Willene Holness
 BA, LLB (Rhodes), LLM (UKZN), LLD (UP)
 Senior Lecturer at the School of Law, University of KwaZulu-Natal, Howard College campus. Member of the Navi Pillay Research Group
  This email address is being protected from spambots. You need JavaScript enabled to view it.

Immense thanks to Dr Janine Hicks for her critical input on earlier drafts of this paper and also to Dr David Barraclough for his technical editing. All errors are my own.

 Edition: AHRY Volume 5
 Pages: 377-400
 Citation:  W Holness ‘Decriminalising vagrancy offences in Africa beyond the ACHPR’s Advisory Opinion: quo vadis?’ (2021) 5 African Human Rights Yearbook 377-400
 Download article in PDF


The African Court on Human and Peoples’ Rights (African Court) advisory opinion in Pan African Lawyers Union on the compatibility of vagrancy laws with the African Charter on Human and Peoples’ Rights and other human rights instruments in Africa (PALU opinion) 1/2018 recommends that states amend or repeal vagrancy offences that criminalise the life-sustaining activities of many persons targeted by these laws. The Court’s main opinion is that states’ national laws and by-laws containing vagrancy offences are vague, overly broad and ambiguous, conferring broad discretion on the police for their enforcement and targets people in terms of their status, not their conduct. These laws violate articles 2, 3, 5, 6, 7, 12 and 18 of the African Charter on Human and Peoples’ Rights; articles 3 and 17 of the African Charter on the Rights and Welfare of the Child; and article 24 of the Protocol to the African Charter on the Rights of Women in Africa. The main opinion insufficiently addresses the colonial root of vagrancy. A separate concurring opinion elaborates on the margin of appreciation and good faith principle that allow states to formulate context-specific answers to amend and repeal vagrancy offences. This article identifies the impact of the COVID-19 pandemic on petty offences enforcement and draws lessons from litigation such as Gwanda v The State and interventions from national human rights institutions such as the South African Human Rights Commission; as well as guidance from the African Commission on Human and Peoples’ Rights Principles on Decriminalisation of Petty Offences in Africa (2017). It is argued that specific institutional and non-material interventions are needed to bring the systemic change required, in tandem with the banning of vagrancy offences, for cities that promote inclusion and sustainable development for all.


Dépénalisation des délits de vagabondage en Afrique au-delà de l’Avis consultatif de la Cour africaine: quo vadis?

RÉSUMÉ: Avis consultatif de la Cour africaine des droits de l’homme et des peuples, (Cour africaine) dans: l’Union panafricaine des avocats sur la convenance des lois sur le vagabondage avec la Charte africaine des droits de l’homme et des peuples et d’autres instruments des droits de l’homme en Afrique (avis de l’UPA) 001/2018, recommande que les États modifient ou abrogent les infractions de vagabondage qui criminalisent les activités de survie de nombreuses personnes ciblées par ces lois. L’opinion principale de l’Cour africaine) constate que les lois et les règlementations nationales des États contenant des infractions de vagabondage sont vagues, trop

larges et ambigus, conférant un large pouvoir discrétionnaire à la police pour leur application et ciblent les personnes en fonction de leur statut, et non de leur conduite. Ces lois violent les articles 2, 3, 5, 6, 7, 12 et 18 de la Charte africaine des droits de l’homme et des peuples; les articles 3 et 17 de la Charte africaine des droits et du bien-être de l’enfant, ainsi que l’article 24 du Protocole à la Charte africaine des droits de l’homme et des peuples relatif aux droits de la femme en Afrique. L’opinion principale aborde insuffisamment le lien colonial du vagabondage. L’opinion concordante séparée donne des précisions sur la marge d’appréciation et le principe de bonne foi qui permettent aux États de formuler des réponses spécifiques au contexte pour modifier et abroger les infractions de vagabondage. Cet article identifie l’impact de la pandémie de COVID-19 sur l’application des infractions mineures et tire des enseignements de litiges tels que Gwanda contre l’état (Gwanda v The State) et des interventions d’institutions nationales des droits de l’homme telles que la Commission sud-africaine des droits de l’homme; ainsi que des orientations de la Commission africaine des principes des droits de l’homme et des peuples sur la dépénalisation des délits mineurs en Afrique (2017). Il est avancé que des interventions spécifiques à impact institutionnel et non matériel sont nécessaires pour apporter le changement systémique requis, en tandem avec l’interdiction des délits de vagabondage, pour les villes qui promeuvent l’inclusion et le développement durable pour tous.

KEY WORDS: petty offences, vagrancy, African Court on Human and Peoples’ Rights, PALU Advisory Opinion, African Commission on Human and Peoples’ Rights Principles on Decriminalisation of Petty Offences in Africa



1 Introduction  

2 Summary of the main findings in the PALU Advisory Opinion  

2.1 The arguments of the requester and intervening parties  

2.2 The African Court’s position  

  3 Discussion of the PALU Advisory Opinion  

4 Conclusion  


Petty offences are found in national penal legislation or local government by-laws and target particular persons on the basis of their status or conduct, with sanctions such as warnings, community service, fines or imprisonment.1 The status of persons as ‘vagrants’ or ‘rogue and vagabond’, ‘idle or disorderly’, ‘a reputed thief’, ‘being without a fixed abode’, beggars, hawkers and vendors, or conduct such as ‘loitering’ and life-sustaining activities such as urinating in public or washing clothes are criminalised through many of these laws in Africa and elsewhere.2 The concept of vagrancy created by colonial masters was not expunged from statutes after the independence of African states. Instead, national and local governments continue to draft and pass criminal laws, including by-laws, with vagrancy related offences that criminalise the life-sustaining activities of poor, homeless and unemployed persons, hawkers and vendors as well as out-groups such as sex workers and LGBTIQ persons and persons with disabilities living within the city limits. Civil society,3 scholars and the international and regional community4 (including the African Commission on Human and Peoples’ Rights (African Commission)), have cited evidence of wide-scale rights violations5 and called for the banning or amendment of such laws to comply with regional and international human rights laws.6 These calls finally found a sounding board.

The African Court on Human and Peoples Rights’ (African Court) seventh advisory opinion was delivered on 4 December 2020 in the Pan African Lawyers Union on the compatibility of vagrancy laws with the African Charter on Human and Peoples’ Rights and other human rights instruments in Africa (PALU opinion).7 The tabling of the issue of the compatibility of vagrancy laws with the African Charter on Human and Peoples’ Rights (Banjul Charter)8 and the African Charter on the Rights and Welfare of the Child (Children’s Charter),9 as well as the Protocol to the African Charter on Human and Peoples’ Rights on the Rights of Women in Africa (Maputo Protocol),10 emanated from a request from the Pan African Lawyers Union (PALU). The PALU opinion focused on a wide array of offences under the rubric of vagrancy.

PALU is a non-governmental umbrella organisation representing Tanzanian national and African regional lawyers’ associations.11 The organisation plays a vital role in extending African human rights

jurisprudence as litigator, consultant and legal aid provider before the African Court.12 PALU has a memorandum of understanding with the African Union (represented by the chairperson of the African Commission).13 Mere observer status with the African Commission does not confer standing to bring a request for advisory opinion as the requirement is that an NGO seeking access must be an ‘African organization’ recognised by the African Union. Six organisations were granted leave to file briefs as amici curiae, and one country, Burkina Faso, made a submission to the African Court. The material jurisdiction of PALU was recognised in that it requested interpretation on particular provisions for the three specified treaties on particular legal aspects related to vagrancy and petty offences.14 Admissibility was granted in light of the African Commission’s response to the African Court’s request about matters pending before it. The African Commission asked the African Court to consider the Principles on the Decriminalisation of Petty Offences in Africa (Decriminalisation Principles) in relation to the legal matters before it. The Decriminalisation Principles set out a normative framework for dealing with outdated, vague and arbitrary criminal laws pertaining to vagrancy.

The African Court considers matters referred to it by the African Commission, legally binding findings are conferred on the subject matter concerned in contentious cases but not if the Commission would request an advisory opinion.15 Advisory opinions are aimed at providing guidance on the interpretation and clarification of treaties and other instruments to domestic courts and tribunals in terms of resolving legal disputes and is not binding.16 The advisory jurisdiction of the African Court is broader than any other court or commission, including in relation to subject matter, as it can advise not only on the interpretation of treaties, but also on human rights instruments.17 Its finesse lies in the soft approach where, due to the lack of dispute of facts and the lack of named perpetrating states or actors, advice and encouragement to change a course of violations and a call for implementing the relevant human rights norms are made.18 The persuasiveness of advisory opinions in the international and regional

arena is promoted as an alternative to coercion.19 The advisory role of the international or regional court (such as the International Court of Justice or the African Court), however, is not limited to the passivity of ‘formal confirmation’, and they can develop the law where there is ambiguity.20 The effect of advisory opinions, however, lies in the adoption of international or regional legal norms in state practice.

The substantive aspects of the PALU opinion are now discussed, including the submissions made by amici and Burkina Faso.


2.1 The arguments of the requester and intervening parties

PALU sought clarity from the African Court on whether vagrancy national laws and by-laws from numerous African countries violate specified articles from the Banjul Charter, the Children’s Charter and the Maputo Protocol based on the assertion that these laws contain offences that;

  1. Criminalise not conduct per se, but rather the ‘status’ of a person as ‘being without a fixed home, employment or means of subsistence; having no fixed abode nor means of subsistence and trade or profession; being a suspected person or reputed thief who has no visible means of subsistence and cannot give a good account of him or herself; and being idle and who does not have visible means of subsistence and cannot give good account of him or herself’.
  2. Mandate ordering a person’s deportation to another area after declaring such a person a ‘vagrant or rogue and vagabond’.
  3. Allow a person’s arrest ‘without a warrant’ on the basis that a person has no ‘means of subsistence and cannot give a satisfactory account’ of him or herself.

In summary, PALU’s submissions to the African Court rested on several arguments. The laws:

  1. Criminalise poverty; punish persons based on an involuntary status, not on the basis of particular conduct; and target or disproportionately impact on poor and vulnerable persons. These outcomes infringe persons’ dignity, equality before the law and the right to non-discrimination under the Banjul Charter.
  2. Allow arrest by the police based on the mere suspicion of having committed an offence or doing so in the future and without evidence or the attempt to obtain evidence. As a result they lead to harassment of particular groups of persons, investigation of ‘unclear offences’ and the removal of ‘undesirable’ populations from the streets. These purposes were argued to be ‘unnecessary for the legitimate purpose of crime prevention’. Furthermore, the mere suspicion of an offence being committed violates the presumption of innocence until proven guilty in terms of the Banjul Charter.
  3. Lead to degrading detention where conditions are appalling, include a lack of food; lead to overcrowding; and such arrests and detentions place a burden on the person’s family to provide food or to pay for bail.
  4. Are imprecise, inaccessible and vague, leaving broad discretion to the police. This results in ‘arbitrary and discriminatory enforcement’ based on their ‘prejudice and social stigma which disproportionately targets poor and marginalised populations’. This therefore violates the rights to equality and equality before the law in terms of the Banjul Charter.
  5. Can result in banishment to a person’s place of origin or deportation where the person is not a citizen, so violating rights to dignity, freedom of movement and protection of the family in terms of the Banjul Charter.
  6. Are used to arrest street children in violation of their rights to dignity and equal protection of the law; children of families where caregivers are detained suffer food insecurity or likely come into conflict with the law; and children are forcibly relocated in violation of their rights to non-discrimination, best interests and a fair trial, which is in conflict with the Children’s Charter.
  7. Disproportionately affect women who often spend more time in pre-trial detention as they are unable to pay fines, bail or obtain legal representation, so conflicting with their entitlement to protection under the Maputo Protocol.

The submissions of Burkina Faso and the amici focused on the effect of the laws. First, the rights violations suffered by the offenders are continuous (during the arrest, before the trial, during and after the trial), and thus flout the prohibition of arbitrary arrest and detention under international human rights law.21 Second, the laws impact the right to a fair trial, so affecting the presumption of innocence.22 Third, unnecessary incarceration exacerbates unsafe prison conditions.23 Fourth, the laws are an ineffective and disproportionate and discriminatory response to social problems such as unemployment, poverty and homelessness and target particular persons, including women, victims of domestic violence and sex workers, and violate the rights to equality before the law and non-discrimination.24 Fifth, this penal response harms the individual and his/her family, and also in socio-economic terms.25 Sixth, the laws are used to arrest and detain persons not for the purpose of prosecution but for intimidation and to remove them from the streets.26 Seventh, the laws criminalise the status of the person and not particular conduct.27 Eighth, the laws ‘are a colonial relic that work to reinforce patterns of discrimination instituted by colonial regimes’.28 Ninth, arrest and detention of street children flouts their best interests and constitutes ‘exploitation, abuse, discrimination and stigmatisation’.29 Tenth, vagrancy laws criminalise women and gender non-conforming persons and prejudice their economic activity, including in the informal sector.30 The broad discretion conferred on law enforcement allows corrupt practices such as exploitation of women’s vulnerability and the extortion of bribes.31 Detained women and their children are harmed by the enforcement of these laws, including where their partners and spouses are detained, as women then carry the household responsibilities. 32

The requester’s second question, whether a positive obligation rests on states to amend or repeal such laws and the nature of such obligations, was supported by the amici. In this regard, the amici and Burkina Faso made submissions relating to mechanisms available to states to implement relevant measures.33 For example, states could fully decriminalise the offences such as with Burkina Faso’s example34 or reclassify them into civil offences, or partially do so where deferred or supervisory sanctions are preferred over detention and incarceration,35 and states could release prisoners convicted of petty offences to ameliorate prison overcrowding as has already been done in Kenya and Egypt.36 Some of the consequences to decriminalisation advocated for by the amici included: decreasing prison overcrowding by releasing inmates, which would also address health concerns such as COVID-19; and respect for the dignity and rights of children and women that would be signalled to law-enforcement agencies.37 PALU highlighted the application of the Kampala Declaration on Prison Conditions in Africa of 1996 and the Ouagadougou Declaration and Plan of Action on Accelerating Prisons and Penal Reforms in Africa to reconsider the use of prisons to prevent crime and to decriminalise petty offences. 38

2.2 The African Court’s position

A unanimous decision was handed down, with a separate opinion filed by Justice Blaise Tchikaya in French.39 The Court’s main opinion was that ‘vagrancy’ as a concept refers to a person’s status and relied on the dictionary meaning and academic writing to articulate the sociological background to this practice of criminalising particular persons.40 Importantly, the African Court describes vagrancy as

a course of conduct or a manner of living, rather than a single act. The term vagrancy is generic. It refers to misconduct brought about by a perceived socially harmful condition or mode of life.41

The colonial root of vagrancy laws and its discriminatory impact are identified as follows: ‘These terms, the Court holds, are a reflection of an outdated largely colonial perception of individuals without any rights and their use dehumanizes and degrades individuals with a perceived lower status.’42 This is the extent of its consideration of the influence of colonialism on these laws - despite the receipt of submissions on this aspect. The colonial origin of vagrancy laws in many provinces, including in KwaZulu-Natal Province, South Africa, where the eThekwini municipality is situated, is well established.43 Perhaps the Court’s omission in further addressing the perpetuation of this colonial legacy is understandable in that the African Commission did not reference this origin in its Decriminalisation Principles. However, it would have been apt for the African Court to expose the colonial premise44 as an added incentive for states to change their stance towards ‘vagrants’ broadly speaking, in order to bring transformative change in pursuit of sustainable, inclusive development.

The main opinion did not offer in-depth examples of vagrancy laws in Africa. Instead, passing reference is made to the existence of such offences in penal codes of 18 states and there are summaries of the definition or conceptions of vagrancy under various codes in one paragraph.45 The African Court identifies that the purpose of its use of examples of some states’ laws or practices during the opinion is to ‘highlight the practical dimensions of the opinion and does not amount to a decision on any factual situation’, which is in line with jurisprudence from the Inter-American Court on Human Rights (Inter American Court) delineating this jurisdictional limitation.46 Its reference to penal codes in particular countries that criminalise vagrancy,47 including a by-law in South Africa, are therefore merely identified examples of potentially impugned vagrancy laws.

The African Court erroneously referred to the example of the by-laws of Ubuhlebezwe local municipality from 2009.48 However, the town of Ixopo and surrounds in this municipality have no beach - as referenced in the by-laws. It is assumed that the African Court meant to refer to the eThekwini Municipality’s by-laws from 2016, which were identified by two of the amici curiae in their submissions: the Centre for Human Rights (CHR) and the Dullah Omar Institute (DOI). The eThekwini Nuisances and Behaviour in Public Spaces By-Law of 2015,49 and its Beaches By-law of 201550 criminalise begging, loitering and life-sustaining activities. Interestingly, the proposed Public Amenities by-laws of Ubuhlebezwe local municipality in 2016 continue to identify loitering as an offence.51

The separate opinion more carefully traversed examples of vagrancy offences and articulates the fluidity of the concept of vagrancy and the ‘African perception’ thereof.52 Justice Tchikaya concludes that the concept of vagrancy under these examples does not ‘specify an act or a commission’ and that ‘being a beggar, poor or wandering cannot in themselves constitute offences’.53

In both advisory opinions, the African Court observed that some state legislatures repealed vagrancy laws.54 Courts have declared these to be unconstitutional in other instances, such as in the Malawian case of Gwanda v The State, where a street vendor was arrested on his way to work under ‘rogue and vagabond’ charges.55

The African Court in its advisory opinion also took judicial notice of the ECOWAS decision in Njemanze and others v Federal Republic of Nigeria56 but curiously not the Kenyan decision in Nyambura & another v Town Clerk, Municipal Council of Mombasa & 2 others.57 The latter decision, however, was not successful, as insufficient evidence was led to convince the court that sex workers’ arrest and detention under a loitering law was unconstitutional on the basis of gender discrimination.58 The ECOWAS decision found multiple violations of the Banjul Charter, the Maputo Protocol and the Convention on the Elimination of all forms of Discrimination against Women (CEDAW), where Nigerian women were arrested on suspicion of being sex workers under loitering laws - simply for being on the street late at night.59 Indeed, Jjuuko and Balya explain that such laws perpetuate the control of sexuality of particular persons such as LGBTIQ and sex workers in particular, including in Uganda.60

On the impugned rights violations, the African Court agreed with all of the submissions of PALU and, unsurprisingly, these findings are also in line with the articulation of such violations in the Decriminalisation Principles. Specifically, the African Court expressed the view that vagrancy laws infringe articles 2 and 3 of the Banjul Charter on the prohibition against unfair discrimination on the basis of a number of statuses and equality before the law. Here the African Court noted that these laws ‘punish the poor and underprivileged, including but not limited to the homeless, the disabled, the gender-nonconforming, sex workers, hawkers, street vendors and individuals who otherwise use public spaces to earn a living’ - primarily based on their ‘economic’ and ‘underprivileged’ status.61 The exacerbation of their socio-economic situation by the enforcement of such laws is also acknowledged.62 The Court agrees that arrest under these laws does not meet the purpose of crime prevention or ‘keeping people off the streets’,63 and, as such, is irrational.64 The Court found that differential treatment on the basis of the status of persons labelled ‘vagrants’ denies them equal protection of the law.65 Furthermore, arresting persons without a warrant and without reasonable suspicion of an offence committed or about to be committed, flouts articles 2, 3 and 5 of the Banjul Charter. This is because such arrests are ‘substantially connected to the status of the individual’ and are both a ‘disproportionate response’ to their poverty and also discriminatory. 66

The African Court identifies that the lack of clarity and precision in the ‘vague, unclear and imprecise language’ and ‘overly broad and ambiguous nature’ of the vagrancy laws do not identify the prohibited conduct to persons subjected to its enforcement by police conferred with broad discretion or the general public.67 As a result, vagrancy laws violate articles 2, 3 and 6 (on the right to liberty) of the Banjul Charter.

The right to dignity and the prohibition of cruel, inhuman and degrading treatment under article 5 of the Banjul Charter are affirmed in the court’s finding that the terminology such as ‘rogue’, ‘vagabond’, ‘idle’ and ‘disorderly’ are ‘outdated’, perpetuate a ‘colonial perception’ of persons as not being rights bearers, and this ‘dehumanizes and degrades individuals with perceived lower status’.68 The court articulated its recognition of a ‘right to enjoy a decent life’ linked to the right to dignity, where it observed that the application of vagrancy laws interferes with peoples’ ‘efforts to maintain or build a decent life or to enjoy a lifestyle they pursue’.69 Furthermore, forceful relocation, sometimes with the use of force, violates articles 5 and 12 (freedom of movement) of the Banjul Charter.

The main opinion finds that the arrest of persons and also requiring them to provide explanations about potential criminal culpability, thus requiring self-incrimination, violates the presumption of innocence in article 7(1) of the Banjul Charter.70 The separate opinion argues that article 7(2) in relation to the principle of legality requiring an offence being an omission or an action, is infringed, as vagrancy laws do not meet this requirement. 71

Limitations to the right to freedom of movement must, according to the court, be proscribed by law and be necessary in order to ‘protect national security, public order, public health or morals or the rights and freedoms of others’, and must be consistent with the other rights in the Charter.72 The first condition is met by vagrancy laws, but the last two are not due to them not meeting their purpose in that ‘there is no correlation between vagrancy and the criminal propensity of an individual’ and because less restrictive measures are available to assist persons affected by these laws.73 Accordingly, the court found that the enforcement of vagrancy laws violates article 12 of the Banjul Charter on the right to freedom of movement.

The right to the protection of the family includes an entitlement to protection from forcible separation.74 The court finds that the arrests and detentions in terms of these laws forcibly removes persons from their families, sometimes relocating them and causing their families ‘deprivation of financial and emotional support’ - particularly for children, the elderly and persons with disabilities - as well as affecting the physical and moral health of the family in contravention of article 18 of the Banjul Charter.75

Children’s right to non-discrimination under article 3 of the Children’s Charter, the court found, is infringed by their arrest, detention and forcible relocation.76 This is primarily based on the discrimination they face based on their status.77 Furthermore, the requirement that they provide an account to avoid arrest affects particular ‘underprivileged and marginalised’ children.78 The court also notes that their removal results in a loss of their ‘community and means of livelihood’ and where their caregivers are relocated or detained, they suffer ‘instability in family relationships and financial problems’.79 As a cross-cutting principle,80 the court holds that the application of the laws infringes on children’s best interests, without explaining the basis on which this finding is made in relation to the specific averments made by PALU or the amici. Children’s fair trial rights under article 17 of the Children’s Charter are also impugned by arrests of children without a warrant, in terms of ambiguous and vague vagrancy laws, and can also be a ‘precursor for further violations of the rights of children’.81 The dignity and ‘worth’ of children in conflict with the law and the recognition of age-appropriate treatment and reintegration is recognised.82

The court explains that the composite obligation of article 24 of the Maputo Protocol protects women who are poor, and who are heads of households and from marginalised populations.83 The court recognises the multiple violations suffered by such women as a result of the application of vagrancy laws, including in relation to their rights to dignity, non-discrimination and equality.84 Furthermore, their economic activities are targeted by these laws.85 The court finds that these laws violate state obligation in terms of the Protocol, to protect them and provide ‘an environment suitable to their condition and their special physical, economic and social needs’.86

As for the requester’s question about the nature of states’ positive obligations to repeal or amend vagrancy laws to conform with the rights under the three treaties, the court disappointingly simply states that the state has a positive obligation to take necessary measures. These include legislative or other measures under articles 1 of the three treaties, and as such they need to ‘amend or repeal all their vagrancy laws, related by-laws and other laws and regulations’ - to bring them in line with the three treaties.87

The separate opinion of Justice Tchikaya illuminates more thoroughly some of the debates that animated the Court in its deliberations and yet which were not traversed in the main opinion. Justice Tchikaya notes that the colonial origin of the laws is indisputable and that they were used for ‘arbitrary arrests and for the excessive and abusive use of colonial power’.88 The vexed issue of succession between the colonial states and current sovereign African states is delineated. Justice Tchikaya advises that while the laws may have had colonial roots, the ‘criminal treatment that States currently administer to the so-called vagrants proceed under their own authority’.89 The Justice explains further that it is up to the states concerned to ‘set the framework and intervene’ - as states have ‘irreducible national jurisdiction in criminal matters’. The inference here is two-fold:

First, states have the discretion on what to make of the obligation to amend and ban vagrancy offences based on the ‘national margin of appreciation’ that exists to ‘temper the obligations of states’.90 Justice Tchikaya refers with approval to the African Commission’s decision in Prince v South Africa, where the doctrine of the margin of appreciation was stated to mean that a state is itself ‘better prepared to adopt [relevant] policies’ as it ‘knows very well its society, its needs, its resources ... and the necessary fair balance between the competing and sometimes conflicting forces that form its society’.91

The second inference comes from the reiteration of the principle of the application of the good faith principle of treaties’ binding nature, which is derived from the Vienna Convention on the Law of Treaties.92 The separate opinion articulates that where a violation is ‘manifestly and objectively’ evident, such as with the ‘sociologically practical domain of vagrancy’, states must act in good faith and ‘respond to a situation of social proximity’.93 In other words, states must act, but how they do so is circumscribed by their own contexts.

Fortunately, Justice Tchikaya identifies that the African Court’s opinion that the states are obliged to take measures ‘as soon as possible’ should have been discussed more in the main opinion - for example in relation to whether this means a ‘reasonable time’ or ‘a short time’.94 However, urgency in taking action is implied by the statement that as ‘a general principle of law, the maintenance of an illegality constitutes a legal uncertainty which must be remedied’.95 The ‘abrupt’ nature of the African Court’s answer to PALU’s request about the nature of the states’ positive obligation to review their laws, is therefore defended in the separate opinion.96


The intervention of PALU to seek advice from the African Court’s comes on the back of a concerted campaign by stakeholders, who have sought social justice for persons affected by petty offences. Civil society has supported the enhancement of the role of national human rights institutions (NHRIs), such as the South African Human Rights Commission (SAHRC),97 in guiding states on how to meet their international obligations in relation to decriminalising petty offences in laws, policies and administrative measures, and exploring different ways of dealing with petty offences that do not violate human rights. Nonetheless, the perspective of NHRIs such as the SAHRC is premised on measures dealing with prison overcrowding,98 and not on the serious rights violations perpetrated by state-sanctioned petty offences that require their unbanning.

That said, the impact of the COVID-19 pandemic on the rights of homeless persons has highlighted some of the acute struggles that they face and the need for social interventions, not criminal interventions, to assist these groups. Promisingly, the SAHRC took a strong stance in its litigation into the Cape Town metropolitan municipality’s egregious approach to shelters for homeless persons during the lockdowns.99 Other states and NHRIs in Africa, however, have started making strides to undo the damage that vagrancy offences cause to affected persons.100 Where law reform is still in process, presidential directives have been used to release persons arrested under petty offences.101 The request for the PALU opinion was lodged in 2018, pre-COVID-19 and it is not clear from public records when submissions closed. This means that it is possible that submissions regarding the impact of COVID-19 on the enforcement of petty offences and its prejudice in respect of particular populations such as the homeless and migrants - constituting violations of a variety of provisions of the three treaties considered - was not properly before the African Court. This is so, except for a limited reference made in the opinion to a submission by an amicus that decriminalisation will address some of the prison overcrowding - considering the COVID-19 pandemic and its impact on prisoner health.102 This advisory opinion, given the timing of the judgment, was an opportune moment for the African Court to reflect on the dire need for decriminalisation of petty offences, particularly as COVID-19 continues to wreak havoc in Africa in relation to the health and movement restrictions of persons. The African Policing Civilian Oversight Forum (APCOF) made a submission to the Special Rapporteur on Prisons, Conditions of Detention and Policing in Africa.103 It identified that the ‘enactment of legislative and regulatory instruments to curb the spread of COVID-19’ disproportionately affects poor and marginalised persons and results in ‘criminal justice sanctions in the context of what are public health and social justice issues’. APCOF argued that the impact of petty offence criminalisation on the poor, including overcrowding, exacerbates pre-existing public health and human rights violations and that states should decriminalise life-sustaining activities in public spaces and rather find permanent alternatives to detention for petty offences during and after the COVID-era.

The findings of rights violations in the opinion are not controversial as they correspond to those in the Decriminalisation Principles. While the African Court was called on to determine the compatibility of petty offences with the three AU treaties, it also referred to its own jurisprudence, United Nations treaties, general comments from treaty monitoring bodies (TMBs), and an ECOWAS decision. The broad mandate of the African Court where it interprets and applies treaties outside of the AU, is not problematic per se, and will not result in

jurisprudential chaos.104 Indeed, the reference to the ICCPR’s conceptualisation of the right to freedom of movement and the African Court’s interpretation that limitations to this right should not ‘nullify its essential content’, are not controversial. Helpfully, in light of a lack of guidance from the African Commission or other TMBs such as the African Committee of Experts on the Rights and Welfare of the Child (African Children’s Committee) on the right to freedom of movement, the African Court’s reliance on the Human Rights Committee’s interpretation of this right as not being subjected to the person’s ‘purpose or reason for staying in or moving out of a specific place’, is crucial. This is to determine the much needed link with the status or reason for a person occupying a public space or using it for life-sustaining activities.

Surprisingly, the African Court did not refer to the report of the Special Rapporteur on Extreme Poverty and Human Rights,105 the guidelines of the Special Rapporteur on the Right to Adequate Housing,106 or the UN Committee on the Rights of the Child’s General Comment on children in street situations.107 These call for the review of laws, policies and measures that discriminate against poor and homeless persons by criminalising vagrancy/loitering, and forcibly evicting homeless persons - including street children. The fact that vagrancy offences, where convicted, left a person with a criminal record, further impacted on their acute marginalisation. This was not traversed in the PALU opinion, perhaps because it was not raised by PALU. However, this argument has been made by other commentators. 108

Of concern is the lack of explicit examples of state vagrancy laws described in the main opinion (though this is partly remedied in the separate opinion), particularly the wording of criminal provisions - except the mention of the South African by-law example. Curiously, the examples provided are summaries of descriptions such as ‘suspected person or reputed thief who has no visible means of subsistence and cannot give a good account’ or being a ‘rogue’ or ‘vagabond’. While one can clearly group many nations’ offences in this way, the distinct impression derived from the main opinion is that the justices shied away from naming and shaming particular states. The South African example is helpful as it is the only by-law that is mentioned in the opinion - as both main and the separate opinion refer to impugned penal codes (national laws). The reference to the Gwanda case by the African Court was brief.109 That case, with three judgments, is helpful to potential litigants, amici, legislatures and other domestic courts for comparative reasons.

Five aspects from these judgments are noteworthy. First, the main judgment in Gwanda by Justice Mtambo lists many cases where rogue and vagabond convictions were overturned but the courts did not invalidate the law.110 Those cases are instructive as the courts identified the poverty targeting aspect of the law, despite making obiter statements about it criminalising poverty. Second, the court only declared one subsection of the law unconstitutional.111 In another obiter statement, the Malawian High Court stated that the ‘reputable thief without visible means of subsistence who cannot give a good account of himself’ offence112 criminalises such suspected persons and is overbroad. However, it did not declare it unconstitutional. Third, the court identified that there are other laws available to law enforcement that can be used to question and arrest persons in a ‘more investigative and/or targeted manner with respect to clear offences’ or on ‘reasonable suspicion’ of committing an arrestable offence.113 Fortunately, this instruction is somewhat tempered by the court indicating that ‘heavy handed’ police conduct will not be countenanced - even under the guise of legitimate offences.114 Fourth, the court accepted evidence from an amicus of a research report which found that the deterrent effect of these laws is not supported and thus the law is not proportional.115 Fifth, the main judgment, concerningly stated, but again obiter, that parliament should draft a ‘new vagrancy law’.116 Justice Kalembera’s concurring judgment elucidated the status targeting aspect of the criminalising law.117 Justice Ntaba’s concurring judgment sought to provide a remedy that is broader than mere declaration of invalidity. It would have: ordered the legislature and executive to fill the gap created by striking down this law, ordered an assessment and review of all vagrancy offences to bring states’ legislation in line with the constitutional prescripts, and directed the police to audit pending cases under this law and to appropriately deal with them.118 All three justices traced the colonial roots of this law.119

Litigators can learn from this case. Relief sought should be couched carefully to ensure that remnants of vagrancy offences do not remain on the statute books. Empirical studies can be tendered as evidence to bolster arguments on discriminatory impacts of these laws and to support a finding of irrationality. Litigators must be mindful that policing attitudes will not change if different laws are used to target persons with the same undesirable status. However, political will is vital to turn the tide against poor persons. Malawi’s penal reform in 2000 did not consider rogue and vagabond offences.120 The Gwanda main judgment still does not provide scope for the broader legislative reform required to address anti-poor sentiments. The law, then, is not the only sword and shield. Policy changes are also needed to address the harm occasioned by criminalising the poor. The Gwanda main judgment and the limited court order keeping strictly to separation of powers bounds, supports an old argument that the judiciary is ‘reluctant to see the courts as an arena for social transformation’.121

While South Africa no longer has vagrancy offences in national legislation, the Criminal Procedure Law renders any ‘offence’ created by municipal by-laws criminal. Killander traces back South Africa’s history of vagrancy laws to 1809 and continuing through successive governments, including colonies, Boer republics, the Union and the apartheid government. This illustrates how both colonial and apartheid sentiments imbued these laws in order to ‘socially [control] the poor’ and as a tool of racism.122 Some by-laws from that era remain, and others have been drafted in recent years - such as the eThekwini municipality example.

The municipality funded a study into homelessness in the city, which was undertaken by the Human Sciences Research Council (HSRC).123 The authors of the study, in a policy brief, argue for a comprehensive policy to be drafted to address the needs of the homeless population in Durban and surrounds, and refer to

unsatisfactory ‘regulatory and programmatic’ responses of the city government.124 The authors do not state explicitly that the rights violations unearthed in the main study (such as experience of violence at the hands of police) are a direct result of the enforcement of the vagrancy by-laws. Holness critically unpacks the eThekwini Municipality’s Nuisances and Behaviour in Public Places By-Laws of 2015 and Beaches By-Laws of 2015,125 and their effect on homeless persons in the city limits. She argues that these laws are contrary to the rule of law and are an irrational extension of local government powers to develop and maintain law and order within municipal boundaries. She posits that these constitute unfair discrimination in respect of a number of protected and unlisted bases that cannot be justified - because they criminalise homelessness and poverty.126 Holness relies on the concept of the ‘right to the city’ as developed by Pieterse127 to explicate the need for a sustainable policy to address the needs of homeless persons, in line with the local government developmental mandate. The eThekwini by-laws criminalise loitering, begging, urinating, washing oneself and one’s clothing, and sleeping in public spaces (including at the beach).128

Evolving jurisprudence in South Africa has successfully attacked some adjunct issues implicated by the enforcement of the similar by-laws, but no successful challenge against the constitutionality of the by-laws has been reported. Some of the issues are: the p olice’s confiscation and destruction of the property of homeless persons in Johannesburg was declared unlawful as it violated their rights to property and dignity;129 and the issuance of fines and destruction of property were interdicted in Cape Town.130

The unhygienic living conditions of a temporary shelter for homeless persons during the hard lockdown in Cape Town, used to mitigate the spread of COVID-19, came before the Western Cape High Court in the Strandfontein case.131 The SAHRC faced off against a stoic city that was apparently unwilling to change its stance on the housing of homeless persons, and blocked human rights monitors from accessing the camp. The Strandfontein shelter was established in March 2020, but after allegations of overcrowding and unsanitary living conditions it was shut down after an interdict was brought by various NGOs, 20 days after it was opened. The High Court issued the interim interdict. It was shut down in May 2021. Unfortunately, alternative accommodation was not offered to homeless persons who occupied the shelter, after the camp was disbanded.132 The City of Cape Town’s conduct in the forced removal of homeless persons to this inadequate shelter and subsequently denial of access to the media and human rights monitors from the SAHRC, was widely criticised.133 The High Court found against the city in the SLAPP suit brought against the SAHRC, finding that the human rights monitors had a lawful right to enter the camp for monitoring purposes.134 Such city responses to the poor during the COVID-19 pandemic evince continuing anti-poor sentiments. Recently, a challenge was filed against the by-laws of the City of Cape Town in the Western Cape High Court and Equality Court.135 That matter is still to be heard.

Learning from the outcome in Gwanda and obtaining a strategic victory from a judgment that declares a by-law unconstitutional, will be unhelpful if instrumental change is not brought and non-material impact is not sought through dedicated training and awareness. Instrumental change occurs where a law is successfully repealed or amended or a policy introduced to change the status quo, and non-material impact is obtained where long-standing prejudices are addressed through, for example, attitudinal changes.136 Court declarations of unconstitutionality are only the starting point for some states in decriminalising petty offences - as the Gwanda decision shows.

There are reports that other offences retained on the statute books such as being ‘idle and disorderly’ are being relied on to arrest and detain people in Malawi since the decision was handed down.137 While some instrumental impact was felt in that instance as the impugned laws were struck down, non-material change has not occurred, as police have not been trained to reverse prejudices against homeless persons, vendors and others targeted by vagrancy laws - and other measures are still being used to repress them. Should a similar challenge against a vagrancy law be successful in a South African court, more will be needed to undo the continuing colonial and racist roots of vagrancy laws in policing practice and local government management of public spaces. The question then is - how do states, including local governments, craft laws that are compliant with their regional and international law obligations as directed in the PALU opinion?

The opinion leaves a wide margin of appreciation to states on how to address this issue of compliance. States’ own contexts will be determinative, and even within states different provinces, cities and towns have differing contexts.138 However, more could have been stated in the opinion in relation to pinpointing which laws are to be repealed and which would still pass muster. This is because although some of the examples listed in the opinion are clear contraventions of the obligations in the Banjul Charter, Children’s Charter and Maputo Protocol, the court left it to states to ‘amend or repeal’ relevant laws and regulations. But where do states obtain guidance on the way forward?

The Decriminalisation Principles provide basic guidelines to states to decriminalise petty offences that are ‘broad, vague and ambiguous’ and that criminalise the status of a person or their appearance and life-sustaining activities in public spaces.139 Any offences referring to the status of a person as ‘criminal’ should be repealed - for example, vagabond, rogue, idle and disorderly. However, what about begging and other life-sustaining activities? To what extent should those be repealed?

The offences that states decide not to decriminalise, the Principles proceed, should be reviewed so that alternatives to arrest and detention can be offered, and such alternatives are to incorporate reasonable accommodation for persons with disabilities and to promote the best interests of children in conflict with the law.140 The question here is - which offences would pass the thresholds created by the Advisory Opinion and the Principles? Furthermore, what alternative can be legitimately pursued? There is very limited literature on best practices as alternatives to arrest and detention in the African context. Meerkotter et al explain that vaunted alternatives such as ‘move on powers’, recording of names, issuing administrative fines and community policing practices, can be discriminatory and violate many other rights.141 As for the latter, the effectiveness of community policing often relies on police using the broader discretion granted to them to uplift and work with communities to find solutions.142 The problem is the legitimacy of such a leadership style in the face of rampant police abuse and corruption in many states such as the sheer brutality experienced at the hands of police enforcing the by-laws in eThekwini.143 The Decriminalisation Principles articulate some alternatives such as diversion, community service, community-based treatment programmes and alternative dispute resolution, as well as declaration of some offences as ‘non-arrestable’.144 A model law should be drafted, after consultation with persons affected by vagrancy laws and related petty offences, to provide guidance to African states on how to craft developmental laws that do not negate the rights of persons criminalised due to their status as homeless, loitering, vending, sex workers, LGBTIQ persons etc.

Measures to address poverty and other marginalisation such as poverty alleviation programmes are mandated by article 22 of the Banjul Charter on the right to development - which was surprisingly not mentioned in either the main or separate opinions. 145


The call for spatial justice recognising the use of free public space for all persons146 will likely continue to fail where attitudes towards undesirable persons are not changed. The change of laws is a start but will only be effective if scaffolded by large-scale and continuous training of law enforcement and relevant civil servants in local and national governments. This training should be on the rights of affected persons in the three treaties and soft law instruments such as the Luanda guidelines, the Decriminalisation Principles, and others, and on positive policing and local government management practices.

The PALU opinion is remiss in failing to provide guidelines to states on how to align statutes and regulations with the three treaties in question. This was a missed opportunity for the African Court to identify what categories of laws should be repealed, such as those criminalising status and vagrancy, and which would comply with states’ obligations in terms of international law. While states are left to amend or repeal their own legislative framework, in accordance with their contexts, minimum standards at best and recommendations at least could have been proposed by this body to eradicate the lingering colonial legacy targeting the poor and homeless and to protect and promote their rights. The Decriminalisation Principles is the vehicle for change at a soft law level. The PALU opinion could have given more credence to these principles to inform states’ best practice.

Civil society stakeholders working with homeless people have generated many practical recommendations on alternative, rights-based approaches to accommodating homelessness in cities that promote the notion of the ‘right to the city’.147 As a starting point, in line with the Decriminalisation Principles, any municipal by-laws or regulations that include petty offences that are ‘broad, vague and ambiguous’ and which criminalise a person’s status or life-sustaining economic activities in public spaces should be struck down.148 Any criminal law response triggered by a person seeking food or shelter will inevitably criminalise poverty. Concepts derived from colonial, racist pasts, such as ‘vagabond’, ‘rogue’ ‘idle and disorderly’, which effectively criminalise a person’s status, have no place in constitutional democracies promoting dignity, equality and freedom.

The following measures could address the systemic rights violations for homeless persons, for example. First, measures should be enacted to guide law enforcement officials on the revised regulatory framework pertaining to homelessness and how to interact with homeless people, and to implement training to transform current policing practice. Such measures include regulations on the confiscation protocol of any goods confiscated, to end the current wanton destruction of homeless people’s property and to ensure the return of their possessions. Regulations prohibiting profiling of homeless people and random searches and arrests without due cause should also be introduced, together with necessary training and complaints mechanisms to ensure their implementation and enforcement. Second, cities should be challenged to adopt a participatory approach to policy formulation and should undertake consultative processes with civil society partners and stakeholders likely to be affected by by-law development and implementation. This would be to yield more responsive public policy that is more likely to be implemented effectively. Third, legal representation to persons affected by vagrancy laws should be prioritised by state legal aid provision and law clinics and private firms acting pro bono in line with the Decriminalisation Principles.149 Similar measures could be adopted depending on the needs and context of other groups affected by these laws, such as sex workers, informal traders, LGBTIQ. In all measures, the rights of persons with disabilities to equality before the law, non-discrimination, dignity and equal participation should be incorporated.

The suggested measures include those of institutional strategic change, in identifying categories of laws and regulations that should be struck down immediately, as well as non-material impact interventions to address stigma and transform discriminatory attitudes, and to promote best practice in managing African cities and their diverse inhabitants. Such responses could be the basis of a transformative approach to foster truly developmental, inclusive, caring cities. The proposed suggestions could be taken up in policy advocacy initiatives with cities and could ultimately be sought by public litigants as court orders following on strategic litigation against recalcitrant cities.

In theory, strategic litigation should not be strictly necessary as states have been issued with the advice of the African Court to ‘as soon as possible’ repeal or review vagrancy offences as described by the court. As the Gwanda case shows, however, litigation is not always the answer. In practice, states are unlikely to prioritise the political will to do so, particularly where alternative approaches to criminalisation and developmental policies are lacking. An African model law drafted with input from affected persons may offer more concrete examples to states on compliant legislative approaches where offences are not repealed outright but are rather amended.

The roles of NHRIs in monitoring and evaluating the systemic change that is needed at state level should be strengthened through research and financial capacitation. The unrelenting nature of the COVID-19 pandemic and its drain on state and humanitarian resources means that now is the time is to decriminalise vagrancy offences and implement developmentally oriented solutions to address poverty and attendant harms. In South Africa, it is hoped that the local governments will heed the call of the African Court and the African Commission’s Decriminalisation Principles and repeal apartheid and colonial era vestiges retained in vagrancy laws.


1. African Commission on Human and Peoples’ Rights (African Commission) Principles on Decriminalisation of Petty Offences in Africa ACHPR/Res. 366 (EXT.OS/XX1) (2017) para 1 (Decriminalisation Principles).

2. Southern African Litigation Center Vagrancy-related provisions in various criminal laws and criminal procedure laws in Africa (2018) https://icj-kenya.org/wp-content/uploads/2018/01/Vagrancy-laws-in-Africa-by-Southern-Africa-Litigation-Center-SALC.pdf (This summary identifies South Africa as not having vagrancy offences in criminal law. However, local government by-laws do contain vagrancy offences).

3. Open Society Foundations (Human Rights Initiative) Campaign on the decriminalisation and declassification of petty offences in Africa www.pettyoffences.org ; Network of African National Human Rights Institutions (NANHRI) Enhancing The role of African national human rights institutions in decriminalisation of petty offences project phase II https://www.nanhri.org/decriminalization-of-petty-offences/ (accessed 7 February 2022).

4. The Kampala Declaration on Prison Conditions in Africa (1996); ACHPR The Ouagadougou Declaration and Plan of Action on Accelerating Prisons and Penal Reforms in Africa (2002).

5. L Muntingh & K Peterson Punished for being poor: evidence and arguments for the decriminalisation and declassification of petty offences (2015).

6. Decriminalisation Principles (n 1) para 14.

7. https://www.african-court.org/cpmt/details-advisory/0012020

8. Organization of African Unity (OAU) African Charter on Human and Peoples’ Rights 27 June 1981, CAB/LEG/67/3 rev. 5, 21 ILM 58 (1982).

9. OAU African Charter on the Rights and Welfare of the Child 11 July 1990, CAB/LEG/24.9/49 (1990).

10. African Union Protocol to the African Charter on Human and People’s Rights on the Rights of Women in Africa 11 July (2003).

11. The registration of an organisation in an African country with sub-regional, regional or continental branches and activities that extend beyond its country of registration, confers jurisdiction. See Request for Advisory Opinion by L’Association Africaine de Defense des Droits de l’Homme 28 September 2017 2 AfCLR 637 para 27.

13. The existence of such a memorandum of understanding confers jurisdiction. See Request for Advisory Opinion by the Centre for Human Rights, University of Pretoria and others 28 September 2017 3 AfCLR 622 para 49.

14. Rule 82(2) of the Rules.

15. AP van der Mei ‘The advisory jurisdiction of the African Court on Human and Peoples’ Rights’ (2005) 5 African Human Rights Law Journal at 27.

16. Art 3 of the Protocol to the African Charter on Human and Peoples’ Rights on the Establishment of an African Court on Human and Peoples’ Rights OAU Doc OAU/LEG/EXP/AFCHPR/PROT (III) Court Protocol).

17. Art 4 of the Establishment Protocol. van der Mei (n 15) 38; R Murray ‘The human rights jurisdiction of the African Court of Justice and Human and Peoples’ Rights’ in CC Jalloh and others (eds) The African court of justice and human and peoples’ rights in context: development and challenges (2019) 967.

18. Van der Mei (n 15) 30.

19. K Oellers-Frahm ‘Lawmaking through advisory opinions?’ (2011) 12 German Law Journal at 1050.

20. F Mayr & J Mayr-Singer ‘Keep the wheels spinning: The contributions of advisory opinions of the International Court of Justice to the development of international law’ (2016) 76 ZaöRV at 447.

21. ICJ-Kenya (n 2) para 52.

22. Burkina Faso (n 7) para 50.

23. NAHRI para 51.

24. Burkina Faso and NAHRI paras 50 & 52.

25. ICJ-Kenya and NAHRI paras 51-52.

26. Centre for Human Rights (CHR) and Dullah Omar Institute (DOI) para 54.

27. HRC-Miami and Lawyers Alert para 55.

28. Open Society Justice Initiative para 56.

29. NAHRI para 111.

30. HRC-Miami, Lawyers Alert, CHR & DOI paras 132 to 133.

31. CHR & DOI para 133.

32. CHR & DOI.

33. ICJ-Kenya flagged the Ouagadougou Declaration’s call for decriminalisation of these offences, including sex work, failure to pay debts and disobedience to parents (these particular aspects were not discussed in the opinion) para 146.

34. Burkina Faso decriminalised the offence of ‘wandering’ para 145.

35. ICJ-Kenya para 146.

36. NAHRI para 147.

37. NAHRI para 147 and OSJI para 148.

38. Main opinion paras 143-144.

39. An official English translation is not available. This separate opinion was translated for the purpose of this case note and is not authoritative.

40. para 59, 72.

41. para 57.

42. para 79.

43. J Martens ‘Polygamy, sexual danger, and the creation of vagrancy legislation in colonial Natal’ (2003) 31 Journal of Imperial and Commonwealth History 24; M Killander ‘Criminalising homelessness and survival strategies through municipal by-laws: Colonial legacy and constitutionality’ (2019) 35 South African Journal on Human Rights 70.

44. For a historical perspective, see AL Beier & P OcoBock Cast out: vagrancy and homelessness in global and historical perspectives (2008).

45. Main opinion paras 60 & 135.

46. Main opinion para 36 and fn 17 referring to IACHR Advisory Opinion OC-18/03 of September 2003 Requested by the United Mexican States, Juridical Condition and rights of undocumented migrants paras 63-65.

47. Main opinion (n 7) para 60.

48. Main opinion (n 7) 25.

49. Clause 5(2).

50. Clause 10(1).

51. Clause 10 of the draft Public Amenities by-laws of Ubuhlebezwe local municipality (2016): ‘No person leading the life of a vagrant or who lacks any determinable and legal refuge or who leads a lazy, debauched or disorderly existence or who habitually sleeps in a public street, public place or other non-private place or who habitually begs for money or goods or persuades others to beg for money or goods on his behalf, may loiter or linger about or sleep on, in or at a public amenity.’

52. Separate opinion paras 17 to 22, referring to penal codes of Senegal, Algeria, Mali, and Ivory Coast.

53. para 24.

54. For example, repeal in Angola, Cape Verde, Kenya, Lesotho, Mozambique, Rwanda and Zimbabwe; and modifications or decriminalisation in Tunisia, Burkina Fasa and Kenya.

55. Main opinion para 61; separate opinion para 39. Gwanda v The State [2017] MWHC 23. Sec of 184(1)(c) of the Penal Code.

56. ECW/CCJ/JUD/08/17 (ECOWAS).

57. [2011] eKLR available at http://kenyaliisync.africanlii.org/node/117667 (no page numbers).

58. W Holness ‘eThekwini’s discriminatory by-laws: Criminalising homelessness’ (2020) 24   Law, Democracy and Development at 484-491, 492.

59. ECOWAS found violations of arts 1, 2, 3 & 18(3) of the ACHPR, arts 2, 3, 4, 8 & 25 of the Maputo Protocol, and arts 2, 3, 5(a) & 15(1) of CEDAW.

60. A Jjuuko & J Balya ‘Taking advantage of political processes to challenge the use of “idle and disorderly” offenses to police sexuality in Uganda’ (2020) 75 University of Miami Law Review Caveat at 44.

61. Main opinion (n 7) 70, 72 & 74.

62. Main opinion (n 7) para 70.

63. Main opinion (n 7) para 72.

64. Main opinion (n 7) para 82.

65. Main opinion (n 7) para 73.

66. Main opinion (n 7) paras 74, 75 & 82.

67. Main opinion paras 71 & 86.

68. Main opinion para 79. This opinion, the court reached, after recalling Purohit and Moore v The Gambia [2003] ACHPR 49 para 59 (dehumanizing terminology meted out to persons with psycho-social disabilities).

69. Main opinion para 80, in line with the decision in Purohit and Moore v The Gambia (n 68) para 61.

70. Main opinion paras 89 and 94. The Court relies also on the interpretation of art 14 of the ICCPR and the African Charter’s Principles and Guidelines on the Right to a Fair Trial and Legal Assistance in Africa, 2003, regarding the principle against self-incrimination.

71. Separate opinion para 46.

72. Main opinion paras 97 to 99 relying on art 7 of the African Charter and art 12(3) of the ICCPR.

73. para 101.

74. para 104 citing the protection of the family unit encapsulated in various international human rights instruments, fn 50.

75. Main opinion paras 105 and 107.

76. para 120.

77. para 119.

78. para 118.

79. para 119.

80. The Court relies on the African Children’s Committee General Comment 5 ‘State obligations under the African Charter on the Rights and Welfare of the Child (article 1) and System Strengthening for Child Protection’ (2018).

81. Main opinion para 126.

82. para 127.

83. para 137.

84. para 138.

85. para 139.

86. para 140.

87. para 154.

88. Separate opinion para 48.

89. para 50 referring to the principle that transfer of laws and regulations is a result of territorial sovereignty.

90. Separate opinion para 53.

91. Prince v South Africa (2004) AHLHR 105; para 51 referred to in the Separate opinion para 55.

92. Arts 26 and 46(2) of the Vienna Convention on the Law of Treaties.

93. Separate opinion para 37.

94. para 33.

95. Van der Mei (n 15) 30.

96. Main opinion para 34.

97. SAHRC National consultation on the African Commission Principles on Decriminalisation of Petty Offences in Africa (21 November 2018) https://www. sahrc.org.za/index.php/sahrc-media/news-2/item/1664-media-alert-national-co nsultation-on-the-african-commission-principles-on-decriminalisation-of-petty-offences-in-africa (no report on this consultation is available publicly); African Policing Civilian Oversight Forum & SAHRC Report on Dialogue on Human Rights and Policing (2 May 2018) 4-6 http://apcof.org/wp-content/uploads/apcofhumanrightsdialoguereport-2-3may2018johannesburg.pdf (this report does not mention the role of banning or amending laws that create petty offences, but engages with the role of the police and policy development).

98. SAHRC Report of the South African Human Rights Commission: The implementation of the OPCAT in South Africa 2019/2020 (2020) 29 https://www.justice.gov.za/ilr/docs/2019-20-NPM-AnnualReport-SAHRC.pdf (accessed 16 February 2022).

99. Discussed below.

100 Uganda’s NHRI calls for decriminalisation of petty offences and amendment of their Penal Code Act; Human Rights Awareness and Promotion Forum ‘When being poor and ‘undesirable’ is a crime: reflections on the impact of the ‘Idle and disorderly’ laws on marginalised groups in Uganda’ (2018) 5 The Human Rights

100. Advocate Magazine https://hrapf.org/index.php/resources/human-rights-ad vocate-magazine/102-fifth-issue-of-the-human-rights-advocate/file ; Jjuuko & Balya (n 60) 43; Sierra Leone’s New Bail and Sentencing guidelines were developed to provide non-custodial sentences for petty offences.

101. IGP orders release of ‘Idle and disorderly’ offenders The Independent 2 October 2019 https://www.independent.co.ug/igp-orders-release-of-idle-and-disorderly-offenders/ (accessed 16 February 2022).

102. OSJI submission para 148.

103. APCOF Submission to the Special Rapporteur on Prisons, Conditions of Detention and Policing in Africa at the 66th Ordinary Session of the African Commission on Human and Peoples’ Rights July 2020 available at https://osisa.org/66th-ordinary-session-of-the-african-commission-on-human-and-peoples-rights/ (accessed 16 February 2022).

104. A Rachovista ‘On new ‘judicial animals’: the curious case of an African Court with material jurisdiction of a global scope’ (2019) 19 Human Rights Law Review at 255.

105. United Nations Office of the Human Rights Commissioner (UNOHRC) Report of the Special Rapporteur on extreme poverty and human rights, Ms Maria Magdalena Sepúlveda Carmona on the penalization of people living in poverty 4 August 2011 A/66/265 para 21.

106. United Nations General Assembly Guidelines for the implementation of the right to adequate housing Report of the Special Rapporteur on adequate housing as a component of the right to an adequate standard of living, and on the right to non-discrimination in this context 29 December 2019 A/HRC/43/43 para 33.

107. Committee on the Rights of the Child General Comment No 21 on Children in Street Situations (2017) CRC/GC21/2017 para 14.

108. K Peterson ‘Law and policy: Barriers to accessing justice for sustainable development’ (2020) 21 ESR Review at 25.

109. The judgments can be found at https://malawilii.org/mw/judgment/high-court-general-division/2017/23.

110. For example Republic v Balala [1997] 2 MLR 67; Chidziwe v Republic, Criminal Appeal 14 of 2013 (unreported).

111. Gwanda (n 55) 26.

112. Sec 184(1)(b) of the Malawian Penal Code.

113. The offence of criminal trespass (sec 319 of the Penal Code); and arrest of persons about to commit an arrestable offence or on the basis of reasonable grounds of suspecting to be about to commit an arrestable offence (sec 28 of the Criminal Procedure and Evidence Code). Gwanda, Mtambo judgment, 26. Justice Ntaba’ judgment para 4.64, however disagrees and opines that sec 28 is procedural and does not create an ‘offence’.

114. Gwanda (n 55) Mtambo judgment, 27.

115. Referring to the study that the amicus Centre for Human Rights Education, Advice and Assistance (CHREAA) cited, namely A Meerkotter and others No justice for the poor: a preliminary study of the law and practice relating to arrests for nuisance-related offences in Blantyre, Malawi (2013) 66.

116. Gwanda (n 55) Mtambo judgment, 26.

117. Kalembera judgment, 13.

118. Ntaba judgment, 31, paras 5.7.1-5.7.3.

119. Mtambo judgment, 4; Ntaba judgment, paras 4.2 to 4.4; Kalembera judgment, 5.

120. C Banda & A Meerkotter ‘Examining the constitutionality of rogue and vagabond offences in Malawi’ in Southern Africa Litigation Centre, judiciary of Malawi, National Association of Women Judges and Magistrates of Botswana (NAWABO) Using the courts to protect vulnerable people: Perspectives from the judiciary and legal profession in Botswana, Malawi, and Zambia (2014) at 71.

121. S Gloppen & FE Kanyongolo ‘Courts and the poor in Malawi: economic marginalization, vulnerability, and the law’ (2007) 5(2) International Journal of Constitutional Law at 290.

122. Killander (n 43) 73-78.

123. HSRC Ikhaya lami: understanding homelessness in Durban: final report (2016) http://www.hsrc.ac.za/en/research-outputs/view/8181 (accessed 7 February 2022).

124. C Desmond et al ‘Towards the development of a contextualised homelessness policy: A Durban case study’ (2017) HSRC Policy Brief 2 (accessed 7 February 2022).

125. eThekwini Municipality Bylaws http://www.durban.gov.za/Resource_Centre/Pages/By-Laws.aspx (accessed 7 February 2022).

126. Holness (n 58) 484-491.

127. M Pieterse ‘Development, the right to the city and the legal and constitutional responsibilities of local government in South Africa’ (2014) 131 South African Law Journal at 149. See also Cf Killander (n 43) 72 (referring to Lefebvre’s coining of the concept).

128. Clauses 5(2)(c), (d), (e), (k), (r) and (u), as well as clauses 12(1)(b) and 12(2) of the Nuisances and Behaviour By-laws. Clause 10(1)(5) of the Beaches By-laws.

129. Ngomane & others v City of Johannesburg Metropolitan Municipality & another 2020 (1) SA 52 (SCA).

130. Gelderbloem and six others v the City of Cape Town case 14669/2019 (not yet finalised).

131. Withdrawn.

132. V Cruywagen ‘City of Cape Town defends dumping of homeless under bridge’ 25 May 2020 The Daily Maverick https://www.dailymaverick.co.za/article/2020-05-25-city-of-cape-town-defends-dumping-of-homeless-under-bridge/#gsc.tab=0 (accessed 7 February 2022).

133. J Cogger ‘The Strandfontein shelter touches a societal and political nerve’ 27 May 2020 The Mail & Guardian https://mg.co.za/opinion/2020-05-27-the-strand fontein-shelter-touches-a-societal-and-political-nerve/ (accessed 7 February 2022).

134. City of Cape Town v South African Human Rights Commission and others Case 5633/2020 WCHC (unreported). See J Stent ‘Court slams City of Cape Town for barring access to homeless camp’ Groundup 17 March 2021 https://www.dailymaverick.co.za/article/2021-03-17-court-slams-city-of-cape-town-for-barring-access-to-homeless-camp/ (accessed 7 February 2022).

135. Gelderbloem & others v City of Cape Town Case no 5708/21 (WCHC); Gelderbloem & others v City of Cape Town Case no 06/21 (WCEC) challenging the Streets, Public Places and the Prevention of Noise Nuisances   (2007) and   Integrated Waste Management   (2009) By-Laws.

136. Open Society Justice Initiative (2018) Strategic litigation impacts: Insights from global experience https://www.justiceinitiative.org/uploads/fd7809e2-bd2b-4f5b-964f-522c7c70e747/strategic-litigation-impacts-insights-20181023.pdf (accessed 7 February 2022).

137. V Mhango & A Meerkotter ‘Policing of petty offences’ The Nation 19 October 2017 https://www.southernafricalitigationcentre.org/2017/10/18/policing-of-petty-offences/(accessed 7 February 2022).

138. L Edwards ‘Africa: A regional campaign to decriminalise petty offences’ World Prison Brief 15 June 2021 https://www.prisonstudies.org/news/africa-regional-campaign-decriminalise-petty-offences (accessed 7 February 2022).

139. Principle 14.1.

140. Principle 14.2.3.

141. Meerkotter et al (n 115) 122.

142. WF Walsh ‘Compstat: An analysis of an emerging police managerial paradigm’ (2001) 24 Policing: an International Journal of Police Strategies & Management at 351, cited in Meerkotter et al (n 141) 119.

143. HSRC (n 123) 26.

144. Principle 14.2.2.

145. But see principle 14.3.1 of the Decriminalisation Principles.

146. I Kriel ‘Engaging with homelessness in the City of Tshwane: Ethical and practical considerations’ (2017) 34(4) Development Southern Africa at 468; V Naidoo (2010) ‘Government responses to street homelessness in South Africa’ (2010) 27 Development Southern Africa at 129.

147. T Görgens & M van Donk 2012 ‘Exploring the potential of the “Right to the City” to integrate the vision and practice of civil society in the struggle for the socio-spatial transformation of South African Cities’, Isandla Institute, Cape Town.

148. Principle 14.1.

149. Principle 14.4.1(b).