves." But applying con-
science through the judicial lie is a moral use of legal power against im-
moral law. Merely applying conscience alone is a version of judicial pro-
test subject to reversal or more severe sanctions. Although the judicial lie
is also a form of protest, it is less visible and thus more likely to be effec-
tive. Dworkin thus tepidly endorsed the lie as the most attractive solution
to the moral-formal dilemma.
The difficulties with the "lie" solution are both practical and theoret-
ical. The solution has the virtue of responding to the only significant flaw
in Dugard's approach, which involves playing down the realities of op-
pressive law in South Africa via manipulation of liberal principles and
rules of construction. The lie recognizes evil legislative intent and con-
sciously mischaracterizes it. Unfortunately, unless other actors in the le-
gal system share the implicit assumption of Dugard and Dworkin that
positive law is identified in some intimate way with morality or natural
law, at least in the absence of a clear legislative statement to the contrary,
the less visible lie will not thereby be less controversial. The practical
difficulty is that the lie is perceived as dishonest, and requires "constant
battle over formal principles"' l such as jurisdiction, appellate authority,
and the bounds of the judicial and legislative functions. The theoretical
difficulty is twofold. A "lie" is a morally tainted thing, and to describe
what liberal judges in South Africa have done as a lie both misdescribes
and demeans their efforts. Thus Dworkin and Hart would not make the
"lie" option a matter of jurisprudence. Dugard denies the lie by saying
that the strange duality of South Africa's legal system, including both
liberal common-law principles and repressive legislation, allows the judge
merely to apply positive law.
While the need to infuse morality into an unjust legal system is so
great that results like those achieved by the judicial lie must be pre-

212. But see supra notes 192-194 and accompanying text.

213. See DUGARD, HUMAN RIGHTS supra note 11, at 294-5 (quoting van Niekerk).
214. See Hart, supra note 2, at 616-17.
215. COVER, supra note 130, at 198.
 DEN. J. INT'L L. & POL'Y VOL. 15:1

served, a new explanation for those results that is more responsive to the
practical and theoretical dilemmas of the lie and more acceptable to juris-
prudence is needed. The following section attempts such an explanation.

E. Law and Judging as Interpretation

The central need for the moral judge in an immoral system like
South Africa is for a jurisprudential explanation that allows the judge to
change the unjust law and make new, just law. Moreover, this need exists
in a system which does not recognize any significant power of judicial
review. As discussed in Part II above, the positivist and natural-law mod-
els are both ambivalent about the judge's law-making function. Dugard
recognizes an interstitial law-making role nominally similar to that ad-
vanced by Hart and other sophisticated positivists, but Dugard deviates
from them in denying positive-law status to evil legislative intent. With
Dworkin, Dugard would inject quasi-natural law values from this world
into the judicial process, and these are characterized as positive law. Such
approaches have an aura of illusion about them. Illusion is not necessarily
a bad thing, as we shall see. But illusion that masks what the law is to
focus on what it could be runs the risk of being empty romanticism.
An alternative approach that may never become official or popular
also makes use of illusion, but to emphasize the illusion of thinking one
can ever know what the law "is." This skeptical alternative recognizes
that the law sometimes seems clear, as when a statute exactly addresses a
situation or a case "on all fours" with the case at hand exists. Neverthe-
less, this approach emphasizes that the law is constantly in flux, never
completely certain, and always indeterminate to some degree. All cases
are hard cases. After all, even easy cases are there because someone
thought they were worth litigating, often for purposes other than delay.
Because the trend of the law, the international and domestic political cli-
mate, and other factors all bear on the current interpretation of what the
law has been and will be, this approach recognizes that the law has been
unjust and could be just.
Before the charge of "empty romanticism" is leveled once again, note
that there is evidence from both the positivistic and natural-law schools
that this approach accurately describes law and judging. Dworkin, like
Hamilton in The Federalist No. 78, speaks of the "judgment" of the judge
between competing "wills" expressed in the legislature. Hart, while
speaking of "discretion," also denies the judge power to exert "mere" will
as the determinative factor in law-making. Yet both characterize the pro-
cess of judging as essentially one of "interpretation." Interpretation sub-
sumes both "discretion" and "judgment," and more accurately describes
the judicial process. Interpretation by judges does not exceed the legiti-
mate limits of the judicial function; it is the essence of that function. In
this sense, interpretation need not be seen as "lying."
If not only judging but the law itself is interpretation, a more expan-
sive and activist conception of the judicial role is possible. Dworkin has
1986 JUDGES IN AN UNJUST SOCIETY

advanced a theory of the "law as interpretation" that envisions the law as

a description of how legal officials perform their tasks.216 He also believes
that interpretation has a prescriptive, normative side which aims at maxi-
mizing political and moral value.2'1 7 Dworkin's theory, which focuses on
judges, relates to my emphasis (one could hardly call it a theory) on law
itself as interpretation. The concepts of law and legal texts of whatever
kind are abstractions from reality before they are anything else. Like
other modes of art and language, such as religion, science, philosophy or
literature, legal structures are, at least initially, fictional structures. Such
interpretative frameworks are designed to help humankind come to terms
with experience and conceive different possibilities for the good life.
While law often crystallizes into an unjust framework for experience, the
fact that law is a fictional creation suggests that it can be malleable for
the many as well as the few, for South African blacks as well as whites.
Take the imaginative creations we call "rights," for example. Moral
lawyers in wicked legal systems are often forced to argue as if pre-ex-
isting rights should be given effect, when they (and the judges) know that,
to date, the rights have not yet "existed" in the sense of having been
recognized by the dominant legal texts subject to interpretation. Yet if
the judge, and thus the legal system, interprets the law so as to recognize
that right, the fiction will solidify and "real" existence will have been ret-
roactively conferred on the right.
This version of law as interpretation differs from that of Dworkin in
three respects other than its more flexible view of law and the rights that
are to be taken seriously. First, it recognizes the subjectivity that Dwor-
kin's legal philosophy tends to deny."' s If justice is an ideal toward which
law should strive and by which it is judged, each judge's personal sense of
justice must play a role. But justice will not come from unrestrained sub-
jectivity, or selfishness. Justice demands that judges be receptive to
points of view other than their own. The impetus toward beneficial inter-
pretation must come from a vision of law as affirmatively advancing jus-
tice, as opposed to merely mediating conflict through judicial compro-
mise. Secondly, this view recognizes the real relativity of law at any given
place or time. The reason Dworkin has problems with the wicked legal
system is that the unjust laws in it are so pervasive that, so to speak,
"something is rotten" and "time is out of joint." Moral judges and moral
lawyers in South Africa must reject any "right answer" thesis grounded in
existing law because they know that the right answer will probably vitiate
rights rather than guarantee rights in their system. Yet once the anachro-
nistic relativity of systems is recognized, the temporary analytical separa-

216. Dworkin, Law As Interpretation, in THE POLITICS OF INTERPRETATION (W.J.T.

Mitchell ed. 1983).
217. Id. at 266.
218. As when he says that judges' background moral convictions enter into a decision
but have no independent force. DWORKIN, TAKING RIGHTS SERIOUSLY, supra note 6 at 118,
124-25.
 DEN. J. INT'L L. & POL'Y VOL. 15:1

tion of "is" and "ought" can be obliterated, as it has been by the Realists,
and arguments of value as to the "right interpretation," as opposed to the
historically-determined "right answer," can play their part. Finally, this
version of law recognizes and actually celebrates the creative, law-making
role of the judge. The law-making role of the judge may be enthusiasti-
cally embraced because interpretation is always subject to constraints
such as plausibility, personality, social context and prior law. In other
words, there are both subjective and objective elements to interpretation,
as we always suspected there were. Unfortunately, activist creativity
could result in change for the worse as well as for the better; thus, the
recognition of the law-making powers of the judicial system must be al-
igned with emphasis on the moral duty of the judge to guide the system
toward moral progress.
The "instructive" function of courts, their role as "moral tutor," is
tied to their duty to promote justice. Formation of values is a social func-
tion, but it is internalized through law.219 At its best, the role is exempli-
fied by decisions such as Brown in the United States and Marechane in
South Africa.2 20 The problem in evaluative interpretation is the source of
values. The fear of such "inarticulate premises" as adherence to the white
status quo is what had prompted Hamilton and others to articulate limits
to the judicial "will." Although positivism may not result in application of
the "proper" value, and natural law may provide too many (possibly con-
tradictory) values, the place of apartheid outside any such value structure
should be clear. Most white South Africans may continue to be racist, but
there is clearly sympathy against apartheid in the country as a whole.
Even the government formally condemns racial discrimination, and thus
implicitly concedes the extent to which such condemnation approaches a
universally objective value. Although one can legitimately be skeptical
about the possibility of a consensus existing or being discernible in any
community, 221 the broad international consensus is that the institutional-
ized system of racism known as apartheid is evil. If broad consensus is the
best test for objectivity we have, anti-apartheid sentiment is one of the
most objective values in the world. For this reason, judges in South Africa
should legitimately be able to appeal to this value in evaluative interpre-
tation. Judges in the United States or other legal systems with closer
questions of just versus unjust law, and even South African judges (on
other issues), will continue to face the problem of standards for defining
laws as just or unjust. Something like Dworkin's moral principles or
Hart's policies would have to provide the answer.
The view of law as interpretation, in the meantime, does no more
than provide an explanation for just results that South African judges
may use to help solve their moral dilemma by establishing their own

219. See TUSSMAN, GOVERNMENT AND THE MIND (1977).

220. See supra notes 166-67 and accompanying text.
221. See, e.g., Brest, The Fundamental Contradictions of Normative Constitutional
Law Scholarship, 90 YALE L.J. 1063, 1109 (1981).
1986 JUDGES IN AN UNJUST SOCIETY

"Grand Style." 2 ' The results could be the same under the "lying" theory
advocated by Dworkin, or the realism-cum-value oriented approach of
Dugard. Yet the interpretative explanation reveals the flaws of describing
the creative, law-making function as a "lie" or "wrong answer" because it
reveals the excess simplicity in the view that legal "truth" is determinate.
The process of interpretation of law is only a mischaracterization of pre-
cedent when one accepts that there is an indisputably "correct" charac-
terization of all the relevant legal factors. The search for "truth" or a
right answer in South Africa will probably yield a fascist "truth." The
explanation also avoids the pitfalls of underestimating the significance of
evil laws and evil legislative intent. Recognizing that interpretation is
evaluative, the explanation urges that judges have a duty to shape the law
in the direction of justice. Of course, once the blithe assumption (at the
heart of this article) that a law is unjust is removed, and genuine debate
over the meaning of justice is reinstituted, the moral judge will have less
guidance as to how to fulfill his duty. That is why this sketch of law as
interpretation works for South Africa, but would have to respond to the
central problem of judicial review - the search for standards of value -
in order to work in the United States.
In South Africa, judges are subject to the possibility of reversal on
appeal, reactionary legislative response, or even more severe sanctions.
For this reason the jurisprudential approach to law as interpretation
should be mentioned, but the subjective aspect should not be extensively
elaborated or publicized in the judge's decisions. Although the extent to
which the creative, interpretative function should be acknowledged is a
difficult issue, the potential for repressive reaction in South Africa is so
great that detailed treatment of the process of interpretation is best left
undeveloped. This is not to say that arguments for more expansive judi-
cial review under the new South African Constitution should not be
made. But at present the chances afforded justice in already weak South
African courts should not be jeopardized. The myth of strictly objective
interpretation - the illusion referred to above - plays a necessary role
in protecting the scanty judicial independence that exists in South Africa.
Allowing judicial rhetoric denying the law-making function to con-
tinue as rhetoric, rather than making reality rise to the level of rhetoric (a
la Dworkin) or vice-versa (a la the Realists) serves a political purpose. To
my knowledge, the significance of the fact that some of the most liberal
South African judges employ the myth of objectivity in liberal decisions
has been overlooked by commentators.22 3 Yet the rhetoric of merely ob-
jective "interpretation" provides flexibility for implementing justice in

222. See K. LLEWELLYN, THE COMMON LAW TRADITION: DECIDING APPEALS (1960) (on
the "grand style" of judicial creativity in mid-19th century commercial cases, as opposed to
the "formal style" of the 1890's).
223. See, e.g., Harris v. Minister of the Interior, 1952 (2) S.A. 428 (A.D.), (Centlivres,
C.J.); Nxasana v. Minister of Justice, 1976 (3) S.A. 745, 747 (N) (Didcott, J.) (Courts consti-
tutionally powerless to veto legislation; they can only "interpret" it.)
 DEN. J. INT'L L. & POL'Y VOL. 15:1

South Africa.
V. CONCLUSION

South African judges, not the author, live within that country's sys-
tem of unjust laws. The liberal principles of the common law have been
seriously subverted by repressive laws that rigidly divide society and
often prevent the feelings that give content to justice from developing.
Consequently, many judges do not believe the system is unjust, or believe
that institutions like parliamentary sovereignty make it necessary to ac-
quiesce in the system's justice. Expanded judicial review is possible under
the new Constitution, but has yet to be asserted.
The reigning schools of Anglo-American legal thought, which have
residual influence in South Africa, give the moral judge in such a system
little guidance. Both positivism and natural law are ambivalent about the
law-making powers of the judge. Not unexpectedly, the judicial response
of twentieth-century South Africa is thus comparable to the judicial re-
sponse in America under slavery. By and large, most judges apply the
law.
Yet an undercurrent of liberal decisions has occasionally but persist-
ently flowed in the direction of justice. Without being unrealistic about
the possibilities for drastic social change through South African courts,
one can say that there are avenues available to moral judges in South
Africa. Applying the law is an evasion rather than a recognition of the
moral dilemma. Resignation not only fails to promote justice, but may
promote further injustices. Protest and civil disobedience are not only in-
effective but dangerous as well.
The judicial lie is thus the most attractive option for the moral judge.
Yet seeing libertarian results as "lies" may be counterproductive and in-
effective. Recognizing the essentially interpretative nature of law and
judging avoids these pitfalls and provides the most legitimate and valua-
ble role for the moral judge in an immoral system.