Revised from my Post Graduate Diploma in Law
Dissertation, UCT (1998), Supervisor: Prof Dirk Van Zyl Smit (Dean)
[Unless otherwise
indicated, all statutory citations relate to the Criminal Procedure Act 51 of
1977]
INTRODUCTION
In respect of more serious offences, the legislature placed the
onus in bail proceedings squarely on the arrested (s60(11). On the other hand,
the implications of the approach adopted in the case law is that, regarding
less serious offences, that one should not have regard to an onus and that the
proceedings are inquisitorial with no
onus on either the state or the arrested. In other words, under no
circumstances is the onus on the State (whether the offences with which the
arrested would be charged relate to either serious or less serious offences).
This is ostensibly an anomaly in a Constitutional Democracy sporting a Bill of
Rights granting the right to bail. But their again, this right is accorded the
arrested only if the interests of justice permit. The catchword is “if”, which implies that the right to be
released on bail is contingent upon what the interests of justice might permit.
In other words, this is a qualified and not an absolute right to be released.
The
Common Law
At Common Law, when an arrested applied for bail, the onus was on them
to prove e.g., that: the interests of justice did not require their further
detention, they did not have the
propensity to commit crimes, would not abscond, endanger the safety of the public
(or any particular person), influence or interfere with witnesses, conceal
evidence, obstruct the cause of justice or, etc.
Howard J set out this position in S v Hlongwa 1979
(4) SA 112D in this an oft-quoted passage that has resonated in some of the cases
I shall discuss below (e.g.,
Ellish en Andere vs. Prokureur-Generaal
van die Witwatersrandse Afdeling 1994(2) SACR 579 (W) at 590e-f; S v Mbele and
Others 1996 (1) SACR 212 (W) at 221b-d).
The accused bears the onus of proving, on a balance of probabilities,
that if bail is granted the interests of justice will not be prejudiced by his
absconding or tampering with State witnesses.
He will have failed to discharge that onus if, on all the evidence,
there is a reasonable possibility that he would tamper with one or more State
witnesses if he were released. The
accused's past record, particularly if it includes a conviction for defeating
or attempting to defeat the ends of justice by tampering with a State witness,
is clearly relevant in this connection and, depending on the circumstances, the
court may rely also on the investigating officer's opinion that the accused
will interfere with State witnesses, if released on bail, even though his
opinion is unsupported by direct evidence. Hlongwa at 113h.
Waddington J in S v Mataboge
and Others 1991
(1) SACR 539 (B) reiterated this view:
The onus is upon the applicants for bail to show on a balance of
probability that they may safely be released. The applicants are required to
show on a balance of probabilities that the interests of justice will not be
prejudiced in the sense that it is likely that they will stand their trial, not tamper with
witnesses or otherwise interfere with the investigation of the case against
them. (at 543f)
In Aitken and Another v
Attorney-General, Zimbabwe 1992 (2) SACR 296 (ZS) Gubbay J held that “[t]he
onus is upon the accused to show on a balance of probabilities why it is in the
interests of justice that he should be freed on bail.” (at 298g). However, the Judge also recognised:
The notion that an accused is presumed innocent until proven
guilty is the cornerstone in an application for bail. Consequently, it is the
tradition of our courts to lean in favour of and not against the liberty of the
subject, and to grant bail where possible (at 298f).
The
Transformation
With the enactment of the Interim Constitution, the principles
relating to the onus of proof in bail applications that had over the years been
taken for granted by our courts, were thrown in disarray. The provision of the
Interim Constitution that caused the flurry of judicial decisions was s 25(2)(d).
Every person arrested for the alleged commission of an offence
shall, in addition to the rights, which he or she has as a detained person,
have the right . . . to be released from detention with or without bail, unless
the interests of justice require otherwise.
The right of an arrested person to be released on bail is at
present entrenched by section 35(1)(f) of
the Republic of South Africa Constitution Act (Act 108 of 1996) (final Constitution). Some decisions discussed in this paper relate to the
interpretation of s 25(2)(d) of the
Interim Constitution and others to s 35(1)(f)
of the final Constitution. However, both sections
are in essence the same. I shall therefore not embark upon an independent
discussion of each section separately. Section 35(1)(f) reads as follows:
Everyone who is
arrested for allegedly committing an offence has the right . . . to be released
from detention if the interests of justice permit, subject to reasonable
conditions)
On 29 April 1994 (a mere two days after the inception
of the Interim Constitution), it befell Froneman J to pronounce upon the burden
of proof in bail proceedings in the matter of S v Maki (1). 1994(2) SACR 630 (E) at p640c-f. As the background to his
judgment preceded the inception of the Interim Constitution, the impact of
s25(2)(d) on the burden of proof in bail proceedings could not be considered as
such. Despite this, Froneman J concluded that the burden of proof was on the
State. The Judge held that the development in our law even in the years
preceding our new constitutional dispensation had gravitated towards placing
the onus upon the person executing the arrest to prove the lawfulness thereof. Even
in the wake of the draconian laws of yore, the then highest court in our
country, the Appellate Division, held that the person who effected the arrest,
had to prove the lawfulness of thereof. In this regard Froneman J referred to Minister of Law and Order and Others v
Hurley and Another 1986 (3) SA 568 (A) at 586j-589i.
The
Judge held that the arrested person was not only relieved of having to prove
the unlawfulness of his arrest, but that the State had to prove the lawfulness
of the detention as well. In this regard the Judge referred to the opinions in Minister van Wet en Orde v Matshoba 1990 (1) SA 280 (A) and During NO v Boesak and Another 1990 (3)
SA 661 (A).
The authorities cited by Froneman J related to the rights of
persons detained under the Security Legislation of our old oppressive legal
order. However, the Judge projected the progressive values expressed in these
decisions onto the law relating to the burden of proof in bail proceedings. There
could find no valid reason why there should be a difference. After all, the
fundamental presumption of innocence extant in our legal system bolsters this argument. Granted, in bail proceedings
regard much be had to considerations of proper administration of justice, i.e.,
the importance to secure the presence of the accused at the trial and avoiding
the obstruction of justice. These
considerations do not necessarily come into play in other forms of depravation
of freedom. This means is that in given situations the burden which the State
is saddled with could possibly be discharged easier in bail applications as in
other matters. In the result, Froneman J concluded that the burden of proof was
on the State.
One of the first cases in which it fell squarely to be decided
what effect s 25(2)(d) had on the Common
law position regarding the onus in bail applications, was Magano and Another v District Magistrate, Johannesburg and Others
(1) (1994(2) SACR 304 (W). Van Blerk J held that s25(2)(d)
brought about a change in the law, i.e., that the burden was no longer on the
arrested, but the Prosecution.
The language of the section does not merely give to an accused person
the right to apply for bail, which he has under the Criminal Procedure Act 51
of 1977, but the right 'to be released from detention with or without bail'.
That right may only be denied an accused person where 'the interests of justice
require otherwise'. The use of the word 'unless' adds weight to the argument
that the onus rests upon the G State to establish that the interests of justice
require the continued detention of an accused. See, for example, R v Liebman 1906 TS 473 at 475; R v Zondagh 1931 AD 8 at 114-15. For
these reasons I am of the view that an accused person does not bear the onus to
prove that he should be released from detention, but that the State is required
to show that he should be refused such bail because the interests of justice
require it (at 306f-g).
Stegman
J in S v Mbele and Others 1996
(1) SACR 212 (W) looked
askance at this approach, labelling
it ‘revolutionary’. According to
Stegman J the mischief which s 25(2)(d)
sought to remedy was to put an end to the powers of the legislature to infringe
upon the right of a detained to approach the court for relief. The intention
was also to put an end to the power of the legislature or the executive to
erode the jurisdiction of the courts to pronounce upon the validity of the
detention. What the framers of the interim Constitution never had in mind, was
to interfere with the basic principles of the law relating to bail and particularly
the incidence of the burden of proof. The Judge held that it was incorrect to
interpret the word ‘unless’ in s
25(2)(d) as assigning law-making
powers to the judiciary. This, Stegman J held, was what Van Blerk J did in Magano. The Judge proceeded to criticise
this approach as follows:
This was, if I may be
permitted a respectful comment, a revolutionary step in the law relating to
bail. By assuming law-making powers, a member of the judiciary reversed the
incidence of the well-established onus of persuading a court in a bail
application. In my respectful view there was no sufficient reason to believe
that the intention of the framers of the interim Constitution was to
revolutionise the settled law relating to bail in that way, or to invest the
judiciary B with such law-making powers. On the contrary,
it is clear to me that the interim Constitution is based upon the doctrine of
the separation of the powers of Legislature, Executive and Judiciary; that,
leaving aside the special position of the Constitutional Court, the judiciary
has not been invested with legislative powers; and C that, by means of s 25(2)(d), the framers of the Constitution wished to protect and entrench
the rights of all persons under the ordinary and settled law relating to bail
by precluding the Legislature or the Executive from preventing anyone from
enjoying the benefit of those rights (save for such legislative adjustments as D would not conflict with the Constitution).
That could only be achieved by ensuring that the power to decide on bail always
remains within the exclusive jurisdiction of judicial officers. That is so
obvious that it hardly needs to be spelt out (at 234i-235d).
The
Counter-Revolution
During 24 and 27 April 1994 (the run-up the first democratic general
election) a group of right wing extremists unleashed a spate of bomb explosions
in the then Transvaal. Many people died, many more were seriously injured, and
damage to the tune of R75 000 000-00 was caused. Upon their arrest, the accused
applied for bail in the Regional Court and their application was heard between
15 and 20 May 1994 by Regional Magistrate C J van Heerden. If by their deeds
the accused had intended to cause a revolution, they had failed. But what they
could not have foreseen when they had hatched their vicious plots, were that
their court battles to be released on bail would unleash a battle of another
kind, i.e., the debate as to whether or not the concept of ‘onus’
had the right of existence in our law relating to bail. This was the backdrop
to two significant judgments delivered by the Witwatersrand Local Division: Prokureur-Generaal van die Witwatersrandse
Afdeling v Van Heerden en Andere (1994(2) SACR 467
(W)) and Ellish
en Andere v Prokureur-Generaal van die Witwatersrandse Afdeling (1994(2) SACR 579 (W)).
But the first volleys in support of the approach adopted in the Van Heerden and Ellish-cases, had already
been fired prior to these judgments in the works of at least two influential
writers: Hiemstra, Suid-Afrikaanse Strafproses
(5th
ed. by J Kriegler),1994, Butterworths)and M.
G. Cowling (Law Reform: Bail
and the search for a new theoretical approach (some comments and suggestions on
the Law Commission's working paper). The
views of these authors questioning the
justification for having a niche in our law relating to bail for a concept such
as a burden of proof, were favourably rererred to in the Van Heerden and Ellish decisions.
The procedure is essentially inquisitorial in character, the presiding judicial
officer (who cannot just sit back and receive whatever evidence the parties
wished or were able to tender) to direct the proceedings and to ensure that
sufficient information is placed before the court so that a proper decision can
be reached. At times corroborating evidence on, e.g., an accused's personal
background or previous convictions might have to actively seek. This inquisitorial-type
procedure renders the question of onus irrelevant although in more general
terms the State has to produce valid reasons why release should be denied. If,
at the end of the weighing-up and balancing process such reasons are found not
to exist, the accused is entitled to be released. (For Hiemstra, see
the Van Heerden-case at 479g,, as well as Ellish-case
at 585 e-f); for Cowling, see the Ellish-case
at 585a-d).
Eloff J in Van Heerden held
that the reason why regard should not be had to the concept of onus is that
bail proceedings were not criminal but judicial proceedings. The question of
onus is important in criminal proceedings against an accused. It not at all has
a comparable role in an investigation into the advisability to release an
arrested person (at
479e).
It was held, however, that the duty to initiate proceedings rested
on the prosecution. The reason was that if the State wished to oppose bail, it
is only logical and fair that the prosecutor should motivate and substantiate
its reasons for its opposition by initiating the proceedings. If the State
presented evidence that was of such a nature that an explanation on the part of
the accused were required, the accused could then adduce evidence in response
to that of by the State. If the accused failed to adduce evidence in rebuttal,
the court could draw a negative inference. Viewed thus, Eloff J held, there
existed a burden of rebuttal, the nature of which had to be shaped by practical
and equitable considerations. (Van Heerden-case (supra) at 480d-e).
Then the Attorney-General of the Witwatersrand Provincial
Division, having succeeded in his appeal to have the judgment of Regional
Magistrate Van Heerden overturned in the Van
Heerden-case, came up against an appeal brought by the accused. The Full Bench
in the Ellish-case (supra) dismissed
the appeal. In the majority judgment Schalkwyk J (Mynardt J
concurring), held that bail proceedings
were sui generis in nature, e.g.,
evidence could be presented on an informal basis, it could be presented by
means of hearsay or documentary evidence
and an accused could motivate his application by means of affidavit.
The test applied to each an
every bail application was, however, aimed at probable future conduct of the
detained person i.e., whether he would stand his trial, whether there is a
likelihood that he would interfere with State witnesses or obstruct the course
of justice, whether their is a likelihood that he would commit offences while
awaiting trial, etc. The court held that in the past, as well as at present, no
bail application could be held without giving due consideration to one or more
of these factors. It was the duty of the presiding officer, the issue of onus
notwithstanding, to ensure that justice prevailed. The presiding officer had to
satisfy him or herself that the right of the arrested to be released should be
weighed up against the interests that the community to protection. This, Van
Schalkwyk J held, could not be achieved when a presiding officer was a mere
onlooker F (at 585f-
586b).
Like Eloff J in Van Heerden,
Van Schalkwyk held that it was incumbent upon the state to initiate the
proceedings. If there were an equilibrium between the interests of the accused
on the one hand and the interests of justice on the other, then the accused
would be entitled to be released. This right, Van Schalkwyk J held, emanated
from the provisions of s 25(2)(d),
and was not the result of the application of an onus. (at 591f-g)
In a minority judgment, Southwood J agreed that the appeal should
be dismissed, but disagreed with the majority of the court that s 25(2)(d) did not create an onus that the
state had to discharge. Having adopted
the approach that the accused had a right to be released on bail in terms of s
25(2)(d) (unless it is proved that
the interests of justice required otherwise), Southwood J proceeded as follows:
It cannot be expected of the arrested person who has a right to be
released from detention with or without bail to prove that his release is not
contrary to the interests of justice or that the interests of justice do not
require his release. This must be shown by the person who, or authority which
seeks his continued detention in the face of such right. This is the only
reasonable construction which can be placed on the wording of the subsection itself. (at 595i-j).
Southwood
J agreed with the view of the majority of the court that it is incumbent upon
the prosecution to initiate proceedings. However, the Judge was of the opinion
that once the State is called upon to lead evidence first, it has a crucial
impact on the whole question as to whether or not s 25(2)(d) created an onus:
I agree that if at the end of the enquiry the scales are evenly balanced
then the person arrested is entitled to be released with or without bail and
that the State must lead evidence first in bail proceedings and this is based
on the proper construction to be placed on the subsection. However, this construction is crucial to the issue which had to be
decided, i.e. whether or not the subsection creates an onus or not. (my
emphasis). (at 596c-d).
Southwood
J held that in a bail application, the question as to whether or not a person
is entitled to be released is of vital interest both to the arrested as well as
the State and both parties were entitled to lead evidence. The fact that bail
proceedings were sui generis and
inquisitorial in nature did not affect the fact that the presiding officer may,
on the evidence presented, find that the further detention of the accused would
be in the interests of justice. To make use of an onus to resolve this issue
would be both practical and juridically sound (see 596d-e).
Southwood J furthermore held that the use of an onus
would not change the nature of the bail proceedings. Historically, the courts
accepted that the onus was on the arrested and this did not have any effect on
the inquisitorial nature of those
proceedings. With the Interim Constitution in force, there was no justification
to believe that the court will now cease to play the role it played prior of
the implementation of the Constitution merely if the onus were shifted to the State. The court only
had to continue the active role it had played over the years prior to the
Constitution:
G As long as the court bears in mind that it is not required to
simply play a passive role, the use of an onus will not result in any I injustice. In any event even on the approach
of Van Schalkwyk J injustices may arise if the court simply plays a passive
role in bail proceedings. I fully agree with Van Schalkwyk J that a court
hearing an application for the release of a detained person must always bear in
mind that its task is to ensure that justice is done (at 596h-i).
Leveson J in the Mbele-case
criticised the judgment of Van Schalkwyk in Ellish
for concluding that bail proceedings were inquisitorial in nature. According to
Leveson J there were no juridical foundation for his conclusion. The Judge held
that, ‘[n]o power was given in the
Constitution Act to the judicial officer to act as inquisitor and as far as my
researches have taken me, nowhere at the time of the Ellish judgment, or since,
has it anywhere been laid down that the proceedings are inquisitorial’ (at 217b).
In dealing with the practical implications of abolition of the
onus in bail applications, Leveson J drew attention to the resemblance opposed
bail proceedings have to actual trials. While the Judge conceded that a court
is called upon to give judgment on ‘the probable future conduct of the detained
person’ (Worded as follows by Van Schalkwyk J
in the Ellish-case (supra) at 585f: ‘die waarskynlike
toekomstige optrede van die aangehoudene.,
he was not convinced that the task of the presiding officer is more onerous
than that performed in any other trial. Even the minority judgment of Southwood
J came in for criticism for concluding that s 25(2)(d) had shifted the onus on to the State.
There are averments to be proved or disproved, as the E case may be, and when there is doubt there is
no reason why the matter should not be resolved by resorting to the expedient
of onus. It is true also that the court is required to form an opinion as to a
man's future conduct. In this connection there is no reason to expect a
judicial officer to be more skilled in clairvoyance than any other person. But
even with future prospects the court must attempt to judge them in F relation to present facts. There has to be
determination of an issue as in any other trial. I am not persuaded that the
factors considered by the learned Judge in the Ellish case carry any weight, first, because there is nothing in s
25(2)(d) which invites their
application, and second, because there was no reason to overthrow at least 60 years of decisions in which it
had been held that the onus lay on the G accused. The correctness of the view of the
majority of the Court is thus open to doubt. For the record it should be added
that in the minority judgment of Southwood J it was said that the onus lay on
the State. With that view I obviously disagree (at
217d-g).
The judgments of Leveson J and Stegman J in S v Mbele were all embracing and relatively lengthy. It dealt with the
common law, it dealt with, and criticised, the ‘revolutionary step’ taken by Van Blerk AJ in S v Mgano, and then proceeded to criticise both the majority and
minority judgments in Ellish. Although Ellish was a full bench decision to which Stegman J conceded, ‘I respectfully acknowledge this Court to be
bound’ (at
237a), both Judges were
nevertheless highly critical of, and rejected the conclusions reached in Ellish.
I have already mentioned the criticism by Stegman J of the Magano-case and the label given to it
i.e., ‘revolutionary’ (supra). This is how Stegman J referred to the judgment of the full
bench in Ellish:
However, the revolutionary
step taken in Magano's case resulted E in what I would respectfully characterise as a
counter-revolutionary step taken in Prokureur-generaal
van die Witwatersrandse Plaaslike Afdeling v Van Heerden en Andere 1994 (2)
SACR 469 (W), and by the majority of a Full Court in Ellish en Andere v Prokureur-generaal, WPA 1994 (2) SACR 579 (W). (at 236d-e).
In Mbele, the Magano-case
was adjudged ‘revolutionary’ by
Stegman J for the following reasons: the law prior to the Interim Constitution
was to the effect that an arrested person in a bail application had the onus to
persuade the court that it would be in the interests of justice to release him;
this, it was held in the Magano-case,
to be in conflict with s 25(2)(d) of
the Interim Constitution. The onus rested on the State. Then Ellish-case, in turn, was pronounced ‘counter-revolutionary’ for holding that
‘a bail application is to be conducted without either the applicant or the
respondent having any burden to persuade the court about how the interests of
justice will best be served.‘ (at p236g).
Under the law as stated before
27 April 1994, replete with the wisdom of two or three generations of
Judges, an applicant for his release with or without bail pending his trial
would fail unless he succeeded in persuading a court that the interests of
justice which favoured the protection of his liberty outweighed the interests
of justice which would be put at risk by his release (at 236i-j).
With the pronouncement of the Ellish-decision,
‘that accumulated wisdom has ceased to
reflect the law’ — at least according to Stegman J (at p237a).. The Judge gave the
following scenario of the practical implications if no onus were applied, and
where the presiding officer took full charge of the proceedings:
The applicant is no longer required to persuade the court to
release him. Neither is the Attorney General or his representative, as the respondent,
B required to persuade the court not to release
him. It is all left to the magistrate or Judge who is to decide the issue. He
must take the initiative and conduct an inquisitorial proceeding. If the first
stages of his inquiry reveal a more or
less equal balance between those interests of justice which favour the release
of the applicant with or without bail, and those interests of justice which do
not, C the inquisitor is presumably required to keep
on digging until his enquiries satisfy him one way or the other. As the present
case has shown, the effective discharge of the role of inquisitor or D investigator is going to make it difficult in
many cases to preserve a judicial impartiality, or even the appearance of it (at 237a-d).
Legislating
and the Burden of Proof
In 1995 legislation was introduced (s 60(11) to make it more
difficult for those charged with serious
offences to be granted bail―if at all. Subsequently the Legislature again
amendment the Criminal Procedure Act (the Criminal Procedure
Second Amendment Act 85 of 1997) with a view to
tightening up our bail law even further. What the legislation in effect did,
was to place the onus on the arrested to satisfy the court that the interests
of justice did not require their further detention. This amendment came into
operation as from 1 August 1998. Some all the authorities cited in this paper
revolve around the interpretation of the original s 60(11). However, the
amended s 60(11) does not impact upon the incidence of the burden of proof.
Only the manner in which the burden should be discharged is being affected
i.e., the arrested person must discharge the onus by adducing evidence.
Therefore, where the old s 60(11) is referred to in passages from decisions
quoted below, the interpretation given to the old measure should, with the
necessary changes in mind, be projected upon the new one.
Section 60(11) as amended by
Act 85 of 1997 now reads as follows:
Notwithstanding any provision of this Act, where an accused is
charged with an offence referred to-
(a) in Schedule 6, the
court shall order that the accused be detained in custody until he or she is
dealt with in accordance with the law, unless the accused, having been given a
reasonable opportunity to do so, adduces evidence which satisfies the court
that exceptional circumstances exist which in the interests of justice permit
his or her release;
(b) in Schedule 5, but not
in Schedule 6, the court shall order that the accused be detained in custody
until he or she is dealt with in accordance with the law, unless the accused,
having been given a reasonable opportunity to do so, adduces evidence which
satisfies the court that the interests of justice permit his or her release.
When the effect of s60(11) on the incidence of proof in bail
proceedings fell to be decided by our courts, Mbele once again featured
prominently. In fact, the effect of s60(11) on the incidence of proof in bail
proceedings was the ratio decidendi of
Mbele. It was only in the course of
interpreting this subsection that the Judges saw it fit to comment on the full bench
decision of Ellish to which they
conceded they were bound anyway. As we have seen, Leveson J and Stegman J (even
without the assistance of this recent legislative directive) were adamant that
s25(2)(d) of the Interim Constitution
had not effected a shift of the burden of proof from the State to the accused.
Given this legislative support, therefore, the Judges had no difficulty to
conclude that the new legislation squarely put the onus on the arrested person.
Thus spoke Leveson J:
There is perhaps a tendency to avoid the use of the expression 'burden
of proof' I because a bail application is not a criminal
proceeding and because the court is not required to weigh proved facts but to
speculate on future conduct on the basis of information laid before it. But it
is clear from the cases cited in the judgment of Van Schalkwyk J in the Ellish case and the cases cited herein
that 'onus' is a well-known concept in bail proceedings. I do not think there
is any purpose in being coy about the use of the word or that courts should
shrink from using it simply because of the nature of the proceeding. If a
statute casts upon a party the duty of satisfying the court on a particular
point then whatever language is used he has an onus. And, for the purpose of a
bail application, the accused has the onus of
J
establishing that the granting of bail is in the interests of
justice and for that limited purpose the word must have the meaning given in Pillay v Krishna and Another 1946 AD 946
at 952, aptly quoted in the minority judgment of Southwood J in the Ellish case.
In my opinion the current legislation squarely places the question
of onus where it B has always belonged - on the accused - . . . (at 220i-221b).
The incidence of proof, having been assigned thus, Stegman J held
that, as far as the facts of the case were concerned, it had been impermissible
for the magistrate to act inquisitorially.
The magistrate, having applied the principles laid down in the Ellish-case, had misdirected himself in
doing this, because Ellish had been
decided before s60(11) was enacted.
The Ellish-case only applied to that
category of cases that did not fall within the ambit of the offences with which
s60(11) was concerned. Stegman J said
the following:
8. In the
present matter the appellants were charged with offences referred to in the new
Schedule 5 of Act 51 of 1977. The magistrate held the consequence to be that
the new
9.
s 60(11) imposed on them the onus of satisfying him that the interests
of justice did not require their detention in custody. In adopting this
approach the magistrate was undoubtedly correct. I
9. Relying on Ellish's
case the magistrate acted inquisitorially or investigatively, and in so doing
he descended for a period into the arena of conflict between the appellants and
the State.
In my judgment this was
an error. Ellish's case was decided
before the new s 60(11) placed an onus
upon a bail applicant who falls within its provisions. Ellish's case continues to govern bail applications that are not
regulated by s 60(11). But where s 60(11) is applicable, the applicant for bail
undoubtedly has an onus to discharge; there is no room for the application of
the decision in Ellish's case; and
there is accordingly no warrant for any inquisitorial or investigative action
on the part of the judicial officer hearing the application (at 237i-238a).
It is now generally accepted by our Courts that s 60(11) places
the burden of proof on the arrested to satisfy the court that the interests of
justice do not require their further detention. (Compare: S v Vermaas 1996 (1) SACR 528 (T) per van Dijkhorst J at p528h; S v Shezi 1996 (1) SACR 715 (T) per Els J at p718a).
The
Standard of Proof
The fact that the s 60(11) placed the onus on the arrested, did not
mean that this burden is now more onerous than it had been before s 60(11) came
into operation. The arrested only has to prove that the interests of justice did
not require their further detention. Traverso J in S v Stanfield 1997 (1) SACR 221
(C) (at 226g-h) held that a test
higher than this would render the onus almost impossible for the person to
discharge and therefore justiciable on constitutional grounds.
If the standard of proof to be applied is the same as it had been prior
to the enactment of s 60(11), then the views expressed in the judgments mentioned
above should apply. This standard is ‘proof
on a balance of probabilities’, not ‘proof beyond reasonable doubt’. See, for example, the judgment of Howard J in S v Hlongwa (supra) which is to the effect that ‘[t]he accused
bears the onus of proving, on a balance of probabilities, that if bail is
granted the interests of justice will not be prejudiced by his absconding or
tampering with State witnesses.’ See
also the passage by Waddington J in S v
Mataboge and Other (supra) that ‘The onus is upon the applicants for bail to show on a balance of
probability that they may safely be released.”
The
Constitutionality of Section 60(11)
The judgments
in Mbele once again feature
prominently during the debate on the constitutionality of s 60(1). As I have
already pointed out, Stegman J adopted the approach that s
25(2)(d) had nothing whatsoever to do
with the incidence of any risk of persuasion. The mischief that the framers of
the Constitution had intended to address, was the removal from our legal system
of the draconian legislation which provided for the detention of persons by the
Executive with impunity, whilst placing prohibitions the arrested person to approach
a court of law to apply for his release.
Thus spoke
Stegman J:
What emerges from this
discussion is the important point that s 25(2)(d) of the interim Constitution does not deal with the E question of onus in a bail application at all.
There is therefore no question of any such conflict between s 25(2)(d) of the interim Constitution and the
new s 60(11) of Act 51 of 1977 as Mr. Stander
postulated in his argument. This would remain the F position even if Ellish's case were one day to be reconsidered. On any basis, s
25(2)(d) of the interim Constitution
does not deal with the question of onus at all, expressly or by implication.
On the other hand, the
new s 60(11) plainly and unambiguously places an onus upon certain accused
persons to satisfy a court G that those interests of justice which favour
their release pending a trial, outweigh the interests of justice which are
adverse to their release (at 237d-i).
In Mbele Leveson J also
held that s 60(11) did not infringe upon the principles embedded in s 25(2)(d). The judge expressed the view that s
25(2)(d) merely essentially recited
the right of the individual to his freedom; there were no prescription for any
procedural provisions for the determination of that right. The subsection did prohibit the introduction of legislation
placing the burden on either the State or the arrested. Being neutral in
character, this subsection neither empowered nor prohibited (at 218b). Leveson
J was. however, alive to the scope and breath of the constitutional debate
unfolding around him.
It is abundantly clear that the section contains a direction that,
pending the outcome of the proceedings and order that might follow, the accused
is to be detained in custody. However he
is given the right to satisfy the court that the interests of justice do not
require his detention. Clearly that
implies that he must initiate the application and be given a hearing although
there is a provision that in a case where the accused is not represented the
magistrate may invite him to apply for bail.
On the face of it a direction that the accused be detained in custody
appears to be in conflict with s 25(2)(d) which prescribes that a detained
person is to be released with or without bail. But I think it more proper
to read the whole section as being adjectival, i.e. as simply requiring that in
order to be released from detention the accused must satisfy the court that it
is in the interests of justice that this be done. (my
emphasis) (at p218h-219a
per Leveson J)
The judgment
of Leveson J (at 218h-219a)
in
Mbele did not appeal to Edeling J in Prokureur-Generaal, Vrystaat v Ramokhosi 1997 (1) SACR 127 (0) (at 152g) either. Edeling J pointed out that Leveson J
conceded (at 218j of
Mbele) that s 60(11) of the
Criminal Procedure Act was on the face of it in conflict with s 25(2)(d) of the Interim Constitution. Edeling
J, however, criticised Leveson J for brushing this conflict off “as simply requiring that in order to be
released from detention the accused must satisfy the court that it is in the interests
of justice that this be done”. C
Edeling J was
of the opinion that s 60(11) is in direct conflict with both s 60(1)(a) as well
as s 25(2)(d) of the Interim
Constitution and that the issue had best be pronounced upon by the
Constitutional Court. The Judge acknowledged that he could not pronounce upon
the constitutionality of s 60(11). He was, therefore bound to work on the
assumption that s 60(11) placed the burden of proof upon the arrested (at
153c-g).
In S v Shezi 1996 (1) SACR 715
(T) Els J was also of the opinion that s
60(11) was not in conflict with s 25(2)(d).
The reason advanced was that s 60(11) was enacted in order to make it more
difficult for certain people who are under arrest and who are charged with
having committed certain serious offences, more difficult to get bail (at 717e).
SUMMARY
The debate on the burden proof in bail proceedings, is confusing.
Van Blerk AJ in Magano, was of
opinion that s 25(2)(d) of the
Interim Constitution reversed the burden of proof i.e., the State is required
to show that an arrested should be refused such bail because the interests of
justice required this. Eloff J in Van
Heerden held that one cannot refer to a concept of ‘onus’. The reason is that neither
the State nor the arrested had the onus to prove that the interests of justice
required further detention or release (as the case may be). The proceedings were
inquisitorial.
The burden to initiate the proceedings, however, rested on the
State. On appeal both Van Schalkwyk J and Mynhardt J in Ellish upheld the decision of Eloff J in Van Heerden. In the minority judgment Southwood J disagreed with
the finding of the majority that the onus should be dispensed with; the Judge
held that the onus rested on the State. He did, however, endorse the view of
the majority that the proceedings were inquisitorial.
Both Judges in Mbele,
disagreed with each and every opinion outlined above. It was held that Magano was incorrectly decided, because
s 25(2)(d) did not shift the onus to
the State. The criticisms levelled at both Van
Heerden and Ellish were that historically
there had not been justification for such an approach: bail proceedings were
inquisitorial and it is furthermore not necessary have regard to a burden of
proof. Although obiter, the judgment in the Mbele
has highlighted the strands of opinion that constitute the slippery nature
of the burden of proof in bail applications. The legislation that was introduced
added another strand, complicating the debate even further. Initially there had
been uncertainty as to the constitutionality of s 60(11). In Mbele and Shezi it was held that s 60(11) was not in conflict with s25(2)(d) of the Interim. Edeling J in Ramokhosi (an orbiter opinion) was of the opinion that s 60(11) was in conflict
with the Constitution and justiciable on that ground.
Where an arrested is charged with an offence not listed in either Schedules
5 or 6 and in respect of which the ‘reverse
onus’ provision of s 60(11) therefore does not apply, there is no onus
either on the state the arrested, because the proceedings would be
inquisitorial. Under these circumstances, the law as laid down in Ellish is applicable—it was a full bench
Decision and thus far the most authoritative constituted court to have
considered the issue. There is, In other words, no burden of proof; although
the prosecution has to initiate proceedings. The proceedings are inquisitorial
and the presiding officer plays an active role throughout. At the end the court
will merely weigh up the quality of the evidence adduced—mindful of course that
there is no onus and neither party receives the benefit in the event of doubt.
Where an arrested is charged with a serious offence listed in
Schedules 5 or 6, the ‘reverse onus’
provision of s 60(11) applies, i.e., they must prove that their further
detention would not be in the interests of justice by adducing evidence that (exceptional circumstances existed which in) the
interests of justice permit their release.” In respect of schedule 5 offences, only the
requirement outside the brackets applies; with regard to schedule 6 offences,
the bracketed text also forms part of the test.
CONCLUSION:
A FootNote’s Worth
The Constitutional Court in S
v Dlamini; S v Dladla and Others; S v Joubert; S v Schietekat 1999 (2) SACR
51 (CC) all but settled the onus debate as regards s60(11), upholding its
constitutionality. But in all fairness to Kriegler J, let me hasten to add that
the Constitutional Court Judge actually gave a wide berth to the term “onus” as such, euphemistically stating
that there is “a burden cast upon an applicant
for bail”. [par 45 note 74―quoted more fully below]. Be this as it may, let
me first deal with Kriegler’s J’s endorsement of the constitutionality of s60
(11) through the following excerpts:
[78] .
. . It was not suggested that the imposition of an onus on an applicant for
bail is in itself constitutionally objectionable, nor could such a submission
have been sustained. This Court has in
the past unhesitatingly struck down provisions that created a reverse onus carrying the risk of conviction despite
the existence of a reasonable doubt; but what we have here is not a reverse
onus of that kind. Here there is no risk
of a wrong conviction, the objection that lies at the root of the unacceptability
of reverse onuses. All that the
subsection does in this regard, is to place on an accused, in whose knowledge
the relevant factors lie, an onus to establish them in a special kind of
interlocutory proceeding not geared to arriving at factual conclusions but
designed to make informed prognoses.
[79] .
. . Section 60(11)(a) therefore does not create an onus where nothing of the
kind existed before. It describes how it
is to be discharged, and adds to its weight.
As in the case of reliance on any other right in the Bill of Rights, if
accused persons wish to rely on the provisions of s 35(1)(f), they must bring
themselves within its ambit. The words
“interests of justice permit” form part of the definition of this right; they
delineate its ambit. The court must be
satisfied that “the interests of justice permit” the release from
detention. Where all the relevant facts
are common cause, the matter is decided by the presiding judicial officer
exercising a value judgment according to all the relevant criteria on the basis
of these facts in the manner described in this judgment. If facts indispensable for establishing that
the interests of justice permit the arrestee’s release are not established, the
arrestee is not entitled to the remedy under the subsection.
On the other hand, perhaps because the question of onus (or “burden cast upon an applicant for bail”
for that matter) with regard to offences not governed by s60(11) had not been
up for consideration as such, Kriegler J (à
la Pontius Pilate) all but washed his hands off this particular aspect―relegating
it to a mere footnote.
I avoid using the word “onus” in
order not to get involved in the debate (see eg Ellish en Andere v Prokureur-Generaal, Witwatersrandse Plaaslike
Afdeleing 1994 (4) SA 835 (W); 1994 (2) SACR 579 (W); S v Vermaas 1996 (1) SACR 528 (T); S v Tshabalala 1998 (2) SACR 259 (C); S v Mbele and Another 1996 (1) SACR 212 (W); S v Shezi 1996 (1) SACR 715 (T)) mentioned at 246A-B of the
judgment in Schietekat. For the present it is unnecessary to resolve
the question whether there is an onus in bail proceedings and, if so, its
incidence. The current cases are governed
by sub-s (11), where there is undoubtedly a burden cast upon an applicant for
bail. [par 45 footnote 74]
So let me conclude that the law on the discussion at hand is that there
is “a burden cast upon an applicant for
bail” in respect of Schedule 5 and 6 offences. However, as for offences not
governed by s60(11) the full bench opinion as laid down in Ellish prevails, i.e., there is
not an onus on either the prosecution or the arrested and the proceedings are
inquisitorial.
