---

IN THE CONSTITUTIONAL COURT OF SOUTH AFRICA

CASE NO CCT 19/94

In the matter of

FARIEDA COETZEE

and

THE GOVERNMENT OF THE REPUBLIC OF SOUTH AFRICA
and Others


and

CASE NO CCT 22/94

In the matter of

N J MATISO
and Others

and

THE COMMANDING OFFICER, PORT ELIZABETH PRISON
and Others



Heard on: 6 March 1995

Delivered on: 22 September 1995


___________________________________________________________________________

JUDGMENT

___________________________________________________________________________

[ 1 ] KRIEGLER J: These cases raise questions concerning the constitutional validity of the provisions of sections 65A to 65M of the Magistrates Courts Act^[1] relating to the imprisonment of judgment debtors.

[ 2 ] The constitutionality of the provisions was first challenged in the Eastern Cape. Shortly after the interim Constitution^[2] came into operation, the applicant in the Matiso case, who had been imprisoned in terms of these provisions, applied to the South Eastern Cape Local Division of the Supreme Court for an order for his urgent release from the Port Elizabeth Prison. The applicant was soon followed by a number of other judgment debtors in the same predicament. The foundation of the applications was that the statutory authority of the orders committing the particular debtors to prison had been vitiated by sections 11(1) and 25(3) of the Constitution. Those subsections, it was argued, made imprisonment without a fair trial unconstitutional. Although they cited the commanding officer of the prison and their respective judgment creditors as respondents, there was no opposition.

[ 3 ] The judges who heard the applications (Melunsky and Froneman JJ) ordered the immediate release of the prisoners and referred the challenge to the constitutionality of the allegedly offending provisions of the Magistrates Courts Act to this Court.^[3] Melunsky J delivered an ex tempore judgment and Froneman J subsequently furnished detailed reasons for the order he made.^[4]

[ 4 ] Some time after the grant of the orders in the Eastern Cape the applicant in the Coetzee case applied to the Cape of Good Hope Provincial Division for similar relief, citing the Government of the Republic of South Africa, the Minister of Justice and the judgment creditor as respondents. The Court (per Van Reenen AJ) stayed committal proceedings pending against Ms Coetzee and referred the constitutional validity of sections 65A to 65M to this Court for determination.^[5] Although the formulation of the constitutional issues in the orders in the Eastern Cape case differs somewhat from that of Van Reenen AJ, the essential issue is one and the same: Is the procedure in the sections mentioned wholly or partially invalid for inconsistency with one or more of the rights guaranteed in Chapter 3 and circumscribed by section 33(1) of the Constitution?

[ 5 ] I have had the opportunity of considering the judgments prepared by my colleagues Didcott and Sachs JJ in these cases. Each of them makes quite plain why the provisions of the Magistrates Courts Act relating to the imprisonment of judgment debtors for contempt of court^[6] must be held to be invalid by reason of their inconsistency with the Constitution. Although I fully agree with that finding, my reasoning is sufficiently different to warrant separate articulation. The grounds for my conclusion are considerably narrower than those set out in the judgment of Sachs J;^[7] and there is some difference of emphasis as between Didcott J and myself.

[ 6 ] Sections 65A to 65M of the Magistrates Courts Act provide a system for the enforcement of judgment debts. Under the system a judgment debtor who has failed to satisfy the judgment debt within 10 days of the date of the judgment can be required to attend a hearing^[8] at which an enquiry will be conducted by a magistrate into the financial position of the debtor, his ability to pay and his failure to do so.^[9] The magistrate may authorise property of or debts due to the judgment debtor to be attached in settlement of all or part of the debt, or the garnishing of emoluments which will accrue to the debtor from his or her employment. The debtor can also be ordered to pay the debt in full or in instalments.^[10] The system does not end there, however. It also provides for the magistrate to issue an order to commit the judgment debtor to prison for contempt of court for failure to pay the debt.^[11] This last option of the magistrate is the issue which has given rise to the constitutional challenge.

[ 7 ] The notice to the debtor to appear at a hearing calls upon the debtor to show cause why he should not be committed for contempt of court and why the judgment debtor should not be ordered to pay the judgment debt in instalments or otherwise.^[12] The notice is drawn up by the creditor, signed by the clerk of the court and served on the debtor in accordance with the rules for service of process.^[13] The magistrate has a discretion whether to order committal to prison unless the debtor proves at the hearing that he or she 1) is under the age of 18, 2) was unaware of the original judgment for debt against him, or 3) has no means of satisfying the judgment debt. In order to show absence of means of satisfying the judgment debt the debtor also must show that such lack of means is not due to wilful disposal of goods in order to avoid payment of the judgment debt, wilful refusal to pay such debt, squandering of money or living beyond his means, or incurring of additional debts (except for household goods) after the original judgment date.^[14]

[ 8 ] On the face of it, the law seems to contemplate that imprisonment should be ordered only where the debtor has the means to pay the debt, but is unwilling to do so. However, on examination of the provisions in detail and taking notice of the actual carrying out of the provisions, it is clear that the law does not adequately distinguish between the fundamentally different categories of judgment debtors: those who cannot pay and those who can pay but do not want to. The system at issue is used most often for the collection of small debts usually of those who are poor and either illiterate or uninformed about the law or both. In the nature of things they do not enjoy legal representation. Imprisonment can and has been ordered without the debtor ever having notice of the original judgment or the notice to appear at the hearing. It can also be ordered without the uninformed or illiterate debtor having sufficient knowledge about the possibility of raising defences or the means of doing so. In the result, the provisions of the law can be used to imprison the debtor who is unwilling to pay his debt even though he has the means to do so, but can also be used (and they are indeed used) to imprison the debtor who simply is unable to pay the debt.^[15]

[ 9 ] This Court has laid down that, ordinarily, one adopts a two-stage approach for determining the constitutionality of alleged violations of rights in Chapter 3 of the Constitution. The first stage is an enquiry whether the disputed legislation or other governmental action limits rights in Chapter 3 of the Constitution. If so, the second stage calls for a decision whether the limitation can be justified in terms of section 33(1) of the Constitution.^[16]

[ 10 ] The first question this Court must answer therefore is whether any of the rights in Chapter 3 of the Constitution are limited by the relevant provisions of the Magistrates Courts Act. The parties argued with regard to the right to dignity (section 10), the right to freedom (section 11(1)) and the right to a fair trial (section 25(3)). Obviously the most fundamental right limited by imprisonment is the right to freedom. Section 11(1) of the Constitution provides:

11. (1) Every person shall have the right to freedom and security of the person, which shall include the right not to be detained without trial.

To determine whether that right is limited by the legislative provisions under scrutiny in these cases, it really is not necessary to determine the outer boundaries of the right. Nor is it necessary to examine the philosophical foundation or the precise content of the right. Certainly to put someone in prison is a limitation of that persons right to freedom.^[17] To do so without any criminal charge being levelled or any trial being held is manifestly a radical encroachment upon such right.

[ 11 ] The remaining question then is whether that limitation of the right to freedom can be justified in accordance with section 33(1) of the Constitution. That subsection, insofar as it is relevant here, provides:

33. Limitation. (1) The rights entrenched in this Chapter may be limited by law of general application, provided that such limitation
(a) shall be permissible only to the extent that it is
(i) reasonable; and
(ii) justifiable in an open and democratic society
based on freedom and equality; and
(b) shall not negate the essential content of the right in
question,
and provided further that any limitation to
(aa) a right entrenched in section ... 11 ...
...
shall, in addition to being reasonable as required in paragraph (a)(i), also be necessary.

In making the determination, especially with regard to a right as fundamental as the one in question, namely personal freedom, one really need not go beyond the test of reasonableness. This is made all the clearer by the criteria for interpretation of the Chapter 3 rights and limitations found in section 35 of the Constitution. Section 35(1) provides, inter alia:

35. Interpretation. (1) In interpreting the provisions of this Chapter a court of law shall promote the values which underlie an open and democratic society based on freedom and equality ....

Clearly that provision applies to the interpretation of both the fundamental right protected and the evaluation of any limitation according to the criteria of section 33(1). In the case of the right and limitation at issue here such interpretation is perfectly simple. At the very least a law or action limiting the right to freedom must have a reasonable goal and the means for achieving that goal must also be reasonable.^[18]

[ 12 ] I accept that the goal of sections 65A to 65M of the Magistrates Court Act is to provide a mechanism for the enforcement of judgment debts.^[19] I also accept that such goal is a legitimate and reasonable governmental objective. The question though is whether the means to achieve the goal are reasonable. In my view, the answer is clearly in the negative.

[ 13 ] The fundamental reason why the means are not reasonable is because the provisions are overbroad. The sanction of imprisonment is ostensibly aimed at the debtor who will not pay. But it is unreasonable in that it also strikes at those who

cannot pay and simply fail to prove this at a hearing often due to negative circumstances created by the provisions themselves.

[ 14 ] There are seven distinct reasons why the provisions are indefensible.

 First, they allow persons to be imprisoned without having actual notice of either the original judgment or of the hearing. It is not only theoretically possible but also quite possible in practice that the debtors first notice of the case against him is when the warrant of committal is executed. In terms of the procedure permitted by the Magistrates Courts Act and the Rules promulgated thereunder there need not necessarily be personal service of any process prior to that.^[20]

 Second, even if a person has notice of the hearing, he can be imprisoned without knowing of the possible defences available to him and accordingly without any attempt to advance any of them. The so-called notice to show cause issued pursuant to section 65A does not spell out what the defences are, or how they could be established.^[21]

 Third, the burden cast on the debtor with regard to inability to pay, although possibly defensible in principle as pertaining to matters peculiarly within his knowledge, is so widely couched that persons genuinely unable to pay are nevertheless struck.

 Fourth, the provisions of section 65F(3)(c), which spell out what the debtor must prove, are not only unreasonably wide, but also unreasonably punitive. The relevant part of the section reads as follows:

(3) No ... sentence shall be imposed ... if the judgment debtor or ... proves to the satisfaction of the court
...
(c) that he has ... no means of satisfying the judgment debts and costs either wholly or in part and that such lack of means is not due to the fact that the judgment debtor

(i) has wilfully disposed of his goods in order to defeat or delay payment of the judgment debt and costs; or

(ii) although he is able to earn sufficient to satisfy the judgment debt and costs in instalments or otherwise to pay such debt and costs, wilfully refuses to do so in order to evade or delay payment of the judgment debt and costs; or

(iii) is squandering his money or is apparently living beyond his means; or

(iv) incurred debts other than for household requirements after the judgment date.

Whatever may be said about a debtor who wilfully frustrates payment (paragraphs (i) and (ii)) the nakedly punitive retribution inherent in the provisions of paragraphs (iii) and (iv) cannot be justified.

 Fifth, the provisions allow a person to be imprisoned without knowing that he has a burden to prove her or his defence or how to discharge such burden. It could possibly be contended that the magistrate ought to explain a debtors rights and duties to an undefended layman and would probably do so. But the fact remains that there is no express obligation on the magistrate to do so.

 In the sixth instance it is hardly defensible to treat a civil judgment debtor more harshly than a criminal. The latter is entitled in terms of section 25(3) of the Constitution to a fair trial with procedural safeguards, including the right to legal assistance at public expense if justice so requires. The debtors, who face months of imprisonment, must fend for themselves as best they can.

 Lastly, the procedure makes no provision for recourse by the debtor to the magistrate or higher authority once an order for committal has been made.^[22] Section 65L, which deals with the release of a debtor from prison, contains no mechanism whereby a debtor, even one against whom a committal order had been made in absentia, is entitled to approach a court for relief.

As a result of these defects, the statute sweeps up those who cannot pay with those who can but simply will not. For this reason, the limitation cannot be justified as reasonable.

[ 15 ] This conclusion obliges one to consider the question of severability. Indeed, there are two questions to be answered with regard to the possible severance of the provisions of the law not consistent with the Constitution. First, can one excise the provisions which render the option of imprisonment unconsitutional because they do not distinguish between those who can pay but will not from those who cannot pay? If not, can the provisions which provide for imprisonment itself be severed from the rest of the system for enforcement of judgment debts?

[ 16 ] Although severability in the context of constitutional law may often require special treatment, in the present case the trite test can properly be applied: if the good is not dependent on the bad and can be separated from it, one gives effect to the good that remains after the separation if it still gives effect to the main objective of the statute.^[23] The test has two parts: first, is it possible to sever the invalid provisions and second, if so, is what remains giving effect to the purpose of the legislative scheme?

[ 17 ] In the present instance, it is not possible to excise only those provisions of sections 65A to 65M of the Magistrates Courts Act which fail to distinguish between the two categories of debtors. In order to do so this Court would have to engage in the details of law making, a constitutional activity given to the legislatures. It is, however, possible to sever the provisions which make up the option of imprisonment. The question then is whether in severing such provisions, the object of the statute will nevertheless remain to be carried out. The answer to this question clearly is yes. The object of sections 65A to 65M of the Magistrates Courts Act is to provide a system to assist in the collection of judgment debts. Removing one of the options available under the system does not render the system that remains contrary to the purpose of the legislative scheme. Accordingly, the infringing provisions can be severed and the balance of the system can usefully remain in force.

[ 18 ] In the course of argument on behalf of the Association of Law Societies,^[24] it was suggested that it would lead to a break down of the whole debt collection procedure under the Magistrates Courts Act if the imprisonment option were to be struck down immediately. Therefore, so it was argued, this Court should exercise the powers vested in it by the proviso to section 98(5) of the Constitution so as to enable the legislature to devise an adequate substitute. I do not believe that the proposal should be entertained. First, it is by no means so that the system is dependent upon the imprisonment sanction for its viability. There are a number of other aids to judgment debt collection in the system, e.g., property attachment and garnishment of wages. But even if I err in that regard the system is so clearly inconsistent with the right to freedom protected by section 11(1) and so manifestly indefensible under section 33(1) of the Constitution that there is no warrant for its retention, even temporarily.

[ 19 ] In the circumstances the following order issues:

1. The following provisions of the Magistrates Courts Act are inconsistent with the Constitution and are declared to be invalid with effect from the date of this order:

a. The following words in section 65A(1)

why he should not be committed for contempt of court and;

b. Sections 65F, 65G and 65H;
c. Paragraphs (a) and (c) of section 65J(1);
d. Paragraph b(ii) of section 65J(2);
e. The following words in paragraph (a) of section 65J(9)
(a) or,

and
and may, subject to the provisions of section 65G, be committed for contempt of court for failing to comply with the said order;
f. Paragraph (b) of section 65J(9);
g. The following words in section 65K(2)
or warrant for the committal of a judgment debtor or a director or an officer of any juristic person or of any sentence imposing a fine on any director or officer representing a judgment debtor who is a juristic person; and
h. Section 65L.

2. All other provisions of sections 65A to 65M of the Magistrates Courts Act remain in force.

3. With effect from the date of this order, the committal or continuing imprisonment of any judgment debtor in terms of section 65F or 65G of the Magistrates Courts Act is invalid.

JC Kriegler
Chaskalson P, Mahomed DP, Ackermann J, Madala J, and ORegan J concur in the judgment of Kriegler J.

[ 20 ] DIDCOTT J: I am by no means convinced at present that it would be unconstitutional, once certain conditions were met, for a debtor who had not paid the amount of a judgment duly awarded against him to be committed to prison for a limited spell like the period allowed by our current legislation or, if the judgment was satisfied before it ended, until the earlier date when that occurred. The conditions which I envisage would be ones requiring that:

(a) the creditor had already exhausted all other lawful means that were usable by him for the execution of the judgment;

(b) the committal was preceded by a full enquiry into the reasons why the debtor had failed to pay the amount that he owed, an enquiry attended by him personally and conducted in compliance with the dictates of procedural fairness by the magistrate from whom the order for his imprisonment was sought;

(c) at the enquiry the debtor had to explain the default, to disclose his financial state and affairs, and to submit to interrogation on those matters, lying largely as they did within his own peculiar knowledge;

(d) in the end, however, the creditor bore the onus to prove directly or inferentially, but positively at all events, the debtors ability in his particular circumstances to pay the amount owed and either a downright refusal by him to do so or the sheer wilfulness of his default;

(e) no order for the imprisonment of the debtor might ensue from the enquiry in the absence of such proof.

In permitting the debtor to be consigned to gaol subject to those conditions, a statutory scheme of that sort would certainly deny him, throughout his sojourn there, the right to personal freedom proclaimed by section 11(1) of the Constitution. That section 33(1) authorised the temporary denial of the right would be an arguable proposition all the same, and no less so owing to the misdescription of the grounds for it when they were artificially called a contempt of court. The denial might be viewed as a reasonable and justifiable measure, indeed as a necessary one, in a final effort to extract from a pecunious but stubbornly defiant debtor the long awaited payment to which the creditor was entitled. And it might not negate the essential content of the right, were that concept to be understood in the sense sometimes dubbed as objective which Chaskalson P and Kentridge AJ discussed in S v Makwanyane and Another 1995(3) SA 391 (CC) (paragraphs 133 at 447C-G and 195 at 470F-471B). I shall say nothing about the wisdom, expediency or efficacy of such a scheme. Nor, even on the narrower question of its constitutional validity, do I express a firm opinion. That topic is beside the point, since the scheme happens not to be the one we now have before us or, for that matter, any other in actual operation here. It has been postulated simply so that it may illustrate why I hesitate to generalise about the imprisonment of debtors, condemning that out of hand and irrespective of the way in which it is regulated.

[ 21 ] Nor, in my opinion, do we need on this occasion to indulge in such generalisations. We can dispose satisfactorily of the issue which has been referred to us without resorting to them. For the legislation that is under attack goes far beyond my imaginary scheme, doing so with no fewer than four draconian effects to which I shall confine my attention.

[ 22 ] The legislation does not, in the first place, insist on the exhaustion by the creditor of his lesser remedies before he throws the book of prospective imprisonment at the debtor. So much he may do a mere ten days after the judgment that remains unsatisfied was obtained by him, and without having taken or had the time to take any prior step in an endeavour to enforce it, by issuing a notice then which calls on the debtor to show cause to a magistrate on a date announced in it, a date as early as seven days later than the one when it was served, why the default should not be visited with committal to gaol. The magistrate is not bound, when the appointed day arrives, to send the debtor there. Some other order may be made instead, an order for the attachment of debts owed to him, or for a garnishee on his wages, or for execution to be levied against his property, or for the payability in instalments of the judgment debt. No doubt that is often done, at first anyhow, in practice and perhaps even as a matter of judicial policy. But it is not enjoined by the statute, which imposes no duty on the magistrate either to follow any of those other courses or to satisfy himself or herself that nothing will be achieved by doing so. Imprisonment is sanctioned as an initial alternative to them, not solely as a sequel to their unsuccessful pursuit.

[ 23 ] The second harsh effect of the legislation is this. It allows the debtor to be imprisoned without a hearing. The notice issued by the creditor, though served in accordance with the rules of court, may have been left with somebody else at one of the places permitted for its service and never have come to his personal attention. He may indeed be unaware of the judgment itself, the same having happened to the earlier notification of that which he was supposed to receive. He may even have known nothing about the action instituted against him which culminated in the judgment, one obtained by default because the summons that started the litigation did not reach him either. A series of accidents like those would be no surprising coincidence, after all, if the same person had accepted service of all the documents in quick succession, but neglected to pass them onto him or knew not where he was. Yet the statute expressly empowers the magistrate to sentence him to imprisonment in his absence, a fate never suffered by convicted criminals.

[ 24 ] Another explanation for the absence of the debtor, even when he has received the notice and the preceding documents, may be his ignorance of the various defences that are available to him in answering it, in particular the important defence of a poverty afflicting him which is not attributable to his own improvidence. He may labour under the misapprehension that no excuse for his failure to satisfy the judgment will be acceptable, that his imprisonment is an inescapable consequence of the default to which he must resign himself, and that his attendance at the proceedings cannot therefore accomplish anything. For the notice did not inform him of any such excuse. It was not required to do so. That is the third obnoxious effect of the statute.

[ 25 ] The fourth ugly feature of the legislation that will confront the debtor if he does appear before the magistrate, on the other hand, is the onus then resting on him to prove that he cannot pay the judgment debt and bears no blame for his impecuniosity on various grounds which are listed. He may not manage to establish that, although it is the truth, especially when his very poverty has prevented him from hiring a lawyer and he has to fend for himself in an unfamiliar environment, bewildered by procedures and a forensic methodology to which he is a stranger. The result may well be, the result must often be, that someone who really cannot pay, through no fault of his own, goes to gaol for his failure to do so.

[ 26 ] The interests of creditors are plainly relevant to any constitutional appraisal of the provisions with those effects. Credit plays an important part in the modern management of commerce. The rights of creditors to recover the debts that are owed to them should command our respect, and the enforcement of such rights is the legitimate business of our law. The granting of credit would otherwise be discouraged, with unfortunate consequences to society as a whole, including those poorer members who depend on its support for a host of their ordinary requirements. That does not mean, however, that the interests of creditors may be allowed to ride roughshod over the rights of debtors. The legislation in question permits that most egregiously, I believe, in the four respects mentioned. I am satisfied that it is unreasonable and unjustifiable on those cumulatively oppressive scores. Its clear invasion of the right to personal freedom which section 11(1) guarantees to debtors like everyone else is therefore, in my judgment, not countenanced by section 33(1).

[ 27 ] The bad parts of the statute are not judicially severable, I consider, from the rest of its provisions that deal with imprisonment. Their roots are entangled too tenaciously in the surrounding soil for a clean extraction to be feasible. The conclusion to which I accordingly come is that we are left with no option but to declare those provisions as a whole to be constitutionally invalid on account of their objectionable overbreadth.

[ 28 ] The incisive judgment prepared by Kriegler J in these two cases came to hand when the preceding parts of this one had already been written. Its thrust, as I read it, is substantially the same as mine. I agree entirely, I now add, with both the focus and
the tenor of it. For the reasons which Kriegler J and I have given, and for those reasons alone, I concur in the order proposed by him.

JM Didcott

[ 29 ] KENTRIDGE AJ: I concur in the judgment of Kriegler J and in the order which he proposes. I also agree with the identification by Didcott J of aspects of the legislation which render it unreasonable and unjustifiable. I would, however, in addition endorse the general critique of the legislation set out in paragraphs [65] to [71] of the judgment of Sachs J.

S Kentridge

[ 30 ] LANGA J: The matter referred to the Court is the constitutionality of certain of the provisions of sections 65A to M of the Magistrates Courts Act^[25] (the Act) in so far as they authorise the imprisonment of defaulting judgment debtors. Inevitably, this raised the question of whether the imprisonment of defaulting judgment debtors can ever be justifiable in an open and democratic society based on freedom and equality. It is important to make a clear distinction between what has been decided and what has not been decided in this case.

[ 31 ] Through the judgments of Kriegler J and Didcott J the Court affirms that those provisions that authorise the imprisonment of judgment debtors in sections 65A - M of the Act are unconstitutional and should therefore be struck down. Sachs J arrives at the same conclusion. I am in respectful agreement with and therefore concur in the order proposed by Kriegler J. That the relevant provisions are overbroad was common cause to all the parties who argued the matter before us. In addition, it was common cause that the provisions were procedurally flawed. Those procedural shortcomings have been crisply identified by Kriegler J at paragraph 14 of his judgment.

[ 32 ] As pointed out by Kriegler J^[26], the provisions hit two categories of defaulting debtors, namely, those who wilfully refuse to settle their debts even though they have the means and those who cannot pay because they do not have the means but who fail to prove their inability to pay. Both categories are subject to civil imprisonment. It is clear that it could never be constitutional to imprison a person who falls within the second category. What is not settled however, is whether, provided certain conditions are fulfilled, it would be unconstitutional to commit a debtor of the first category to prison. Because the impugned provisions are clearly overbroad and procedurally flawed, it is not necessary to address that question here.

[ 33 ] Although I concur with the judgment of Kriegler J, I wish to add a few comments concerning section 11(1) and its interpretation. It is trite that imprisonment, whether as a civil or criminal sanction, is a drastic curtailment of a persons liberty, which is the essence of the freedom and security provision in section 11(1) of the Constitution. In the criminal law, it is generally accepted that imprisonment should be resorted to only after the most anxious consideration. Twenty years ago Hiemstra J remarked:^[27]

The views of the Courts in regard to imprisonment have however undergone modification in the last ten years. Imprisonment is seen more and more as a harsh and drastic punishment to be reserved for callous and impenitent characters. We wish to adopt a more enlightened approach in which the probable effect of incarceration upon the life of the accused person and those near to her is carefully weighed.

Thirion J, in a later judgment observed:^[28]

Imprisonment is the form of punishment which may detrimentally affect not only the offender but also his family and his employment and because of its duration it can seldom be kept from becoming general public knowledge. It ... can have a lasting demoralising effect on the character and personality of the offender. The loss of liberty, tedium, regimentation ... which prison life entails, have a greater potentiality than a whipping for destroying the offenders self-esteem and the integrity of his character and for changing, for the worse, his way of life.

Reynolds J^[29] refers to the  deleterious effects of penal institutions ... and the unfortunate results that regularly follow the imposition of custodial punishment. Goldstone J^[30] refers to the need to ... avoid exposure to the negative consequences of imprisonment.

[ 34 ] The language of section 11(1), which guarantees freedom and security of the person and the right not to be detained without trial, is an implicit recognition and rejection of some of the practices of the past. Despite the existence of common-law provisions protecting personal freedom and security, many people were imprisoned and detained without the application of principles of procedural fairness and in circumstances where they had committed no offence which would warrant the deprivation of liberty. Thousands of South Africans each year were, for instance, imprisoned for breaches of influx control legislation after summary trials which carried few, if any, of the characteristics of a fair trial. In addition, imprisonment was also used to curtail other fundamental freedoms unjustly, including those of association, expression and belief, and, as an instrument of coercion, in order to extract information to be used for prosecutions and various other official purposes. It has therefore been a powerful weapon in the hands of officialdom. In terms of the challenged provisions, this weapon is placed at the disposal of creditors for use against defaulting debtors.

[ 35 ] The difference between the past and the present is that individual freedom and security no longer fall to be protected solely through the vehicle of common law maxims and presumptions which may be altered or repealed by statute, but are now protected by entrenched constitutional provisions which neither the legislature nor the executive may abridge. It would accordingly be improper for us to hold constitutional a system which, as Sachs J has noted, confers on creditors the power to consign the person of an impecunious debtor to prison at will and without the interposition at the crucial time of a judicial officer.^[31]

[ 36 ] For the reasons articulated in Kriegler Js and Didcott Js judgments, I agree that the impugned provisions constitute an unreasonable limitation on the freedom and security provision and that they are therefore clearly unconstitutional. In view of the conclusion I have come to in concurrence with that of Kriegler J, it is not
necessary to finally resolve the question of whether it would be unconstitutional to imprison wilfully defaulting debtors.

PN Langa

[ 37 ] SACHS J: Is imprisonment for debt in itself unconstitutional, or does it all depend on how it is done and against whom it is directed? This, to my mind, was the major issue raised in the present matter.
It was common cause amongst counsel for the Applicants and Respondents as well as the representatives of the Association of Law Societies - although their reasons differed - that the imprisonment of judgment debtors in terms of the provisions of Sections 65A to 65M of the Magistrates' Courts Act, was unconstitutional. There was no agreement, however, as to the order which they thought should be made as a result.
[ 38 ] Mr Navsa, who was briefed by the Legal Resources Centre to appear on behalf of the Applicants, argued that the provisions in question flew in the face of the international prohibition against civil imprisonment, and were so profoundly ridden with unconstitutionality, and so inextricably linked up with the remaining provisions of Sections 65A to 65M, that the whole cluster had to be invalidated.
[ 39 ] Mr Potgieter, who appeared on behalf of the Government and the Minister of Justice, accepted that the unconstitutionality was broadly-based, but said that the provisions dealing with imprisonment for alleged contempt of court could be excised without destroying the remaining portions.
[ 40 ] Mr Du Plessis, on the other hand, contended in the name of the Association of Law Societies, that the unconstitutionality rested on narrow procedural grounds, more particularly, on the lack of a hearing and a consequent violation of the well-known principle of audi alterem partem. He argued that this defect could easily be corrected by the legislature if properly directed. He agreed with Mr Navsa that the impugned provisions were so intrinsic to the scheme of Sections 65A to 65M that the whole set should be invalidated. In order to avoid a situation in which all court-supervised debt collecting became toothless and ineffective, however, he urged us to require Parliament, in the interests of justice and good government, to correct the defect in the law within a period of one year^[32]; Sections 65A to 65M should then remain in force until such correction had been made or the year had elapsed. In effect, he was arguing that the scheme for imprisoning recalcitrant judgment debtors was rescuable, and should be rescued. Implicit in the arguments of counsel for the Applicants and the Government, on the other hand, was the notion that the institution of sending non-paying debtors to jail was intrinsically beyond repair and had to be ended forthwith. It was this disagreement that has prompted my exploration of the question of whether or not imprisonment for debt is in itself unconstitutional, or, whether, properly controlled and focused, it could pass constitutional muster.
[ 41 ] A perusal of the admirably, and I might say, enviably, succinct judgments of Didcott J and Kriegler J respectively, shows that they have not found it necessary to go beyond considering the reasonableness of the procedures involved. I agree with their analysis and with the order that Kriegler J proposes. I feel however that a proper answer to the request from the Association of Law Societies that we use our powers to keep the committal proceedings alive pending rectification, requires a fuller analysis of the institution of civil imprisonment than they have considered appropriate. If there is nothing in principle constitutionally objectionable in sending people to jail for not paying their debts - as their judgments indicate or imply - then there would be considerable merit in the argument of the Association of Law Societies in favour of retaining committal proceedings pending rectification. If, on the other hand, we are dealing with an institution that is intrinsically suspect then the justification for using our powers in terms of Section 98(5) becomes weak indeed. The matter is of considerable importance not only for creditors and debtors, but for the administration of justice, inasmuch as it affects the daily work of attorneys, magistrates and prison officers. I will accordingly complement the judgments of my colleagues with some views of my own. I will start at the beginning, namely, with the nature of the right allegedly infringed, and then proceed step by step until reaching the final question of whether or not to keep the institution alive.
I THE QUESTION OF CONSTITUTIONALITY
[ 42 ] The first task is to decide whether Sections 65A to 65M are in whole or part unconstitutional. In the present case, they were said to violate the right to freedom and security of the person in Section 11, the prohibition against detention without trial in the same section, the requirements of a fair trial specified in Section 25 and the right to dignity contained in Section 10.
[ 43 ] Section 11(1) bears directly on the subject. It reads:

Every person shall have the right to freedom and security of the person, which shall include the right not to be detained without trial.

It is tempting to regard the absence of a hearing as indicating that there is a direct violation of the right in Section 11(1) not be detained without trial. Given the specific meaning that the phrase detention without trial has acquired in South Africa, however, I prefer not to apply the words literally to the situation under discussion, but rather, for the purposes of this case, to view them as protective buttresses for the broader structure of personal freedom. I feel that this approach opens the way for a richer and more sophisticated exploration of the values embodied in the concept of personal freedom, which in turn will facilitate the discovery and delineation of what could be appropriate limitations consistent with these values. It also maintains the relative impermeability of the concept of detention without trial, as generally understood; the narrower and more deeply anchored the right, and the closer it is kept to its special purpose, the more easily can it be defended against invasion.^[33] Similarly, rather than attempt to force the situation of imprisoned judgment debtors into the matrix of a criminal trial, which has different objectives^[34], I will regard Section 25 as a relevant background source which furnishes values helpful in the interpretation of the elusive notion of freedom. Thus, although Section 25 is not directly applicable to the present case in that defaulting civil debtors are neither persons arrested nor accused persons as provided for in that section, it does indicate fundamental standards of fairness regarded as appropriate before penalties, including imprisonment, are judicially imposed. I propose, also, to treat the right to dignity contained in Section 10 as a right which is intertwined with and helps in the interpretation of the rights of personal freedom and security protected by Section 11^[35], rather than as an independent right violated by the statute in question. In this way I will attempt to locate the issue in what I regard as its proper constitutional framework.
The right to Freedom and security of the person
[ 44 ] My principal focus is on the rights subsumed in the expression freedom and security of the person. The issue of determining the precise limits and content of these words will no doubt exercise this Court for a long time to come. Other jurisdictions have battled with the problem of whether the phrase should be construed as referring to one right with two facets, or two distinct, if conjoined, rights.^[36] Another jurisprudentially controversial matter has been whether the words should be considered as applying only or mainly to the absence of physical constraint^[37] or whether it should be regarded as having the widest amplitude^[38] and extend to all the rights and privileges long recognized as central to the orderly pursuit of happiness by free men and women.^[39] Even more fundamental (and even more difficult) are questions relating to the nature of citizenship and civic responsibility in a modern industrial-administrative state, the degree of regulation that is appropriate in contemporary economic and social life and the extent to which freedom and personal security are achieved by protecting human autonomy on the one hand and recognizing human interdependence on the other.^[40] The present case does not, however, compel us to penetrate into any of these complex areas. On any analysis, using any approach, there can be no doubt that committing someone to prison involves a severe curtailment of that person's freedom and personal security. Indeed, the very purpose of committal is to limit the freedom of the person concerned. Given the manifest and substantial invasion of personal freedom thus involved, the real issue that we have to decide is whether such infringement can be justified in terms of the general limitations on rights permitted by Section 33 of the Constitution. This is the nub of the problem before us.
[ 45 ] Yet the second, and for our purposes, crucial step of the investigation, is by no means unrelated to the first. Although notionally the court proceeds in two distinct analytical stages,^[41] there is clearly a relationship between the two curial enquiries. The more profound the interest being protected, and the graver the violation, the more stringent the scrutiny; at the end of the day, the court must decide whether, bearing in mind the nature and intensity of the interest to be protected and the degree to which and the manner in which it is infringed, the limitation is permissible. The President of this Court has outlined the basic balancing process in the following words:

The limitation of constitutional rights for a purpose that is reasonable and necessary in a democratic society involves the weighing up of competing values, and ultimately an assessment based on proportionality. This is implicit in the provisions of Section 33(1). The fact that different rights have different implications for democracy, and in the case of our Constitution for an open and democratic society based on freedom and equality, means that there is no absolute standard which can be laid down for determining reasonableness and necessity. Principles can be established, but the application of those principles to particular circumstances can only be done on a case by case basis. This is inherent in the requirement of proportionality, which calls for the balancing of different interests. In the balancing process, the relevant considerations will include the nature of the right that is limited, and its importance to an open and democratic society based on freedom and equality; the purpose for which the right is limited and the importance of that purpose to such a society; the extent of the limitation, its efficacy, and particularly where the limitation has to be necessary, whether the desired ends could reasonably be achieved through other means less damaging to the right in question. In the process regard must be had to provisions of Section 33(1), and the underlying values of the Constitution, bearing in mind that, as a Canadian Judge has said, the role of the Court is not to second-guess the wisdom of policy choices made by legislators.^[42]

If I might put a personal gloss on these words, the actual manner in which they were applied in Makwanyane (the Capital Punishment case) shows that the two phases are strongly interlinked in several respects: firstly, by overt proportionality with regards to means, secondly by underlying philosophy relating to values and thirdly by a general contextual sensitivity in respect of the circumstances in which the legal issues present themselves.
[ 46 ] I make these points because of what I regard as a tendency by counsel, manifested in this case, to argue the two-stage process in a rather mechanical and sequentially divided way without paying sufficient attention to the commonalities that run through the two stages. In my view, faithfulness to the Constitution is best achieved by locating the two-stage balancing process within a holistic, value-based and case-oriented framework.^[43] The values that must suffuse the whole process are derived from the concept of an open and democratic society based on freedom and equality, several times referred to in the Constitution. The notion of an open and democratic society is thus not merely aspirational or decorative,^[44] it is normative, furnishing the matrix of ideals within which we work, the source from which we derive the principles and rules we apply, and the final measure we use for testing the legitimacy of impugned norms and conduct. If I may be forgiven the excursion, it seems to me that it also follows from the principles laid down in Makwanyane that we should not engage in purely formal or academic analyses, nor simply restrict ourselves to ad hoc technicism, but rather focus on what has been called the synergetic relation between the values underlying the guarantees of fundamental rights and the circumstances of the particular case.^[45] There is no legal yardstick for achieving this.^[46] In the end, we will frequently be unable to escape making difficult value judgments, where, in the words of McLachlin J, logic and precedent are of limited assistance. As she points out,^[47] what must be determinative in the end is the courts judgment, based on an understanding of the values our society is being built on and the interests at stake in the particular case; this is a judgment that cannot be made in the abstract, and, rather than speak of values as Platonic ideals,^[48] the judge must situate the analysis in the facts of the particular case, weighing the different values represented in that context. In the present matter then, we are called upon to exercise what I would call a structured and disciplined value judgment, taking account of all the competing considerations that arise in the circumstances of the present case, as to whether in the open and democratic society based on freedom and equality contemplated by the Constitution, it is legitimate/acceptable/appropriate to continue to send defaulting judgment debtors to jail in terms of the procedures set out in Section 65 of the Magistrates Courts Act.
The Limitations Clause
[ 47 ] Section 33, commonly known as the Limitations Clause, is central to our enquiry and bears repeating:

33 (1) The rights entrenched in the Chapter may be limited by law of general application, provided that such limitation -

(a) shall be permissible only to the extent that it is -

(i) reasonable; and

(ii) justifiable in an open and democratic society based on freedom and equality; and

(b) shall not negate the essential content of the right in question,

and provided further that any limitation to -

(aa) a right entrenched in section ... 11 ...

shall, in addition to being reasonable as required in paragraph (a)(i), also be necessary.

[ 48 ] There are in fact a multiplicity of situations where the limitations clause might be invoked to justify physical restrictions on personal freedom. They were not argued before us and it would be inappropriate to express any opinion whatsoever on the validity of other proceedings presently treated by the law as permissible. They would include such matters as: detention of illegal immigrants, segregation of persons with highly infectious diseases, custodial orders in terms of mental health legislation, and arrests to establish or confirm jurisdiction of a person seeking to flee the country so as to avoid civil liability.^[49] In each case, the law limiting the exercise of the rights contained in Section 11 would have to pass the tests of reasonableness, justifiability and necessity laid down in Section 33.^[50] I will not touch the complex question of not negating the essential content of the right. Many jurisdictions, our own included, allow imprisonment of persons who fail to meet court-ordered maintenance payments.^[51] Here, too, we are not called upon to give any ruling. Nor are we called upon to make a ruling on other statutes which impose criminal liability for failure to pay monies owing.^[52] What we are required to decide is the narrow question of whether the Sections 65A to 65M procedures for the committal of non-paying judgment debtors to prison for up to ninety days are constitutionally permissible; more particularly do they meet the Section 33 criteria? Put in summary form, Section 33 requires us to ask: is the limitation reasonable, is it justifiable and is it necessary?
[ 49 ] The tests of reasonableness, justifiability and necessity are not identical, and in applying each one individually we will not always get the same results. Frequently, however, it is convenient to look at and assess them together.^[53] Normally, if a limitation fails to pass the test of reasonableness, there is no need to consider whether it could be justified or regarded as necessary; it falls at the first hurdle. My colleagues have demonstrated convincingly that on the assumption that sending defiant judgment debtors to jail was a legitimate objective, present procedures are manifestly overbroad in furthering that purpose, and as such are unreasonable and unconstitutional. As I have said, I agree with them. In the present case, however, we are required to do more than decide on the constitutionality of certain statutory provisions. We are asked to use our discretion in terms of Section 98(5) to keep constitutionally invalid provisions alive. In concrete terms, I consider this to be the real issue before us. In making our assessment, I accordingly feel it is appropriate to examine whether, even if the procedural defects could be cured, as Mr Du Plessis argued, the limitation would pass the tests of justifiability and necessity. If committal proceedings are in essence both justifiable and necessary, but vitiated merely because the means used are unreasonable in relation to the objective to be achieved, the case for giving Parliament a chance to remedy the defect is a strong one. If, however, they would fail the tests of justifiability and necessity, however well tailored, then there would be no point in attempting to correct the procedures. I will accordingly deal with the distinct criteria both separately and globally.
Reasonableness
[ 50 ] The requirement that limitation be reasonable presupposes more than the existence of a rational connection between the purpose to be served and the invasion of the right. Thus, a limitation logically connected to its objective could be unreasonable if it undermined a long established and now entrenched right;^[54] imposed a penalty that was arbitrary, unfair or irrational;^[55] or, as in this case, used means that were unreasonable.^[56] My colleagues have dealt in detail with this aspect, and I need say no more than that the procedures are manifestly unreasonable.

Justifiable in an Open and Democratic Society

[ 51 ] In deciding whether or not sending people to jail for not paying their debts is justifiable in an open and democratic society based on freedom and equality, we need to locate ourselves in the mainstream of international democratic practice.
[ 52 ] At first sight, it would appear that imprisonment for debt is totally prohibited in international law and practice. Paul Sieghart writes in a much-quoted passage that:

In the international instruments there are ... some exceptions of choice such as the freedoms from torture, slavery and imprisonment for debt, which are declared absolutely, without restriction or limitation of any kind, and not subject to derogation even in the most extreme circumstances.^[57]

Without further analysis, however this statement might be misleading. The point the author is making is that, like torture and slavery, imprisonment for debt is one of the prohibited practices in relation to which no derogation is permissible. The question that still has to be determined is exactly what is meant by imprisonment for debt; in other words, the concept or definition of imprisonment for debt can be qualified, even if its practice is absolutely forbidden. A close look at international instruments shows that far from resolving the dilemma posed in the opening sentence of this judgment, they replicate it. Thus, the American Declaration of the Rights and Duties of Man provides in broad terms that:

XXV. No person may be deprived of liberty for non-fulfilment of obligations of a purely civil character.

The American Convention on Human Rights similarly states in Article. 7(7) that:

no one shall be detained for debt. This principle shall not limit the order of a competent judicial authority issued for non-fulfilment of duties of support.

[ 53 ] On the other hand, the prohibition in the UN International Covenant on Civil and Political Rights (ICCPR), which is repeated verbatim in Protocol 4 of the European Convention, is somewhat narrower. It reads:

11. No one shall be imprisoned merely on the ground of inability to fulfil a contractual obligation.

According to the Explanatory Report on the Fourth Protocol to European Convention,^[58] freedom from civil imprisonment must be understood in the following context:

[T]he obligation concerned must arise out of contract; the prohibition does not apply to obligations arising from legislation in public or private law. Nor does the prohibition apply if the debtor acts with malicious or fraudulent intent; or if a person deliberately refuses to fulfil an obligation, irrespective of his reasons therefor, nor if his inability to meet a commitment is due to negligence. In these circumstances, the failure to fulfil a contractual obligation may legitimately constitute a criminal offence.

The aim of the Protocol was said to be to prohibit, as contrary to the concept of human liberty and dignity, any deprivation of liberty for the sole reason that the individual had not the material means to fulfil his or her material obligations.^[59] Similar points are made in connection with the ambit of Article 11 of the ICCPR, where it is stressed that the prohibition relates expressly to contractual obligations; that it does not cover deprivations of liberty based on non-fulfilment of statutory obligations, nor does it include criminal offences related to civil law debts, nor does it protect persons who simply refuse to honour a debt which they are able to pay.^[60]
[ 54 ] The only conclusion that I can draw from these materials is that international instruments strongly repudiate the core element of the institution of civil imprisonment, namely, the locking-up of people merely because they fail to pay contractual debts, but that there is a penumbra relating to money payments in which imprisonment can be used in appropriately defined circumstances.
'Necessary'
[ 55 ] By adding the requirement that limitations on Section 11 be not only reasonable and justifiable, but also necessary, the framers of the Constitution were emphasizing the status of Section 11 as one of the core provisions requiring special solicitude. It would thus not be sufficient for defenders of a renovated set of committal proceedings to show that they were reasonable and justifiable in an open and democratic society. The use of prison would also have to be sustained on the grounds that it was necessary.
[ 56 ] The element of necessity thus tightens up the scrutiny in respect of what would be reasonable and justifiable. It is a question of degree rather than of kind. Investigation of alternatives becomes more important and the tolerance given to the legislature in its choice of means to achieve reasonable objectives is reduced.^[61] The burden of persuasion is a higher one, and the balance is tipped more sharply in favour of upholding the infringed rights. Although this might not involve an onus of proof in the sense that the term is used in criminal and civil trials,^[62] it does presuppose that at the end of the day, and after having considered all argument and done its own intellectual research, the court must be satisfied that the limitation in fact meets the requirements of Section 33. Clearly, not every form of regulation or each impediment to the exercise of free choice would qualify as a violation of freedom.^[63] Yet once there is a manifest infringement of the right, as in the case of civil imprisonment, such invasion would have to satisfy the special test of being necessary.
[ 57 ] How are we to interpret the word necessary? Section 35 invites us to have regard to international experience where applicable when seeking to interpret provisions relating to fundamental rights. As I understand it, this section requires us to give due attention to such experience with a view to finding principles rather than to extracting rigid formulae, and to look for rationales rather than rules. Because of its importance and its relative novelty in South African jurisprudence, I will set out references to international instruments in some detail. The phrase 'necessary in a democratic society' appears frequently in the European Convention for the Protection of Human Rights and Fundamental Freedoms.^[64] To determine whether a particular restriction is necessary, a number of guidelines have been developed which the European Court summarized in Silver v United Kingdom^[65] as follows:

(a) the adjective 'necessary' is not synonymous with 'indispensable', neither has it the flexibility of such expressions as 'admissible', 'ordinary', useful, 'reasonable' or 'desirable'.
(b) the Contracting States enjoy a certain but not unlimited margin of appreciation in the matter of the imposition of restrictions, but it is for the Court to give the final ruling on whether they are compatible with the Convention.
(c) the phrase 'necessary in a democratic society' means that, to be compatible with the Convention, the interference must, inter alia, correspond to a 'pressing social need' and be 'proportionate to the legitimate aim pursued'.
(d) those paragraphs of Article (sic) of the Convention which provide for an exception to a right to be guaranteed are to be narrowly interpreted.^[66]

[ 58 ] The term 'necessary' is also used in the ICCPR in relation to permissible limitations on fundamental rights specified on an article by article basis. This has been interpreted to mean that a restriction is necessary only if it responds to a pressing public and social need, pursues a legitimate aim and is proportionate to that aim.^[67] It has also been stated that the requirement of necessity implies that the restriction must be proportional in severity and intensity to the purpose being sought, and may not become the rule. Unlike the European Convention, the ICCPR does not relate the element of necessity to a democratic society; accordingly, the relevant criterion for evaluating whether interference is necessary is not a common, democratic minimum standard, but rather solely whether it was proportional in the given case.^[68]
[ 59 ] The Siracusa Principles drawn up by a group of experts to guide the interpretation of the limitations clauses in the ICCPR state that:

10: Whenever a limitation is required in terms of the Covenant to be necessary, this term implies that the limitation:

(a) is based on one of the grounds justifying limitations recognised by the relevant article of the Covenant,
(b) responds to a pressing public or social need,
(c) pursues a legitimate aim, and
(d) is proportionate to that aim.

Any assessment as to the necessity of a limitation shall be made on objective considerations.^[69]

Commenting on the general use of the word 'necessary' in international instruments, Paul Sieghart says that the principle of proportionality is inherent in the adjective 'necessary'. This means, amongst other things, that every 'formality', 'condition', 'restriction', or 'penalty' imposed must be proportionate to the legitimate aim pursued.^[70]
[ 60 ] What all the above citations indicate is that the term necessary is not made the subject of rigid definition, but rather is regarded as implying a series of inter-related elements in which central place is given to the proportionality of the means used to achieve a pressing and legitimate public purpose. Turning to the South African Constitution, I will not attempt a full definition of the word 'necessary', but, bearing international experience in mind, make the following observations. The requirement that the limitation should be not only reasonable but necessary would call for a high degree of justification. It would also reduce the margin of appreciation or discretion which might otherwise be allowed to Parliament. Personal freedom would have to be regarded as a core value not lightly to be interfered with. In particular, any physical restraints imposed by State coercion would have to be looked at very closely. In lay language, a strong case indeed would have to be made out in favour of a law which allowed people to be locked up other than through the pre-trial and trial procedures provided for in Section 25. Put more technically, it would not be enough that suitably amended Sections 65A to 65M served the public interest in a rational way by enforcing legitimate claims of creditors, and using justifiable methods before to do so. The public interest served by these sections would have to be so pressing or compelling as clearly to outweigh the indignity and loss of freedom suffered by the judgment debtors, not to speak of the costs to the public purse. In negative terms, the law would not be permitted to impose restrictions or burdens going beyond what would be strictly required to meet the legitimate interests of judgment creditors and society as a whole. This is not to say that an impossibly high threshold would have to be established which effectively ruled out genuine weighing by Parliament of reasonable alternatives within the broad bracket of what would not be unduly oppressive in the circumstances.^[71] The requirement of finding the least onerous solution would not therefore have to be seen as imposing on the court a duty to weigh each and every alternative with a view to determining precisely which imposed the least burdens. What would matter is that the means adopted by Parliament fell within the category of options which were clearly not unduly burdensome, overbroad or excessive, considering all the reasonable alternatives. The question could would than have to be asked: could the societal reasons in favour of imprisonment of judgment
debtors be said to be sufficiently acute and forceful to pierce the protective constitutional armour provided by the word necessary?
Civil imprisonment or contempt of court?
[ 61 ] One justification of the necessity for retaining committal proceedings is that what we are really dealing with is not civil imprisonment at all but contempt of court. This indeed is the descriptive justification given in the texts of Sections 65A to 65M themselves for imprisonment of debtors in default. The institution of contempt of court has an ancient and honourable, if at times abused, history. If we are truly dealing with contempt of court then the need to keep the committal proceedings alive would be strong, because the rule of law requires that the dignity and authority of the courts, as well as their capacity to carry out their functions, should always be maintained. Yet are we in truth dealing with contempt of court? In answering this question it is useful to look at the context in which Sections 65A to 65M were adopted and the manner in which they have been interpreted until now.^[72] Legal history shows that Sections 65A to 65M are based on a confluence of two common law principles that were previously separate and to some extent even in conflict with each other. The first related to imprisonment for civil debt, which went back to Roman times; the second was the concept of contempt of court, in terms of which persons could be fined or committed to prison for challenging the dignity or authority of a court, usually because of defying a court order. In respect of contempt of court, the common law drew a sharp distinction between orders ad solvendam pecuniam, which related to the payment of money, and orders ad factum praestandum, which called upon a person to perform a certain act or refrain from specified action. Failure to comply with the order to pay money was not regarded as contempt of court, whereas disobedience of the latter order was. Thus, civil imprisonment for failure to pay a debt was a remedy in its own right, not dependent on proof of contempt of court. Conversely, contempt of court proceedings were not used against defaulting judgment debtors.
The purport of legislation adopted in the mid-1970's was to reverse the situation: civil imprisonment as an institution was to be abolished, while failure to pay a judgment debt was to give rise to liability to be imprisoned for contempt of court. Sections 65A to 65M, introduced into the Magistrates' Courts Act in 1976, authorized the committal to prison for contempt of court of debtors who had defaulted on judgment debts. The Abolition of Civil Imprisonment Act 2 of 1977, on the other hand, purported to get rid of civil imprisonment, though it did keep alive committal proceedings in the Magistrates Courts.^[73] Judges of the Supreme Court were, however, unconvinced either that civil imprisonment had been abolished or that the real reason why debtors in the Magistrates Courts were being committed to prison was for contempt of court. Looking at the legislative history, Van Dijkhorst J felt compelled to declare that "die daad wat strafbaar gestel word is ... die wanbetaling van die vonnisskuld" (the act that is made punishable is the failure to pay judgment debt), and that in reality civil imprisonment was re-introduced "onder die dekmantel van minagting van die hof" (under the cloak of contempt of court).^[74] In another case,^[75] the court commented that if regard was had to the wording of Section 65A(1) and 65F(1) "the so-called contempt of court is a failure to satisfy a civil judgment".^[76] In both cases, the court observed that the sections concerned made drastic inroads into the freedom of the individual and had accordingly to be interpreted restrictively rather than extensively.
[ 62 ] The mere fact that what the statute refers to as contempt of court could be considered civil imprisonment under another name, (a matter which will be discussed further below), would not, of course, per se make it unconstitutional. Nor does the judicial characterization of the law as being one that makes severe inroads into the freedom of the individual mean that such inroads could not be justified in terms of Section 33. The function of this Court is limited to declaring unconstitutionality in relation to matters properly brought before it, and then only where the legislation concerned clearly resists being construed in a manner which would save it.^[77] This latter principle does not, of course, imply the opposite, namely that fundamental rights would have to be narrowly interpreted in order to keep legislation alive. Section 232 (3) would permit a pared-down construction of legislation so as to rescue it from being declared invalid; it would not require a restricted interpretation of fundamental rights so as to interfere as little as possible with pre-existing law.^[78] Furthermore, it would not be the function of the court to fill in lacunae^[79] in statutes that might not have been visible or regarded as legally significant in the era when parliamentary legislation could not be challenged, but which would become glaringly obvious in the age of constitutional rights; the requirement of reading down would not be an authorization for reading in.
Critiques of Sections 65A to 65M
[ 63 ] Mr Du Plessis contended on behalf of the Association of Law Societies that save for one fatal defect, the procedures outlined in Sections 65A to 65M were not only not unfair, but necessary to ensure that people paid their debts and that debt-collecting was conducted in an orderly way and not through what he termed the law of the jungle. The essence of Mr Du Plessis argument can be summed up as follows: The threat of committal for a short period is not an inappropriate sanction for debtors who are able to pay, but refuse to do so. Without some penalty of this kind, the whole of debt-collecting can come to be regarded more as a matter of benign entreaty than of serious law enforcement. Worse still, strong-arm methods of debt-collecting, far more deleterious in the result than a period in prison, would inevitably follow. Far from being over-severe, a well-focused process could be quite appropriate for the objective to be achieved, namely to separate out the reprobate from the unfortunate. The correct balance between the rights of creditors and debtors would be maintained. The rule of law would be upheld. Any limitation on personal freedom that might result would be the consequence not of a harsh law, but of a conscious decision by the recalcitrant debtor to defy the court order; it would not be too drastic in the circumstances; and it would be under judicial control and function according to clearly prescribed criteria. It was reasoning along these lines which underlay Mr Du Plessis request, on behalf of the Association of Law Societies, that we exercise our discretion to keep the current debt-collecting procedure alive while Parliament remedied what he regarded as a technical and procedural defect in a well-tried, legitimate and socially-necessary legal institution.
[ 64 ] As far as counsel for the Applicants and the Government were concerned, however, the institution was intrinsically bad because it represented a continuation of civil imprisonment, under another name. In their view, it was profoundly violatory of fundamental rights in its application, and beyond repair by Parliament. For the purposes of this judgment, it is not necessary to recapitulate all their arguments or to analyse the supporting materials they made available to us. Nor is this Court obliged to make a definitive finding on whether or not the committal proceedings in Sections 65A to 65M are constitutionally retrievable or not. Yet it is appropriate to examine Mr Du Plessis arguments with some attention, since if I am convinced that his overall evaluation of the committal proceedings is correct, then I could be more easily persuaded than otherwise to accede to his request to give an order in terms of Section 98(5) which would enable the committal proceedings to be rescued by Parliament.
[ 65 ] If we look at the text not in abstract, but in its actual legal-historical setting and socio-economic context, and if we are sensitive both to its purpose and to its impact,^[80] we find strong suggestions to the effect that it does indeed represent a form of civil imprisonment in disguise, retained as a relatively quick and inexpensive means of frightening small debtors into paying up without following the procedures regarded as appropriate in the case of larger debtors. In other words, the defects might be symptomatic of a deeper unconstitutionality, so that even if each imperfect procedural detail were to be corrected, we might still be left with an unconstitutional legal institution. The picture of the operation of the provisions, as painted for us by all three counsel, was that of an institutionalized and systematic instrument of debt collecting, rather than that of a badly-tailored, yet nevertheless individualized, back-up process to deal with occasional recalcitrant and contumacious debtors; the difference between counsel was that Mr Du Plessis, in the name of the Association of Law Societies, thought the system as such was necessary and justifiable, while counsel for the Applicants and the Government thought it was not.
[ 66 ] As I have said, Sections 65A to 65M do indeed describe the penalty imposed on a defaulting debtor as being based on contempt of court, which is a well recognised legal institution of manifest virtue if properly utilized. Yet even in technical terms, there must be doubts as to whether this description is accurate. The proceedings lack the essential elements of criminal contempt of court, in that the imposition and continuation of the penalty is dependent on the will of the judgment creditor and not the court (other than through imposing the sentence).^[81] It is also doubtful whether it properly qualifies as civil contempt of court. A judgment debtor should in principle not be held liable through his or her person, life or liberty, for the payment of a debt, but only through the aggregate of his or her means. The long-standing distinction made in common law between orders ad pecuniam solvendam and those ad factum praestandum is therefore founded on logic and principle.^[82] Thus, whatever terminology may be used, we could well be dealing in reality with civil imprisonment and not with contempt of court. The essence of civil imprisonment, even in its milder forms, has always been that the debtor pays with his or her body. The Afrikaans word gyselaar (hostage) comes from the contract recognized in Roman Dutch law in terms of which a freeman pledged his person as suretyship for performance. Behind its verbal description, the committal process embodied in Section 65A can be said still to amount in practice to a form of ransom which family and friends are forced to pay to secure the release of the debtor, the only two differences being that the period is limited to ninety days, and that the State pays for maintenance rather than the creditor.^[83] Viewed historically, civil imprisonment can hardly be regarded as a tried and tested remedy deeply rooted in progressive legal tradition and necessary in a democratic society. Over the centuries and decades, its ambit has been progressively restricted so that now all that is left of it is its attenuated existence in relation to debtors hauled before the Magistrates Court; like the Cheshire cat, it has disappeared bit by bit leaving only, not a smile, but a frown. The broad question before us would be whether, in the open and democratic society contemplated by the Constitution, it could ever be appropriate to use imprisonment as a means of ensuring that creditors got paid in full, bearing in mind that the amount to be collected would often fall below the costs of collection, not to speak of the costs to the taxpayer of keeping the debtor in prison.^[84] It is evident from the statistical data presented to us^[85] that committal to prison is in reality mainly for relatively small amounts and largely for debt in respect of goods purchased, services rendered and money borrowed. Mr Du Plessis argued that the expense to be considered would not be that of sending people to prison for trifling amounts, but rather the cost of keeping the spectre of prison sufficiently alive and deterrent (afskrikwekkend) to compel the great majority of debtors to pay up. When properly examined, however, this argument seems to condemn rather than support the institution of committal proceedings, under Sections 65A to 65M. The persons most vulnerable to committal orders would be precisely those who were unemployed, and thus could not be subject to emoluments orders,^[86] and those who did not have any property which could be attached.^[87] To penalize the workless and the poor so as to frighten those a little better off would be exactly the kind of instrumentalising of human beings which the concept of fundamental rights was designed to rebut.^[88] To suggest that thousands of people would rather go to jail than satisfy relatively small debts within their capacity to pay, strains the imagination. There is thus support for Mr Navsas claim that the object of the system would be to send to jail those who could not pay in order to get money out of those who could pay. The borderline between ability to pay and refusal to pay would be a shadowy one; resigned and bewildered debtors, confused by complicated and technical notices, would inevitably get caught up with the truly recalcitrant debt-dodgers who defiantly refused to pay even when they could.
[ 67 ] Furthermore, even if the corrected law were to be overtly neutral in its language, its operational effect would to a degree be discriminatory^[89] in that the rich who did not pay their debts would in practice be dealt with in the Supreme Court by bankruptcy procedures which respected due process, while the non-paying poor would continue to be faced with summary committal in the Magistrates Court. It seems strange indeed that the lower courts, using attenuated procedures in relation to smaller debtors less able to defend themselves, would have greater coercive powers than would the superior courts using normal due process in relation to larger debtors, better able to assert their rights.^[90]
[ 68 ] Finally, we must take into account the fact that other efficacious remedies would be available to judgment creditors. It would not be easy to substantiate the existence of an imperative need to use committal orders. The civil law, in fact, would provide a series of remedies for non-payment of contractual debts. These would vary depending on the nature of the contract: repossession or holding on to goods in some cases, evictions from premises, cutting-off of services, attachment and sale of property and deduction from wages in others. Where the assets were insufficient to cover liabilities, bankruptcy proceedings could be instituted with a view both to recovering hidden assets and to ensuring appropriate distribution of what was available. The specific remedies, other than imprisonment, which Sections 65A to 65M themselves would provide, would include: sale in execution of goods; attachment of debts due; emoluments orders and an order to pay in instalments. Another section would provide for what would amount to sequestration.^[91] Furthermore, creditors could arrange different forms of security for debts, ranging from mortgages to pledges to sureties. Rather than extend credit freely and then rely on the threat of imprisonment to ensure that the debt is paid, persons could prudently calculate the risks they undertook, and then depend on normal methods of securing payment where the means for such payment existed. This need not require their denying credit to the poor, but, rather, their treating the poor with the same circumspection they would apply to the better-off.
[ 69 ] For the purposes of this judgment it is neither necessary nor desirable to make definitive findings on any of the above matters. Suffice to say that the constitutional vice at the heart of the committal proceedings cannot be identified with total assurance as being limited merely to the failure to provide a hearing, nor in my view, simply to the defects listed by Didcott J and Kriegler J. There are weighty arguments in favour of considering the institution as being more profoundly vitiated.
[ 70 ] Having rejected the minimalist position of contended for by Mr Du Plessis, however, I feel it equally necessary to refuse to accept the maximalist claims of Mr Navsa. As I have stated above, the answer to the problem of constitutionality cannot be found in an abstract, either/or decision over whether the practice in the Magistrates Court can be defined as civil imprisonment and as such automatically fall to be rejected as unacceptable (argument for both the Applicants and the Government tended to be along these lines). Rather, it would depend on an evaluation of whether, in their actual setting and operation, the provisions would involve concretely identifiable and constitutionally-indefensible invasions of the right to personal freedom. Looked at in relation to the request by the Association of Law Societies, which does not relate to constitutionality but to the appropriate order to be made, the issue presently before us is whether the institution under consideration is in itself so non-problematic and worthy of being kept alive that we should exercise our discretion under Section 98(5) in favour of this course.
[ 71 ] My conclusions, on this point, are as follows: when the Law Commission says committal of judgment debtors is an anomaly that cannot be justified and should be abolished; when it is common cause that there is a general international move away from imprisonment for civil debt, of which the present committal proceedings are an adapted relic; when such imprisonment has been abolished in South Africa, save for its contested form as contempt of court in the Magistrate's Court; when the clauses concerned have already been interpreted by the courts as restrictively as possible, without their constitutionally offensive core being eviscerated; when other tried and tested methods exist for recovery of debt from those in a position to pay; when the violation of the fundamental right to personal freedom is manifest, and the procedures used must inevitably possess a summary character if they are to be economically worthwhile to the creditor, then the very institution of civil imprisonment, however it may be described and however well directed its procedures might be, in itself must be regarded as highly questionable and not a compelling claimant for survival.
[ 72 ] This is not to say that there could never be circumstances which could justify the use of the back-up of prison to ensure that court orders for payment of judgment debts were obeyed in the same way as other orders. We are not called upon to decide this question at the moment, nor do we have sufficient material before us to make a definitive finding. The legislature, if it so chose, would be better placed than ourselves to do the requisite research, canvass opinions and receive information; it could give full consideration to relevant, inter-related factors, such as the proper management of debt collection, the way in which credit is extended, remedies for ensuring fulfilment of obligations and the proper use of court time and prison facilities. It could weigh up all the competing considerations and take account of cost implications and the availability of court and prison officials. If it chose to undertake such an investigation it would, in my opinion, have to operate within the following framework:

(i) The process should not permit the imprisonment of persons merely because they were unable to pay their contractual debts;
(ii) The procedures adopted would have to be manifestly fair in all the circumstances;
(iii) Imprisonment, involving as it does a major infringement of the right to personal freedom, would have to be the only reasonably available way of achieving the stated objectives.

II THE APPROPRIATE ORDER
[ 73 ] In the light of the above evaluation of the use of committal proceedings for non-payment of judgment debts, I proceed to answer the question raised at the beginning of this judgment, namely, whether or not this Court should use its powers in terms of Section 98(5) to keep such proceedings alive. If my overall assessment is correct, then the necessity for retaining what amounts to a sanitized form of civil imprisonment has not been established. There accordingly seems to be little reason for pressurizing Parliament into considering these questions as a matter of priority, which use of Section 98(5) powers would require it to do. The Association of Law Societies did suggest a course of action which would result in the coming into existence of such a reason. They argued that the committal procedures were so bound up with and central to the application of the remaining debt collecting provisions, that removing imprisonment and the threat of prison would lead to the collapse of the entire system. They accordingly urged us to strike down Sections 65A to 65M as a whole and, then, in order to avoid a chaotic situation from arising in the entire area of debt-collection, to use our powers in terms of Section 98(5) to put Parliament on terms to correct the defects. Committal proceedings would then continue, pending appropriate remedial action by Parliament.
[ 74 ] This raised the question of severability, namely, whether the impugned provisions could be excised from the rest of Sections 65A to 65M, or whether these sections must fall in their totality. If we were to follow the proposal of the Association of Law Societies, (surprisingly, in this respect, supported by the Applicants), then no debt-collecting procedures in the Magistrates Court would remain, and the need to exercise our life-saving discretion would indeed be great.
[ 75 ] Severability is an important concept in the context of the relations between this court and Parliament; like reading down, it is an instrument of judicial restraint which reduces the danger of producing an overbroad judicial reaction to overbroad legislation. I agree with Kriegler Js analysis of the matter, subject to one methodological qualification I feel worth mentioning. It is the following: in deciding whether the legislature would have enacted what survives on its own, we must take account of the coming into force of the new Constitution in terms of which we receive our jurisdiction, and pay due regard to the values which it requires us to promote. We must, accordingly, posit a notional, contemporary Parliament dealing with the text in issue, paying attention both to the constitutional context and the moment in the country's history when the choice about severance is to be made. It is in this context that we must decide whether the good can be separated from the bad. In the instant case, the excisions which my colleague proposes would leave a statutory provision that in my view is linguistically sustainable, conceptually intact, functionally operational and economically viable; I agree with them.
[ 76 ] Having separated the good from the bad, would it then be in the interests of justice and good government to keep the bad in existence to give it a chance to become part of the good? The words in the interests of justice and good government are widely phrased and, in my view, it would not be appropriate, particularly at this early stage, to attempt a precise definition of their ambit. They clearly indicate the existence of something substantially more than the mere inconvenience which will almost invariably accompany any declaration of invalidity, but do not go so far as to require the threat of total breakdown of government. Within these wide parameters, the Court will have to make an assessment on a case-by-case basis as to whether more injustice would flow from the legal vacuum created by rendering the statute invalid with immediate effect, than would be the case if the measure were kept functional pending rectification. No hard and fast rules can be applied. In the present case, we are dealing with one of the core values of the Constitution. As I have endeavoured to show at some length, we cannot say with confidence that all that is needed to rectify the defect in the sections concerned is a simple set of technical amendments. It is intolerable, once the unconstitutionality of imprisonment of judgment debtors has been established, that persons should continue to be detained under the impugned provisions. It has not been established that ending committal proceedings will impair justice or interfere with good government in any drastic or irreparable way. The other remedies provided for in Sections 65A to 65M remain available to creditors. There is no reason why we should insist on a rapid decision by Parliament, one way or the other, either to accept the continuance of Sections 65A to 65M in their truncated form, or else to modify them in the light of the principles enunciated by this Court. Many issues which were raised before us could be considered at the appropriate time in that forum, basing itself on the kinds of broadly-based enquiry we are not in a position to undertake: for example, whether or not the whole area should be decriminalized,^[92] or whether a procedure should be developed in terms of which failure to attend a debt enquiry hearing, or the deliberate concealment of assets, should be made criminal offences to be prosecuted in the ordinary way.^[93] Policy choices of this kind, provided they are resolved within constitutional limits, belong to Parliament, not to this Court, and it would be invidious for us to pre-empt the issue by making an order keeping the present system alive pending legislative modifications. I accordingly do not think it right to accede to Mr Du Plessis request, and for the reasons advanced above, agree fully with the order proposed by Kriegler J.

AL Sachs

[ 77 ] MOKGORO J: I have had the opportunity to read the judgments of Kriegler J, Didcott J, Langa J and Sachs J. I respectfully agree with the order proposed by Kriegler J. To the extent that he articulates the values which underlie the fundamental rights and interests at stake in the circumstances of the issue before us, I concur in the approach and conclusions of the judgment of Sachs J.

Y Mokgoro

For the Applicants in both matters:

MS Navsa SC and L Mpati instructed by the Legal Resources Centre.

For the First and Second Respondents in the Coetzee matter:

D Potgieter instructed by the State Attorney.

For the Association of Law Societies (as amicus curiae):

JC du Plessis of Du Plessis & Eksteen.

---

^[1]No. 32 of 1944. The particular sections at issue were inserted by section 2 of Act No. 63 of 1976.

^[2]Constitution of the Republic of South Africa, No. 200 of 1993. In terms of section 251 of the Constitution, the Constitution came into operation on 27 April 1995.
^[3]The provisions targeted by the order of Froneman J are:

(a) the phrase "why he should not be committed for contempt of court" in section 65(1);
(b) the whole of sections 65F, 65G, 65H and 65L;
(c) subsections (1)(c), (2)(b)(ii), 9(a) and 9(b) of s 65J; and
(d) section 65K(2).

^[4]The judgments have been reported as Matiso and Others v The Commanding Officer, Port Elizabeth Prison and Others 1994(3) BCLR 80(SE); 1994(4) SA 592 (SECLD).
^[5]The learned judge formulated the constitutional question as follows:

Are sections 65A to 65M of the Magistrates' Courts Act, No 32 of 1994, as amended, or any parts of the said Sections, invalid on the ground of their inconsistency with Sections 10, 11 and 25 of the Constitution of the Republic of South Africa Act, No 200 of 1993, or any other provision of the said Constitution?

^[6]In my view, it is not important whether the system is termed imprisonment for contempt of court for not paying a debt or civil imprisonment or some other word or phrase. The task of this Court is to determine whether the system, whatever it may be called, is or is not consistent with the Constitution.

^[7]Because I base my decision on the examination of the specific provisions of the sections at issue and not the overall concept of imprisonment for failure to pay a judgment debt, I do not find it necessary in this judgment to comment on the procedures of other countries used for the enforcement of judgment debts or the judicial decisions regarding such procedures. Nor do I find it necessary to consider the impact of the international human rights instruments so instructively canvassed by Sachs J.

^[8]See Sections 65A and 65B of the Magistrates Courts Act. The notice to the judgment debtor must be served at least 7 days prior to the hearing. Section 65B of the Magistrates Courts Act.

^[9]See Section 65D of the Magistrates Courts Act. In determining the ability of the debtor to pay, the magistrate is required to take into account the debtors and his dependants necessary expenses, other court orders to pay, and other commitments of the debtor. Section 65D(4)(a) of the Magistrates Courts Act.

^[10]See Section 65E of the Magistrates Courts Act.

^[11]See Section 65F of the Magistrates Courts Act. The magistrate may also suspend a sentence for committal. Section 65F(2) of the Magistrates Courts Act.

^[12]Section 65A(1) of the Magistrates Courts Act.

^[13]Section 65B of the Magistrates Courts Act. In accordance with the rules of service the notice need not be served personally. Rule 9 of the Magistrates Courts Rules.

^[14]Section 65F(3) of the Magistrates Courts Act.

^[15]South African Law Commission, Debt Collecting (Project 74): Imprisonment for Debt, Interim Report dated August 1994 at paragraph 4.2.2.

^[16]See, e.g., S v Williams and Others 1995(7) BCLR 861 (CC), 879D-G.

^[17]It is not necessary to address whether the rights in sections 10 and 25(3) are limited. It would only become necessary to do so should analysis of the limitation with regard to the right to freedom in accordance with section 33(1), infra, validate the provisions vis-a-vis the right to freedom. Section 10 provides - 10. Human dignity. Every person shall have the right to respect for and protection of his or her dignity. Section 25(3) provides - (3) Every accused person shall have the right to a fair trial ....

^[18]See S v Makwanyane and Another 1995(6) BCLR 665 (CC), 748A-B.

^[19]The Association of Law Societies argued as amicus curiae that the imprisonment option is defensible because putting some judgment debtors in prison coerces other debtors to pay their debts. If indeed, this is the purpose of the law, then it would fail to be consistent with the Constitution because the goal of the statute would be unreasonable. For the purposes of this judgment, we do not accept this as the purpose of the law.

^[20]Substituted service of some kind is possible in respect of all process prior to judgment. See Rule 9 of the Magistrates Courts Rules. Even where it was a default judgment, Section 65A(2) does no more than require that a notice be given by registered post. Section 65F(3)(b) renders unawareness of the original judgment a defence, but that is cold comfort to the debtor who also has no knowledge of the hearing.

^[21]See Rule 45 and Form 40 of the Magistrates Courts Rules.

^[22]Admittedly section 65F(2) contemplates subsequent suspension of a committal order but there is no procedure established for the debtor to enforce such right as the subsection may be said to afford him.

^[23]Johannesburg City Council v Chesterfield House 1952(3) SA 809 (AD), 822D-E. See also, S v Lasker 1991(1) SA 558 (CPD), 566.

^[24]Afforded an audience as amicus curiae by virtue of its special interest and expertise in the matter and represented by two members.

^[25]Act No. 32 of 1944 (as amended).

^[26]In paragraph 14 of his judgment, Kriegler J states: As a result of these defects, the statute sweeps up those who cannot pay with those who can but simply will not.

^[27]In S v Benetti 1975(3) SA 603 (T) at 605G.

^[28]This was a dissenting judgment in S v Motsoesoana 1986(3) SA 350 (N) at 372F - G. Thirion J was comparing imprisonment with corporal punishment for juveniles as sentencing options.

^[29]Reynold Js remarks, made in S v Chirara; S v Hwengwa; S v Pisaunga; S v Muzondiwa 1990(2) SACR 356 (ZH) at 358i - j, were in the context of a statement he quoted from by Ashworth in Sentencing and Penal Policy (at 318) that custodial sentences should be used as sparingly as possible.

^[30]In S v Kumalo 1984(4) SA 642 (W) at 644H.

^[31]Sachs J opines at paragraph 66 of his judgment that [A] judgment debtor should in principle not be held liable through his or her person, life or liberty, for the payment of a debt, but only through the aggregate of his or her means.
^[32]Using our powers in terms of Section 98 (5) of the Constitution, which provides that:

In the event of the Constitutional Court finding that any law or any provision thereof is inconsistent with this Constitution, it shall declare such law or provision invalid to the extent of its inconsistency: Provided that the Constitutional Court may, in the interests of justice and good government, require Parliament or any other competent authority, within a period specified by the Court, to correct the defect in the law or provision, which shall then remain in force pending correction or the expiry of the period so specified.

^[33]P Hogg, Constitutional Law of Canada (3rd ed. 1992) at Chapter 4.

^[34]See Hicks v Feiock [1988] USSC 72; 485 US 624 (1988) where the US Supreme Court re-affirmed the distinction between imprisonment for a fixed period as a punishment for doing something forbidden, and imprisonment as a flexible remedial instrument for failure to fulfil an obligation, with full due process being required for the former, but not for the latter.

^[35]See comments on interacting values by Wilson J in R v Morgentaler 44 DLR (4th) 385 (1988) at 493; See also S v Makwanyane [1995] ZACC 3; 1995 (6) BCLR 665 (CC) per Chaskalson P at 702D and 722H-723A, and ORegan J at 777E.

^[36]Hogg at 1022; Garant in Canadian Charter of Rights and Freedoms (2nd ed. 1989, eds Beaudoin and Ratushny) at 334; Re Singh and Minister of Employment and Immigration 17 DLR (4th) 422 (1985) per Wilson J at 458; R v Morgentaler, supra. The issues are discussed by Du Plessis and De Ville in Rights and Constitutionalism - The New South African Legal Order, (1994, eds Van Wyk et al) at 234 and Cachalia et al in Fundamental Rights in the New Constitution (1994) at 35.

^[37]For the tendency in Canada, see Garant supra at 342 et seq; Hogg at 1029, and also in Germany, as well as in the judgments of the European Court of Human Rights, see P Sieghart, The International Law of Human Rights (1992) at 141-42. Useful information is to be found in Du Plessis and De Ville, Rights and Constitutionalism supra at 236 and Cachalia et al supra at 35.

^[38]For the approach in India, see Kharak Singh v State of U.P. and Others [1964] 1 SCR 332; See also Maneka Gandhi v Union of India AIR 1978 SC 597 quoted in Davis, Chaskalson and De Waal in Rights and Constitutionalism supra at 46.

^[39]For the position in the US see Board of Regents of State Colleges v Roth, [1972] USSC 168; 408 US 564 (1972).

^[40]There is extensive literature on the subject which we are not compelled to explore in the present case.

^[41]S v Zuma and Others [1995] ZACC 1; 1995 (4) BCLR 401 (SA) and S v Makwanyane and Another [1995] ZACC 3; 1995 (6) BCLR 665 (CC).

^[42]Per Chaskalson P in Makwanyane supra at 708D-G.

^[43]See the warning Dickson CJC gives against a mechanical, formula-driven application of the principles in R v Oakes 26 DLR (4th) 200 (1986), and of his emphasis on the concept of a free and democratic society which, in his words, is the commonality which links the guarantee of rights and freedoms to their limitation. R v Keegstra 3 CRR (2d) 193 (1990).

^[44]In the words of Dickson CJC, in Keegstra supra, they are no mere incantation, rather, they are central to the methodology to be adopted. In the circumstances of the evolution of South African society as alluded to in the Epilogue to the Constitution, they could have special technical relevance in at least three respects: our jurisprudence has many admirable features, but has not always evolved in the direction of supporting openness and democracy, hence the need for selective utilization of decisions by our courts; the deference which courts normally give to political acts and to legislative outcomes of the democratic process, might be more tenuous in the case of decisions and legislation of the pre-democratic period; and we might be required to use a wider range of source material than traditionally has been the case. None of these issues have been argued before us, and none need to be decided for the purposes of the present case, so I express no opinion on them.
^[45]By Dickson CJC in Keegstra supra at 30 where he points out that factual circumstances shape the courts view of both the right or freedom at stake and the limit proposed by the state, neither of which should be viewed in abstract, and cites with approval the following statement by Wilson J in Edmonton Journal v Alberta AG 45 CRR 1 (1989) at 26-27.

... a particular right or freedom may have a different value depending on the context. It may be, for example, that freedom of expression has greater value in a political context than it does in the context of disclosure of the details of a matrimonial dispute. The contextual approach attempts to bring into sharp relief the aspects of any values in competition with it. It seems to be more sensitive to the reality of the dilemma posed by the particular facts and therefore more conducive to finding a fair and just compromise between the two competing values ....

and the observation of La Forest J in United States of America v Cotroni 42 CRR 101 (1989) at 117 that:

In the performance of the balancing task ... a mechanistic approach must be avoided. While the rights guaranteed by the Charter must be given priority in the equation, the underlying values must be sensitively weighted in a particular context against other values of a free and democratic society sought to be promoted by the legislature.

^[46]Per Gubbay CJ of the Zimbabwean Supreme Court:

There is no legal yardstick, save that the quality of the reasonableness of the provision under attack is to be adjudged on whether it arbitrarily or excessively invades the enjoyment of the guaranteed right according to the standards of a society that has a proper respect for the rights and freedoms of the individual.

Woods and Others v Minister of Justice, Legal and Parliamentary Affairs 1995 (1) BCLR 56 (ZS) at 59; 1995 (1) SA 703 (ZS) at 706E.

^[47]Keegstra supra at 109. How difficult this judgment is, is brought out by the fact that, applying an approach cast in almost identical terms, the majority judgment given by Dickson CJC, supported by three judges, upheld the statute while McLachlin J, supported by two judges, would have struck it down.
^[48] Trakman, Reasoning with the Charter (1991) at 201:

Rights are not self-explanatory. They are principled constructions informed by social history, communicative experience and normative practice.

^[49]In terms of Section 16(1) of the Aliens Control Act 96 of 1991 and Section 33(1)(c) and (m) of the Health Act 63 of 1977 as amended (keeping under medical surveillance and restriction of movement of persons with communicable diseases); Sections 16(1) and 19(1)(a) of the Mental Health Act 18 of 1973 as amended; Section 30 of the Magistrates Courts Act 32 of 1944 as amended provides for the arrest of persons tanquam suspectus de fuga. See also African Realty Trust v Sherman 1907 TH 34 quoted in Jones and Buckle, The Civil Practice of the Magistrates Courts in South Africa (8th ed. 1988) at 416.

^[50]In each matter, too, if litigation were to ensue, then, in my view, more than an ad hoc technical analysis of procedural fitness would be required if the correct order was to be made; sooner or later we will have to grapple with the fundamental values underlying the rights set out in Chapter 3.

^[51]In terms of Sections 11(2)(a) and 14C of the Maintenance Act 23 of 1963 as amended. Although there are some similarities with procedures under Sections 65A to 65M, there are great differences, and I wish to underline that nothing in this judgment should be seen as impinging on these sections of the Maintenance Act, which was dealt with in argument only on the basis that it was clearly distinguishable.

^[52]For example, in terms of Section 25(2) of the Basic Conditions of Employment Act 3 of 1983 as amended; Section 37 of the Wage Act 5 of 1957 as amended; Section 61(1) of the Unemployment Insurance Act 30 of 1966 as amended; and Section 50 of the Manpower Training Act 56 of 1981 as amended.

^[53]See Kentridge AJ in Zuma supra at 420A-B.

^[54]Zuma supra at 420A.

^[55]Makwanyane supra at 709E.

^[56]S v Williams and Others [1995] ZACC 6; 1995 (7) BCLR 861 (CC) at 880C.

^[57]Sieghart supra at 87, note 1.

^[58]Sieghart supra at 159.

^[59]See decision of the European Commission of Human Rights in the case of X v the Federal Republic of Germany, Case No 6699/74, given on 18 December 1971, where it was held that a provision in the German Code of Civil Procedure permitting imprisonment for up to 6 months (at the creditors expense) of debtors who refused to make an affidavit of means, did not violate Protocol 4. The question of onus of proof in relation to ability to pay was the central issue in the more recent case in the US Supreme Court of Hicks v Feiock supra which concerned imprisonment of a father for failure to pay maintenance. All members of the court agreed that if the proceedings were civil rather than criminal, then the 14th Amendment due process requirement of proof beyond reasonable doubt of ability to pay would not apply, and a legislative presumption of ability to pay would not be unconstitutional. The court divided on whether the proceedings in question were shown to be civil. The court, however, re-affirmed a long-standing distinction between imprisonment as a punishment for a limited period (criminal contempt), and purgeable imprisonment for remedial purposes to compel performance of an obligation (civil contempt), where the person concerned carried the keys of the prison in their own pockets. The leading cases cited, however, dealt with refusing to produce documents, and refusing to testify under a grant of immunity, and not with failure to pay a contractual debt. The case itself turned on failure to pay maintenance, where the obligation arose from law, not contract, and where the need to protect the interests of children was particularly compelling.

^[60]M Nowak, UN Covenant on Civil and Political Rights - CCPR Commentary (1993) at 193-6.

^[61]See ORegan J in Makwanyane supra at 780E-F.

^[62]The Canadian Charter speaks of a limitation having to be demonstrably justifiable. There is no equivalent word in Section 33, in respect of which the phrase burden of persuasion might be more apposite than onus of proof. Even here, I would be reluctant to see the fundamental rights of citizens becoming too dependent on how adroit or maladroit counsel happen to be.
^[63]See Wilson J's caution about regarding any tenuous restriction as a violation of liberty, in Thomson Newspapers v Canada [1990] 1 SCR 425 at 186. Also, her remarks in Operation Dismantle Inc. v The Queen 18 DLR (4th) 481 (1985) at 516-7. See also Garant supra at 352:

Countless standards, provisions and measures which affect the security of individual citizens are established by public authorities. Would it be necessary to see in each case an interference with or a threat to the security of the individual or corporation?

^[64]European Convention, Articles 8 to 11 and Article 2 of Protocol No. 4.

^[65][1983] ECHR 5; (1983) 5 EHRR 347 at para. 97.

^[66]An important distinction to be borne in mind is that the European Convention does not have a general limitations clause such as Section 33, but rather identifies permissible limitations on a clause by clause basis. The nature of acceptable limitations is spelt out in each clause, which makes the object of the limitation relatively easy to identify, and the application of the proportionality test a comparatively straightforward exercise. The concept of margin of appreciation also has a special meaning. It goes beyond the legitimate tolerance normally granted to the legislature to decide on matters such as budgetary priorities and the due weight to be given to competing social, moral, political and economic claims. It relates to an acknowledgment of the need to accommodate the cultural, philosophical and political diversity of the states accepting the court's jurisdiction. Robertson and Merrills in Human Rights in Europe (3rd ed. 1993) at 198-204 indicate that the width of the margin of appreciation varies a good deal. This is inevitable because situations, claims and justifications vary considerably. The margin will usually be broad if some restriction would normally be expected, or if the case presents a controversial political, economic or social issue. They point out that the cases are not always easy to reconcile, but the result is not so much an inconsistency in the Strasbourg jurisprudence, as a demonstration of a point which is fundamental to an understanding of the Convention, that decisions about human rights are not a technical exercise in interpreting texts, but judgments about political morality.

^[67]Cf Nowak supra at 211. See also Chaskalson P in Makwanyane supra at 710G-711B citing the case of R v France [1992] ECHR 40; (1993) 16 EHRR 1 and Langa J in Williams supra at 880F.

^[68]Nowak supra at 325; see also at 379 and 394 where he states that the principle of proportionality requires a precise balancing of the intensity of a measure with the specific reason for interference.

^[69]See (1985) 7 Human Rights Quarterly 1, quoted by Erasmus in Rights and Constitutionalism supra at 644.

^[70]Sieghart supra at 94; on margin of appreciation at 99-102.

^[71]See the remarks of Wilson J in Re Singh supra at 467.

^[72]The information on which the following observations are based was culled from the Report of the South African Law Commission on Committal to Prison in Respect of Debt, May 1986. The Report refers to such committal as an 'anomaly' and recommends that it be abolished.

^[73]Section 3.

^[74]Quentin's v Komane 1983 (2) SA 775 (T) at 778. See also Grosskopf JA in T`dt v Ipser 1993 (3) SA 577 (A) at 588 describing the whole process as being in effect one of civil imprisonment.

^[75]Van der Bergh v John Price Estates and Others 1987 (4) SA 58 (SE).

^[76]See also Knott v Tuck 1968 (2) SA 495 (D) at 496H; Hofmeyr v Fourie; BJBS Contractors (Pty) Ltd v Lategan 1975 (2) 590 (C) at 590-600; Erasmus v Thyssen 1994 (3) 797 (C).

^[77]Section 232(3) provides that if a restricted interpretation of the law concerned is possible, which would save it from making unconstitutional inroads into fundamental rights, then such interpretation must be favoured, even if it went against the prima facie meaning of the words in question. This section gives expression to the principle well known in other jurisdictions as reading down. Hogg points out that reading down allows the bulk of the legislative policy to be accomplished, while trimming off those applications that are constitutionally bad. See Hogg supra at 393-4. Like severance, it mitigates the impact of judicial review, but reading down achieves its remedial purpose solely by the interpretation of the challenged statute, whereas severance involves holding part of the statute to be invalid. It is still primarily the task of Parliament, not this Court, to adapt the laws of the country to the new democratic and rights-based dispensation.

^[78]See Kentridge AJ in Zuma supra at 411E-G.
^[79]See Hunter et al v Southam Inc 11 DLR (4th) 641 (1984) per Dickson J (as he then was) at 659:

It should not fall to the courts to fill in the details that will render legislative lacunae constitutional.

^[80]See cases referred to in note 89 below.

^[81]See, however, Jones and Buckle supra at 273, where the contrary position is argued.

^[82]Specific performance, which requires the person concerned to do or to refrain from doing an identified act (such as handing over a motor car or ceasing to molest someone) by its nature can only be carried out in a particular way, whereas in the case of debt, there are other means of ensuring compliance with the court order, and if these means fail because it is impossible for the debtor to perform, then there is no real contempt of court.

^[83]Cf Hofmeyr v Fourie supra at 599-600.

^[84]This could have been a factor in producing the unusual situation where it was government that asked for a simple striking down of the offensive portions of the statute, while the Association of Law Societies urged us to keep them alive pending rectification.

^[85]In the period 1977 to 1984 the number of civil summonses for debt issued each year rose from 587,000 p.a. to 666,000 p.a. while the number of committals increased sharply from 3,600 p.a. to 9,000 p. a. A random sample showed that 37% of imprisoned debtors owed less than R100, and 83% less than R500. On average, the debtors were sentenced to 31 days each, and served 9 before being released. The causes of debt were principally goods purchased (62%), professional services - mainly to doctors and lawyers (12%), money borrowed (9%) and other services (8%). Unfortunately, the detailed statistics made available to us were not up to date, but even allowing for inflation, the amounts involved would still be relatively trivial. We were informed that the number of committals increased to approximately 18,000 p.a., or, as Mr Navsa put it, two every hour. It appears, however, that in 1994, when the new Constitution came into force, the number dropped sharply to 3,700 p.a.. See the affidavit of Johan Jacob Arno Botha submitted on behalf of the Association of Law Societies.

^[86]Section 65J.

^[87]Section 65E.

^[88]Chaskalson P in Makwanyane supra at 723A; Langa J in Williams at 886F-G.
^[89]Though judgment creditors in the superior courts can and do transfer judgment debts to the Magistrates Courts for enforcement in terms of section 65M, the statistics quoted indicate that the overwhelming majority of cases are for relatively small sums. As Dickson CJC said in the Canadian case of Morgentaler supra at 408.

As is so often the case in matters of interpretation ... the straightforward reading of this statutory scheme is not fully revealing. In order to understand the true nature and scope of (the section), it is necessary to investigate the practical operation of the provisions.

See also his observation in Thomson Newspapers supra at 241:

The courts ... cannot remain oblivious to the concrete, social, political and economic realities within which our system of constitutional rights and guarantees must operate.

For the need generally to look not only at the purpose of a statute but its effect, see Pentney in Canadian Charter of Rights and Freedoms supra at 32-34. See also, White J's dissent in City of Mobile, Alabama v Bolden 446 US 55 (1980) at 102 on the importance of looking at the totality of circumstances to ensure that the 'design and impact' of a challenged legal scheme is appraised in the light of past and present reality, political and otherwise as discussed in L Tribe, American Constitutional Law (2nd ed. 1988) at 1502 et seq. The latest trend in the US Supreme Court has been the other way. See the criticism by Tribe at 1502.

^[90]The fundamental problem would seem to be that if, as was pointed out by Ackermann J in Makwanyane supra at 728G-729A, due process is almost impossible to achieve de maximis because of the severity of the outcome (capital punishment), it is equally difficult to accomplish de minimis (jail for collecting small debts), where the relative triviality of the interest involved is overwhelmed by the cumbersome machinery required for its protection, bearing in mind that jail is involved. See comment by Jansen and Brand, Civil Imprisonment, Debt Collection and Section 65 of the Magistrates Courts Act>, Centre for Human Rights Occasional Papers, No 7, 1995 at para 9.

[91]Section 74(1) of the Magistrates Courts Act 32 of 1944 as amended provides for the appointment of an administrator of a debtors estate where the debtor inter alia has insufficient assets capable of attachment to satisfy a judgment or the debtors financial obligations.

^[92]In the comment referred to in note 90, Jansen and Brand propose that the entire system be de-criminalised; that the law concentrate on effective means of attachment; and that simple and effective measures be designed to extinguish debts that cannot realistically be paid.

^[93]See South African Law Commission Further Report of August 1994.

---