City of Cape Town v Various Occupiers and Another (21101/2022) [2024] ZAWCHC 173 (18 June 2024)

1


IN THE HIGH COURT OF SOUTH AFRICA
(WESTERN CAPE DIVISION, CAPE TOWN)

CASE NO: 21101/2022
In the matter between:

CITY OF CAPE TOWN


and


THOSE PERSONS IDENTIFIED IN ANNEXURE “A
TO THE NOTICE OF MOTION WHO ARE
UNLAWFULLY OCCUPYING THE ERVEN WITHIN
THE CITY OF CAPE TOWN CENTRAL BUSINESS
DISTRICT AND SURROUND AS MORE FULLY
DESCRIBED IN PARAGRAPH 2 OF THE NOTICE OF
MOTION


THOSE PERSONS WHOSE FULL AND FURTHER
PARTICULARS ARE UNKNOWN TO THE APPLICANT
WHO ARE UNLAWFULLY OCCUPYING THE ERVEN
WITHIN THE CITY OF CAPE TOWN CENTRAL
BUSINESS DISTRICT AND SURROUND AS MORE
FULLY DESCRIBED IN PARAGRAPH 2 OF THE
NOTICE OF MOTION

Applicant





First Respondent








Second Respondent

Coram: Bishop, AJ

Date of Hearing: 9 and 10 October 2023

Date of Further Submissions: 10 and 28 November and 8 December 2023

Date of Judgment: 18 June 2024
2

JUDGMENT


BISHOP, AJ

[1] In 1997, Chaskalson P observed that, “[w]e live in a society in which there are
great disparities in wealth”, and in which “[m]illions of people are living in deplorable
conditions and in great poverty. ”1 A commitment to address those conditions, and
ensure that all South Africans live lives of dignity, equality and freedom “lies at the
heart of our new constitutional order. For as long as these conditions continue to exist
that aspiration will have a hollow ring.”
2
[2] What was true in 1997 remains true today. While conditions for millions have
improved, still m illions of people in South Africa live in poverty with inadequate
housing, water, healthcare and food. The Constitution’s call to remedy those
conditions remains no less urgent.
[3] This case concerns the rights of some of the most vulnerable people in our
society – people living on the pavements of downtown Cape Town. The conditions in
which they live are deplorable. They live next t o busy roads in tents or structures
constructed of plastic sheets and cardboard. They are compelled to live their lives in
public, with little or no privacy. They struggle for food, for shelter, and for warmth.

1 Soobramoney v Minister of Health (Kwazulu-Natal) [1997] ZACC 17; 1998 (1) SA 765 (CC); 1997 (12)
BCLR 1696 (CC) at para 8.
2 Ibid.
3

[4] The Applicant – the City of Cape Town – has a duty to these people. “It is
irrefutable that the State is obliged to take positive action to meet the needs of those
living in extreme conditions of poverty, homelessness or intolerably inadequate
housing.”
3 It has a duty to remedy their conditions of living , to take reasonable steps
to realise their right to housing, and to ensure they can live lives with dignity and
privacy. Grootboom reminds us that “the Constitution requires that everyone must be
treated with care and concern”, and that those “whose needs are the most urgent and
whose ability to enjoy all rights therefore is most in peril, must not be ignored”.4
[5] The homeless people of the City do not exist separately from the rest of the
residents. We are all part of the same whole. The City consists of and belongs to those
who work in its corner offices and live in the mansions of Higgovale and Clifton, just
as much as it does to those who eke out a living on the City’s streets and sleep on its
pavements and in its parks. We are all entitled to the same level of respect and
concern from our City and from each other.
[6] This case is a reminder that the Constitution dares us all “to care for the people
on the edge of the night ”. It also dares us “ to change our way of caring about
ourselves.”5 We can only care about the homeless when we see ourselves in them ;
we can only realise our own humanity if we commit to realizing theirs ; when we see

3 City of Cape Town v Commando and Others [2023] ZASCA 7; [2023] 2 All SA 23 (SCA); 2023 (4) SA
465 (SCA) at para 5.
4 Government of the Republic of South Africa and Others v Grootboom and Others [2000] ZACC 19;
2001 (1) SA 46; 2000 (11) BCLR 1169 (CC) at para 44.
5 Queen Under Pressure (1981).
4

that we are all a few bad decisions and some bad luck from life on the pavements.
Umntu ngumntu ngabantu.6
[7] To its great credit, this is largely the attitude the City has adopted in this
litigation. It asks this Court to evict approximately 200 people (the Occupiers) that live
on seven pavements or road reserves that it owns around the City’s central business
district (the Properties). It does not, as other municipalities have, seek to remove
them without a court order and without alternative accommodation. It applies under
the Prevention of Illegal Eviction and Unlawful Occupation of Land Act 19 of 1998
(PIE). Nor has it sought to banish them to the City ’s periphery; it offers all the people
it seeks to evict alternative accommodation in “safe spaces” that it has developed in
the City centre.
[8] The accommodation in the safe spaces is rudimentary. But it is undoubtedly
better than the Occupiers’ current accommodation. It includes toilets and showers, two
meals a day, blankets, and access to clothes. And it comes with a range of services
designed not only to give them somewhere to sleep, but to help them to get off the
streets and into permanent homes. The City commits to helping those who use its safe
spaces to overcome addiction, to find jobs, and to reconnect with their families.
[9] Some of the occupiers are willing to take up the City’s offer. But others are not.
They insist that the City has not adequately engaged with them – it presented them
with a binary choice: safe spaces or nothing. They want the City to explore alternatives

6 See New Nation Movement NPC and Others v President of the Republic of South Africa and Others
[2020] ZACC 11; 2020 (8) BCLR 950 (CC); 2020 (6) SA 257 (CC) at para 25, fn 31: “IsiXhosa for: ‘One
becomes a fulfilled human being because of others. ’ Literally, ‘A person is a person because of other
people.’”
5

with them, and to find joint solutions. They also argue that the safe spaces are not
suitable alternative accommodation. They do not meet the ordinary requirements for
temporary accommodation following eviction, and they impose restrictive rules that
separate families and restrict freedom.
[10] Against this background, this case raises the following primary questions for
determination:
[10.1] Has the City meaningfully engaged with the Occupiers?
[10.2] Are the safe spaces suitable alternative accommodation?
[10.3] Is eviction just and equitable and, if so, on what terms?
[11] I conclude that the City has meaningfully engaged, that the safe spaces are
suitable alternative accommodation, and that eviction is just and equitable, but subject
to a detailed order to ensure the Occupiers’ rights are fully respected.
[12] Granting the eviction raises another question: the City seeks an interdict
preventing a list of named Occupiers from re-occupying the Properties or any other
City-owned property. It argues that, without this relief, the eviction will not achieve its
purpose. The Occupiers who do not already wish to move to the safe spaces will have
little incentive to do so; they can simply occupy a different pavement that was not
subject to the eviction order. The City cannot afford, it argues, to keep evicting people
from the pavements.
[13] The Occupiers contend that an interdict would authorize further evictions
without the protection of a court order as required by s 26(3) of the Constitution. It will
6

place them outside the ordinary protections of PIE and subject to removal at the whim
of the City. They argue such an order would be unlawful and unconstitutional.
[14] I agree with the Occupiers. While I understand the City’s difficulty, and believe
that it is motivated by a desire to assist the homeless amongst us, not to marginalize
them, an interdict is not the appropriate mechanism. If the City believes it needs the
power to evict people outside of PIE, it must defend the existing mechanisms it has
created to do so, or create new ones that are constitutionally consistent and apply
equally to everyone.

THE FACTS
[15] It is always useful to start with the facts. To understand the outcomes this Court
reaches, I need to explain where the properties are and why the City is particularly
concerned about people living on them. Next, I look at the City’s approach to
homelessness generally, and particularly its use of safe spaces; what do they provide,
and what restrictions do they impose. I then describe the Occupiers – where they come
from, how they live, and what they want. That leads to how this dispute came to court
and the shape of the litigation.

THE PROPERTIES
[16] The City seeks the eviction of people living on seven defined sites in the inner
city. These are defined by clear descriptions, and by maps attached to the notice of
7

motion. They are all pavements or road reserves along major roads leading into and
out of the City’s centre. They are:
[16.1] The pavement to Buitengracht Street (inclusive of the corn er of Rose
Lane and Buitengracht Street behind the wooden bollards along the
edge of the pavement);
[16.2] Either side of the road on FW De Klerk Boulevard (inclusive of the
pavements, centre island and road reserve);
[16.3] The corner of FW De Klerk Boulevard and Heerengracht leading into the
harbour area (inclusive of Foregate Square, Taxi Rank and Foreshore
as well as outside and opposite Customs House and along Heerengracht
and the pavements and road reserves in front of, opposite and along
Heerengracht);
[16.4] Helen Suzman Boulevard (inclusive of the pavements and road reserves
on either side of the road and the centre island);
[16.5] Strand Street (inclusive of the pavements and road reserve on both sides
of the road after the station outbound and over the entire width of the
pavement on both sides of the road) and the Strand Street side of the
Castle on the pavement and road reserve and grass area outside the
Castle;
[16.6] Foreshore N 1 (near Turbines) inclusive of the pavement and road
reserve and area surrounding the Roggebaai Gas Turbines; and
8

[16.7] Virginia Avenue and Mill Street Bridge (inclusive of the pavement and
road reserve).
[17] The City owns all the Properties. The City does not seek the eviction of persons
living in similar conditions on land it does not own. For example, there are other
homeless people living next to the Castle on land owned by the national government;
the City does not seek their eviction.
[18] The City emphasizes that the Properties are not fit for human habitation. They
are adjacent to busy roads and therefore unsafe. There is no access to water,
sanitation or electricity. The structures – which are either tents , or are made out of
plastic and cardboard – are unfit for long- term habitation. They provide limited
protection from the elements, and little or no safety or security.
[19] The City describes the problems at each site specifically. But several problems
are common across most or all of them.
[20] First, there are risks to state infrastructure. The Occupiers – understandably –
make fires to cook and to keep themselves warm. But this can damage pavements,
bridges and other infrastructure. A particularly severe example is the risk posed by
people living and making fires near the Roggebaai Gas Turbines . Signs and fences
have been vandalized to use for shelter or gain access to better areas. Waterpipes
and drains have been obstructed or damaged. Simply put, the road reserves contain
a wide range of infrastructure, none of which is designed to operate with people living
there. The Occupiers’ presence there threatens this infrastructure creating a risk for
them, and for all who rely on it.
9

[21] Second, the Occupiers obstruct pedestrian and vehicular traffic. Because they
live on the pavement, people using the pavement (including the occupiers) are forced
onto the road. This creates a risk both for them and for motorists. It also denies other
residents of the City the ability to freely use the pavements that were designed for that
purpose.
[22] T
hird , the conditions are unhealthy and hazardous for the Occupiers
themselves. They suffer from malnutrition , physical and psychological health r isks
from li ving such unsheltered lives, and diseases cause d by food wast e, lack of
sanitation, exposure to fires and vermin. People living on the streets are also generally
less likely to seek he althcare when they become sick , or to take their medication as
directed.
[23] F ourth, the conduct of the Occupiers a ffects people living and working in the
City. The City argues that homeless people are forced to conduct normal human
behaviour in public, including urinating, defecating, bathing and having sex. The City
alleges that this undermines the Occupiers’ privacy and dignity, as well as i mpacting
on others in the area.
[24] To a degree, I accept that these concerns justify the City’s decision to seek an
eviction. It is no criticism of the Occupiers to say that they are compelled to live in
public – they have no choice. The limitation on their rights by being compelled to do
so is far greater than any impact on those who must observe them. But there are
reasons why the law generally prohibits urinating, defecating and washing in public. It
is unsanitary and unpleasant for those who must observe others doing in public what
should be done in private. It creates health risks for the Occupiers and for others.
10

[25] T o make its point, the City relies on complaints that it has received from
members of the public about homeless people in the CBD. These complaints range
from an inability to use bus stops , litter and the vandalism of public infrastructure, to
the depreciation of property values, and the flight of tourists and businesses from the
CBD.
[26] I accept that the Occupiers’ occupation of the properties make the use of the
City more difficult for others who use the City , and may make the downtown less
desirable location. I think this is already established by the City’s other evidence.
[27] But I do not think much weight should be placed on these complaints. They are,
largely, complaints about the existence of homeless people in the CBD. But
homelessness is a reality without any simple solution. Ms Pillay SC, who appeared for
the City, repeatedly reminded me that this case was not about “solving” homelessness.
I agree. But that means that the complaints will persist for as long as people have
nowhere else to live and gravitate to the CBD to make a living. That will happen with
or without this eviction order.
[28] There is also a degree of nimbyism about the complaints – “Please move these
homeless people somewhere else and make them somebody else’s problem.” But
homelessness is our problem. Homeless people are part of the City as much as all its
other residents. Homelessness is a result of some factors beyond the City’s control,
and some choices we have made as a society about the distribution of resources.
None of the causes or symptoms of homelessness will be solved by moving homeless
people from one place to another, even to a safe space.
11

[29] F ifth, there is a complaint that homeless people are responsible for various
forms of crime in the City, including selling drugs, petty theft, mugging and sex work. I
do not place any weight on this complaint for two reasons:
[29.1] The City did not explain why, if the Occupiers are responsible for these
crimes, they will stop committing them if they are evicted from the
Properties and move to safe spaces.
[29.2] I accept that some homeless people may commit crimes. But to use that
as a justification for eviction without specific proof linking any Occupier
to crime seems unwarranted. The Indian Supreme Court described a
similar charge against homeless people as “born of prejudice against the
poor and the destitute. Affluent people living in skyscrapers also commit
crimes varying from living on the gains of prostitution and defrauding the
public treasury to smuggling. But, they get away. ”
7 If belonging to a
category of people who commits crimes was relevant to eviction,
homeless people are not the only ones who would face an insecure
future.
[30] The Occupiers accept that living on the street creates risk for their physical and
mental well-being. They do not contend that they have a right to occupy the properties
indefinitely. They accept that, ultimately, their eviction may be warranted – but only
once it has engaged meaningfully and offered suitable alternative accommodation.


7 Olga Tellis v. Bombay Municipal Corporation 1986 AIR 180, 1985 SCR Supl. (2) 51.
12

HOMELESSNESS IN THE CITY AND THE SAFE SPACES
[31] There are no clear figures of how many homeless people live in Cape Town, or
in the CBD. The estimates range from about 6 175 in 2015 (including those living in
shelters) to 14 357 in 2020. The parties agree that the economic impact of the
Covid-19 pandemic significantly exacerbated the problem. But there were no reports
to estimate just how much the population of homeless people has increased.
[32] Whatever the precise number, there are thousands of people living on the
streets in Cape Town. Most of them live in the CBD, Belville and Mitchell’s Plain
because that is where the opportunities exist for them to make a living. Homeless
people perform a number of valuable services in these areas – they wash and guard
cars, they recycle litter. Some also resort to crime – petty theft, muggings, sex work,
drug dealing.
[33] What causes homelessness? The parties agreed that the causes are multi -
faceted. The City emphasized general factors beyond its control – organized crime,
national economic hardship, inadequate mental health care, the prevalence of
domestic violence, and persistent poverty and inequality. No doubt these all contribute
to people leaving or being forced to leave their homes to live on the street. None of
them are directly within the City’s power to address (although it has a role).
[34] But the City’s own policies and practices must also affect the extent and nature
of homelessness. Most homeless people live on the streets because there is no better
option available to them. It is fundamentally the City’s constitutional role to provide
better options. It is also the City’s role to assist homeless people. It is also th e City’s
role – which it accepts – to undo the legacy of apartheid spatial injustice. Part of the
13

reason the City struggles to offer affordable housing in the inner city is precisely
because apartheid policies forced Black and Coloured people to the peripheries.
[35] The City has adopted what it calls a “holistic and multi -faceted” approach to
addressing homelessness. It has a Street People Programme Unit that is designed to
reduce the number of people living on the streets by reintegrating them into the
community. The City says its approach has three “pillars”: rehabilitation; reintegration
and an immediate alternative to living on the streets. Rehabilitation addresses drug
addiction, and treatment for mental or other illnesses. Reintegration focuses on
employment, reunification with families, assisting people to obtain ID documents, and
developing skills to enable people to lead functional lives.
[36] It is the third pillar that is at the centre of this application – an alternative to living
on the streets. The City provides funding to NGO-run shelters that provide a place for
homeless people to sleep. But t he City also run s its own safe spaces to provide
temporary accommodation for homeless people. The City first opened the Culemborg
Safe Space in the City Centre on 29 June 2018 with space for 230 people. By 2020, it
had opened two more – Culemborg II and Paint City in Belville. Collectively they can
accommodate 700 people. The City plans to expand the safe spaces and open a new
one at the bottom of Ebenezer road for a further 300 people. It has allocated R142
million over three years to expanding and operating safe spaces. It estimates that it
spends R41 000 per occupant, per year.
[37] The safe spaces include the following practical amenities:
[37.1] Beds and shelter. They are not “a housing structure” but “more akin to a
dormitory-type structure, although each person has their own personal
14

space”. In Culemborg I it appears that people sometimes sleep outdoors
but protected from rain.
[37.2] Shared ablution facilities, including towels and toiletries provided by the
City. In addition to toilets, there are bucket showers and access to water.
The sites are cleaned daily.
[37.3] Two meals – breakfast and dinner – are provided per day, but no cooking
is allowed on the site because of the risk posed by fires.
[37.4] Locker space for residents to store their goods.
[37.5] 24-hour support, security and medical assistance is available.
[38] The City also provides a range of social services to people in safe spaces to
help them to reintegrate into society:
[38.1] The City provides job opportunities under the expanded public works
programme. It also links residents with other job opportunities and claims
“an excellent success rate”.
[38.2] It assists people to obtain identity documents, including facilitating the
funding for the applications.
[38.3] It provides access to a substance abuse rehabilitation programme. It
seems undisputed that street people have high levels of substance
abuse. The City claims its drug rehabilitation has an 80% success rate.
[38.4] The City offers development programmes geared for reintegration.
These include trauma therapy and family strengthening program mes. It
15

also provides computer skills training, helps people write their CVs, and
apply for jobs. It even gives them appropriate clothing for, and transport
to, job interviews.
[39] The goal of the safe spaces is to be temporary because the City wants to help
people off the streets, not send them back to the streets. The hope is that the services
they provide will allow people to reintegrate with their families or find employment that
will enable them to move off the streets.
[40] These facilities are offered free of charge to any person willing to take up the
City’s offer and accept social support. All three are located near business districts. The
two Culemborg sites are in the CBD, and the Paint City site is near the Belville
business district.
[41] But there are strings attached.
[41.1] Safe spaces are weapon free, drug free and alcohol free. People are
searched on entry. If people arrive intoxicated, they are not allowed entry
until they are sober. The City explains that these rules are essential to
“protect the dignity and health and safety of all the persons utilising” the
safe spaces. Without them, there would be “social discord” which would
negatively affect those trying to rehabilitate and reintegrate.
[41.2] Residents of safe spaces are either required or encouraged to leave the
safe spaces during the day. The ordinary rule is that they must leave
between 8:00 and 17:00. But that rule is flexible, and – as I detail later –
the City has relaxed it even further for the Occupiers.
16

[41.3] The accommodation is primarily divided by gender for safety reasons.
But some couples’ accommodation is also available. The City has
guaranteed that couples’ accommodation will be available for all the
Occupiers if they are evicted.
[41.4] Safe spaces are not meant to be permanent. For those who use them
voluntarily, they are ordinarily required to leave after six months. The
purpose of the safe spaces is to provide a base to enable street people
to rehabilitate and reintegrate, not to provi de a permanent home. The
City has, again, been willing to modify that rule for the Occupiers who
choose to take up its offer of access to safe spaces.
[42] The City’s approach should not be mistaken as being designed solely to benefit
homeless people – although I have no doubt that it is and that it does. The City is plain
that its aims are also to ensure that all its residents can use and enjoy streets and
other public places. They cannot do so – the City claims – if those spaces are occupied
by homeless people. The safe spaces are part of its plan to reduce the number of
homeless people living on the streets. But the C ity realises it can only achieve that
goal by helping them find somewhere else to live.
[43] In the year from July 2021 to June 2022, 1 813 people were helped off the
streets and 2 799 people participated in development programmes at City -run safe
spaces. The City offered 936 EPWP referrals and 566 referrals for social grants,
identity documents, specialized care or substance abuse.
[44] Independent NGOs have also supported the increased use of safe spaces. The
Occupiers provide a report prepared for the Hope Exchange which advocates for more
safe spaces and refers to “the current success of Safe Space[s]”. While I place limited
17

value on the Hope Exchange report as it was prepared by students, not experts, the
Occupiers did not provide expert evidence to support their contention that the safe
spaces were an inappropriate or ineffective measure in helping homeless people off
the streets.
[45] In addition to lauding the benefits of its safe spaces, the City argues that the
ordinary Emergency Housing Programme ( EHP) that it ordinarily offers when people
are evicted is not suitable for the Occupiers. It argues that the “holistic” intervention it
offers is better suited to the needs of homeless people, than a house in a temporary
relocation area (TRA). It offers three reasons.
[46] The first (and best reason) is that TRAs are all located far from the city centre;
there are none in the CBD. But the Occupiers – like all street people living in the CBD
– depend on proximity to the CBD for their livelihoods. That is where the opportunities
for them to earn a living exist, not in Blikkiesdorp or Wolwerivier. Even if spaces were
available in TRAs, the City contends the Occupiers could not live there and would
return to the CBD. The Occupiers do not dispute this – they seek other temporary
accommodation in the city centre.
[47] Second, the City argues that the safe spaces are targeted to assist homeless
people to reintegrate and rehabilitate. They come with a range of support – from the
material (food, blankets and toiletries), to the developmental (job- finding, trauma
support, identity documents, and substance abuse rehabilitation). Assignment of a
structure in a TRA comes with none of that support, making it less likely the homeless
person will escape homelessness. The basic point is that what is keeping many people
homeless is not just access to a house, but the ability to function effectively off the
streets.
18

[48] Third, it argues that there is no space in its existing TRAs and that it has no
funds to construct new ones for the Occupiers. It claims the cost for each emergency
housing opportunity will be between R55 000 (for an 18m2 structure) and R81 000 for
a (30 m2 structure), assuming the City already owns the land. But that is not for land
located in the CBD . By contrast, the safe spaces – in which the City has already
invested – are a far more efficient mechanism. The numbers do not seem drastically
different – R55 000 for a TRA structure and R41 000 per year for a safe space . The
efficiency seems to lie in the proximity to the CBD and the additional services safe
spaces provide.

THE OCCUPIERS
[49] Who are the occupiers? This is a question of some complexity.
[50] Attached to the notice of motion is a list of 114 people who the City described
as the First Respondent. They are all the people who the City had identified as living
on the Properties. There were 30 people living on Buitengracht, seven on FW De Klerk
Boulevard, three at Foregate Square, 20 on Helen Suzman Boulevard, 44 on Strand
Street, four by the Roggebaai gas turbine, and six at the Mill Street bridge.
[51] But homeless people are, understandably, transient. Since the notice of motion
was filed in December 2022, people have moved onto the Properties while others have
left.
[52] In their answering affidavit, the Occupiers – who are represented by the
Socio-Economic Rights Institute of South Africa ( SERI) – provided the details of 54
19

people who were living on the various properties , as well as some of their partners .
Many of them also had partners or family living with them.
[53] I cannot recount all their stories in this judgment. But I think it is important to
recall some of them. They explain how the Occupiers became homeless, what their
lives are like, and what they want for their futures:
[53.1] Mr Mquqa has been living on the streets for five years. He had to leave
his home because he lost his job and could not afford rent. He did not
finish school because of abuse by his family. He now lives on FW De
Klerk Boulevard, where he has been since 2017. He explains that
although the structure he lives in “is constructed with salvaged wooden
boards and plastic sheets” and “does not boast the strength of a brick -
and-mortar building, it is my only home and holds profound significance
for me.” He makes about R80 a day by assisting people to park their
cars, and by collecting recycling.
[53.2] Kashifa Williams is 37. She used to live with her family in Manenberg.
She was “kicked out by her family” and came to the inner city to find
work. She and her partner Leighton Vlok live at the FW De Klerk
Boulevard site. She makes a living collecting recycling and makes
roughly R100 a day. She is willing to move to the safe spaces but has
concerns about their rules and location.
[53.3] Zeinab Sutria is 43 and lives on Buitengracht Street with her partner,
Niezaar Abdula. She moved there in January 2023. Ms Sutria used to
live in Manenburg, but came to the inner city to find employment, without
any luck. They survive off informal work and make R80- R100 per day.
20

They support children who do not live with them. They are not willing to
move to the safe spaces because they would not be able to live together
and would be locked out during the day.
[53.4] Denis Fortuin has lived at the Buitengracht site since 2020. She was
rendered homeless when her family home was sold. She moved to the
inner city to try to find accommodation and employment. She has applied
for an RDP house but has not had any feedback. She lives with her
partner Immanuel Adams. They do informal work and make about R80
a day. She is reluctant to relocate to the safe spaces because she is
worried about whether she will be able to make a living.
[53.5] Unathi Noyi used to live with his family. But he became embroiled in
activities that disrupted his schooling and forced him to leave his family.
He used to live in a safe space, and on the street in District 6 and now
lives on the Strand Street site. He earns about R90 a day and has no
problem relocating as long as he can make a living and have safe and
secure accommodation.
[53.6] Sithembiso Kupiso also lives on the Strand Street site. He ended on the
street because he lost his job, could not pay his mortgage, and lost his
house. He was left homeless. He moved to the inner city to seek
employment. He earns about R80 a day and would be willing to relocate
if he could continue to earn a living.
[53.7] Terisa Townsend is 41 and has been living in the inner city since 2011.
She makes just R30 a day. She used to live with her family but left when
their relationship broke down. She previously lived near a McDonalds,
21

but was evicted and placed in S afe Space 1. She was forced to leave
the safe space because of capacity constraints and returned to the
streets. She does not want to return to the safe spaces because she
fears again being forced to leave and because she was poorly treated
last time she was there.
[54] Many of the Occupiers struggle with drug or alcohol addiction. Most suffer from
chronic illnesses. Some have skills but no jobs. Most make a living doing odd jobs in
the City – guarding or washing cars, collecting recycling or “skarreling”. Some of them
have been at the same site for years, others have moved multiple times, either for their
own reasons or because they were chased away, sometimes by the City’s officials.
Some have applied for housing from the City but have not received it.
[55] As the examples above show, the Occupiers have different attitudes to the safe
spaces. In the answering affidavit, several indicated that they would be willing to go to
the safe spaces if the rules around access and/or partners were amended or relaxed.
Others had had bad experiences or had heard negative reports from those who had
been to the safe spaces. Some did not want to give up the comparative freedom they
enjoy living on the pavements.
[56] This is just a snapshot. As Mr Mquqa – the deponent to the answering affidavit
– put it: “Each of us has experienced a great deal more than we are able to say in this
affidavit. Each of us hopes for a great deal more than we have experienced in our lives
so far.” They are stories of suffering, setbacks and failure. But they are also stories of
perseverance in the face of extreme obstacles, and of survival against all odds.
[57] In the replying affidavit, the City provided an updated list of the Occupiers who
were named in the answering affidavit (it did so as part of its further engagement with
22

the Occupiers that I describe below). The City engaged with 87 people in total. The list
provides the person’s name, the names of those they lived with, their employment,
and their attitude to the City’s offer of a safe space.
[58] Some were amenable to relocating to the safe spaces, others would be if the
rules were altered, while some were completely opposed. The concerns about the safe
spaces were generally about their ability to work, whether they would be allowed to
live with their partners, and concerns about the rules or conditions either based on
prior experience or what they had heard from others. While the 54 occupiers are the
same in the answering affidavit and the list provided in reply, there are differences in
the description of the partners, children and family members with whom that cohabit.
[59] On 5 October 2023, shortly before the hearing, at the Court’s request, the City
conducted a further process where it sought to count the number of structures at each
of the sites. There were a total of 107 structures across the seven sites, ranging from
one at Mill Street Bridge to 22 at Strand Street. The City estimated there were 214
people occupying all seven sites.
[60] At the hearing, the Occupiers produced a “composite” list that included all the
people who had been mentioned in the notice of motion, the answering affidavit, or the
replying affidavit. It has a total of 272 people. The list just provides basic details –
name, age, which site they occupy – and in which document they were mentioned.
[61] Following the hearing, the City and SERI conducted another engagement (more
details to follow) that again sought to determine how many people were occupying the
properties and, particularly, how many Occupiers were elderly or disabled, how many
children were present, and how many Occupiers were living with their partners. There
23

was some disagreement between the City and SERI about the numbers . But the
following emerged:
[61.1] There were 113 structures across the seven sites.
[61.2] The City reported that there were 99 people on the sites who engaged
with the City, and a further four who refused to engage. SERI counted
125 people, including those not present. Ultimately, it seems there were
between 99 and 140 people living on the sites.
[61.3] Of th e 99 the City engaged with, 37 accepted the City’s offer of
accommodation at the safe space (37%). According to SERI, 64 were
willing to move to the Safe Spaces out of (at most) 140 (46%).
[61.4] The City identified 14 couples. SERI identified a further six couples,
making a total of 20.
[61.5] There was one person with a disability. The oldest Occupier was 57 or
58 years’ old.
[61.6] The City identified three minor children – VD and ZD aged 5 and 2, and
PA aged 14. SERI claimed there were a further four minor children, but
further investigation by the City revealed they had all moved elsewhere.
[62] The Court is deeply appreciative of the work that both the City and SERI did to
identify the Occupiers. The information cannot be precise. The nature of the sites is
that people move in and out of them. There has also been a delay of six months from
the time of the latest report to the date of this judgment. The position may now be very
24

different. But the reports do give a sense of the scale of the problem, the demographics
of the Occupiers, and their attitudes towards the safe spaces.
[63] There is a final issue to address in describing the Occupiers – who represents
them? For this purpose, there are two groups of Occupiers: those named in the
answering affidavit; and those not named. The Occupiers who are expressly named
in the answering affidavits and who have filed confirmatory affidavits are plainly
represented by SERI. But SERI also initially claimed to represent all other unnamed
Occupiers, and the answering affidavit was purportedly filed on behalf of all the
Occupiers. Is that permissible?
[64] When questioned, SERI properly explained that it could only formally represent
those Occupiers who had given it express instructions. It could not claim to represent,
in the ordinary sense, those Occupiers who had never instructed them, and whose
details they may not even be aware of. Howev er, it contended that it still acted in the
interests of all the occupiers whose fates are tied together. As Ms Bhengu put it: “SERI
considers itself under a general ethical duty to seek to prote ct the interests of as -yet
un-named and unidentified persons who may have come to live at one of the seven
sites”.
[65] Public interest litigation is often messy. Representing people who are
vulnerable and disempowered is not easy. While it was inaccurate for SERI to say that
it formally represented unnamed Occupiers, I do not think it should be criticised. There
was no reason to believe that the interests of its specific clients would differ from the
interests of the Occupiers as a whole. In my view, SERI eventually adopted the correct
approach – it was on record only for specific named Occupiers. But it made
submissions in the interests of all the Occupiers. It did so in the best tradition of public
25

interest litigation in this country which seeks to make sure that all people have a voice
in litigation that concerns them.

The Shape of the Litigation
[66] The application to evict was launched in December 2022. On the City’s version,
people began occupying the seven properties in April 2020, largely as a result of the
lockdown imposed during the Covid- 19 pandemic.
8 The City explains that it did not
want to evict the Occupiers during Covid- 19, and so waited until the pandemic was
over and the various restrictions had been lifted. It had previously offered all the
Occupiers it could identify a spot at a safe space. Some accepted. Others turned down
the offer and remained.
[67] The notice of motion in the eviction application was not initially linked to the
provision of alternative accommodation; although the founding affidavit made it clear
that the City believed the safe spaces were an adequate alternative. The Occupiers
initially complained that the City sought their eviction without any guarantee of even
accommodation at the safe spaces. By the hearing, this issue had fallen away and the
City accepted that the Occupiers could only be evicted if it could provide them with
accommodation at a safe space.
[68] T he Occupiers were initially unrepresented and the matter was set down for
hearing on 19 April 2023. By that time, the Occupiers (or at least some of them) had

8 Some of the Occupiers claim they have been occupying the properties for longer than this, but nothing
turns on the dispute.
26

obtained legal representation. The parties agreed to afford the Occupiers until 30 June
2023 to file their answering papers.
[69] The answering papers oppose the eviction on t wo broad grounds. First, the
Occupiers allege the City had not meaningfully engaged. Second, they argued that the
safe spaces were not suitable alternative accommodation because of the rules they
impose, and their failure to cater for families and couples.
[70] The Occupiers also, rightly, complain that while this application was pending,
the City attempted to demolish some of their homes. On 16 May 2023, the City’s
officials demolished about 15 structures at the FW De Klerk site. The City initially
denied the demolition, then admitted it and claimed it was permitted under a court
order. However, the City eventually offered R1 700 to each affected occupier as
recompense and undertook that no further demolitions would be conducted at the
Properties while this application was pending. It was improper (and probably unlawful)
for the City to unilaterally demolish the structures while seeking an eviction order. The
City appears to have, grudgingly, admitted its error.
[71] The Occupiers also filed a counter-application challenging the constitutionality
of certain provisions of the City’s Streets, Public Places and Prevention of Noise
Nuisance By-Law. The City had relied on the By -Law as one of the bases on which
the Occupiers’ occupation of the Properties was unlawful. The Occupiers argued that
the provisions of the By -Law criminalized homelessness and permitted eviction
contrary to s 26(3). The City abandoned reliance on the By-Law and the Occupiers did
not persist with their counter -application. Part of the reason is that there was a
separate challenge to the constitutionality of the By-Law on similar grounds.
27

[72] As I narrated earlier, the parties had also engaged between the filing of the
answering and replying affidavits. More on that and whether it was adequate later.
[73] That was the shape of the litigation when it was allocated to me. I must next
explain the steps I took, beyond merely hearing the application.
[74] A court’s role in an eviction application is not the same as its role in other civil
litigation. It has an additional duty to ensure it has all the information it needs to satisfy
itself that an eviction will be just and equitable. As the Constitutional Court explained
in PE Municipality, courts are “entitled to go beyond the facts established in the papers
before it” particularly where the evidence “leaves important questions of fact obscure,
contested or uncertain ”.9 When necessary, a court must engage in “ active judicial
management according to equitable principles of an ongoing, stressful and
law-governed social process”.10
[75] But there are limits on how far a court can legitimately go. “ A more active role
in managing the litigation does not permit the judge to enter the arena or take over the
running of the litigation. ”
11 While a court asked to evict people can ask for more
information and can construct remedies that ensure evictions are executed only when
appropriate, it cannot redefine the issues between the parties.12

9 Port Elizabeth Municipality v Various Occupiers [2004] ZACC 7; 2005 (1) SA 217 (CC); 2004 (12)
BCLR 1268 (CC) at para 32.
10 Ibid at para 36.
11 City of Johannesburg v Changing Tides 74 (Pty) Ltd and Others [2012] ZASCA 116; 2012 (6) SA 294
(SCA); 2012 (11) BCLR 1206 (SCA); [2013] 1 All SA 8 (SCA) at para 27.
12 Fischer and Another v Ramahlele and Others [2014] ZASCA 88; 2014 (4) SA 614 (SCA); [2014] 3 All
SA 395 (SCA) at para 13.
28

[76] This will sometimes be a tightrope that judges must walk with care: Intervening
enough to ensure they can do justice in the case, but not so much that they become
an active participant in the litigation. In this case, I intervened on two occasions.
[77] I held a meeting with the legal representatives for the parties on 3 October 2023,
a week before the hearing. At the meeting, I raised questions about issues that were
“obscure, contested or uncertain” on the papers. I sought clarity either immediately, or
through affidavits the parties were invited to file before the hearing. The issues on
which I sought the parties’ assistance were:
[77.1] On whose behalf the S ERI was acting. As I mentioned earlier, SERI
claimed to be acting on behalf of both the named Occupiers, and the
unnamed Occupiers.
[77.2] The details of the persons occupying the properties. Having considered
the papers, there was not only a lack of clarity, but also an inconsistency
in the information provided by the City and the Occupiers.
[77.3] The availability of couples’ accommodation at the safe spaces.
[77.4] Whether the City would permit Occupiers who were evicted and decided
to move to a safe space to stay there beyond six months and until they
found alternative accommodation.
[78] Both the City and the Occupiers filed affidavits detailing their responses. I deal
with them thematically when I reach those questions.
[79] After the second day of hearing, I held a meeting with counsel for the parties. I
enquired whether the parties would be amenable to further engagement on the make
29

up of the Occupiers, and the operation and availability of the safe spaces. T here
seemed to be a disconnect between the City’s explanation to the Court about what the
safe spaces entailed, and the Occupiers’ understanding of what relocation to a safe
space meant. The City ’s position had also shifted to guarantee couple’ s
accommodation and stays beyond six months. As a result it was very difficult to predict
how many people were likely to take up the offer of safe space if they were evicted.
There was also uncertainty about the number of couples, children, elderly and disabled
people, and the capacity of the City to accommodate them.
[80] The parties engaged in correspondence about my suggestion subsequent to
the hearing. There was a tension between them about the ambit and purpose of further
engagement. The City was willing to engage on a defined basis, but worried about
re-opening the litigation, or allowing the Occupiers to raise new complaints. The
Occupiers saw it as an opportunity for a more open engagement about all the options
that could be made available to the Occupiers, and the suitability of the safe spaces.
[81] Following the exchange of correspondence which was provided to the Court, I
issued an order on 20 October 2023 that reflected what the parties had agreed on. It
required the parties to engage on:
[81.1] Ascertaining the total number of persons occupying the subject
properties who intend taking up the City’s offer of accommodation at the
safe spaces;
[81.2] Ascertaining the number of persons who require couples’
accommodation at the safe spaces;
30

[81.3] Determining the number of elderly persons and disabled persons, and
whether the safe spaces pose any reasonable constraints to them
accepting the offer of alternative accommodation; and
[81.4] Determining whether there are any more minor children on the subject
sites (in addition to those identified in the affidavits filed to date) and to
obtain full details of the minor children.
[82] It also required the City’s officials to be available to engage with the Occupiers
on: (a) what is offered at the safe spaces; (b) the rules applicable at the safe spaces;
and (c) the details of the operation of the safe spaces. The order then provided a
timeline for the filing of further affidavits by the parties on the results of their
engagement. Finally, paragraph 9 of the order expressly stated that it did not preclude
the parties from engaging on other issues, and reporting to the Court on the outcome
of those engagements if they deemed it appropriate.
[83] The City filed a report, to which SERI responded, and to which the C ity filed a
reply. I deal with the contents where they are relevant. Overall, the exercise elicited
significant useful information.
[84] I do not think I exceeded my powers in the manner I engaged with the parties.
I needed clarity on certain issues in order to determine whether eviction was just and
equitable and, if so, what order I should make. This is not an ordinary eviction. There
was little to guide me on how to manage the eviction of a transient group of people
who were currently homeless. This was a challenge not only for the court, but for the
parties too. I needed more information. At the same time, I have been very careful not
to introduce new issues on either side – not to give the City new bases to justify
eviction, and not to raise new defences for the Occupiers. Rather I sought clarity on
31

the parties’ positions, or better and updated evidence that was relevant to the issues
as they were already defined by the parties.
[85] In my view, it would not have been useful to refuse eviction because, for
example, the City had not provided sufficient information about the number of children,
elderly or disabled people. While in an “ordinary” eviction application where the
occupiers are stable over time, it could and should provide that information in its
founding papers, the City faced legitimate difficulties in obtaining that information for
these Occupiers. So did the Occupiers’ own legal representatives. Similarly, it was not
the City’s fault that it could not predict how many Occupiers would take up its offer for
safe spaces if they were evicted.

EVICTION
[86] Is an eviction order just and equitable in these circumstances? To answer that
primary question, I consider the following topics:
[86.1] The basic requirements for eviction are met;
[86.2] The City has meaningfully engaged with the Occupiers;
[86.3] The City was not obliged to offer housing under the Emergency Housing
Programme;
[86.4] The safe spaces constitute suitable alternative accommodation;
[86.5] In all the circumstances, eviction is just and equitable; and
[86.6] The timing and details of the eviction order.
32


THE BASIC REQUIREMENTS FOR EVICTION
[87] The starting point is s 26(3) of the Constitution which prohibits evictions without
an order of court made after considering all relevant circumstances. Parliament has
enacted legislation to give effect to s 26(3) – PIE.
[88] Prior to the Constitution and PIE, eviction was characterized by abuse, paid no
heed to the impact on those evicted, and was generally used as a tool to enforce White
domination and control. The “manifest objective” of PIE is “overcoming [those] abuses
and ensuring that evictions in future took place in a manner consistent with the values
of the new constitutional dispensation.”
13
[89] PIE creates two distinct mechanisms for eviction. Section 4 permits evictions
by the owner or person in charge of a property. Section 6 permits eviction at the
instance of an organ of state, even if it is not the owner of the property. The basic
requirements are the same – the occupier must occupy the property unlawfully, and
the eviction must be just and equitable. And the procedures for eviction under both
sections are the same.
14 But there are differences:
[89.1] An organ of state can only apply for eviction under s 6 if its consent is
required for occupying the property, or “it is in the public interest to grant
such an order”.
15 Public interest “includes the interest of the health and

13 Port Elizabeth Municipality v Various Occupiers [2004] ZACC 7; 2005 (1) SA 217 (CC); 2004 (12)
BCLR 1268 (CC) at para 11.
14 PIE s 6(6).
15 PIE s 6(1).
33

safety of those occupying the land and the public in general.”16 There is
no similar requirement under s 4.
[89.2] Unlike s 4, s 6 specifically lists factors a court must consider in deciding
whether eviction is just and equitable: “(a) the circumstances under
which the unlawful occupier occupied the land and erected the building
or structure; (b) the period the unlawful occupier and his or her family
have resided on the land in question; and (c) the availability to the
unlawful occupier of suitable alternative accommodation or land.”
[90] This application was brought under both s 4 and s 6. The City claims that both
thresholds are met. Given the approach I take to the case, nothing turns on whether
the application is evaluated under s 4 or s 6. I accept that the City has brought this
application not only to vindicate its rights as owner, but also in the public interest, and
to vindicate the rights of members of the public who use the CBD.
[91] There are two central factors a court must almost in variably consider in an
eviction application by an organ of state , and that take centre stage in this case. I
consider each in more detail below, but mention them briefly now:
[91.1] When a municipality is seeking the eviction, a court will be reluctant to
grant an eviction if the municipality has not meaningfully engaged with
the occupiers.
17 Meaningful engagement is meant to avoid the need for
eviction by finding alternatives. As Yacoob J explained in Olivia Road :

16 PIE s 6(2).
17 Occupiers of 51 Olivia Road, Berea Township and 197 Main Street Johannesburg v City of
Johannesburg and Others [2008] ZACC 1; 2008 (3) SA 208 (CC) ; 2008 (5) BCLR 475 (CC) at paras
9-22.
34

“Engagement has the potential to contribute towards the resolution of
disputes and to increased understanding and sympathetic care if both
sides are willing to participate in the process.”18
[91.2] Generally, a court will not grant an order of eviction if there is a risk that
the order of eviction will render the occupiers homeless. That is not an
absolute rule – the Supreme Court of Appeal put it this way: “an eviction
order in circumstances where no alternative accommodation is provided
is far less likely to be just and equitable than one that makes careful
provision for alternative housing.”
19 Before it grants an eviction, a court
must know where the evictees will live, and will very rarely grant the
eviction if no suitable alternative accommodation is available.
[92] This case is somewhat unusual because the City seeks to evict people who are
already homeless. Yet it was common cause that this was the appropriate legal route
for the City to follow. I agree, but should briefly explain why.
[93] Section 26(3) whenever a person is evicted from a “ home”. To give effect to
that, PIE defines eviction widely as “to deprive a person of occupation of a building or
structure, or the land on which such building or structure is erected, against his or her
will”. It gives a similarly broad meaning to “building or structure” , so that the term
“includes any hut, shack, tent or similar structure or any other form of temporary or
permanent dwelling or shelter”.

18 Ibid at para 15.
19 Changing Tides (n 11) at para 15.
35

[94] The Occupiers shelters, while rudimentary, were undoubtedly their homes.
They saw them as their homes. They are also “structures” as defined in PIE. While all
the parties referred to the Occupiers as “homeless” that is, in some sense, inaccurate;
they do have homes. Depriving them of occupation of their homes , however basic,
constitutes an eviction.20
[95] The City was right to recognize that the consequence of removing the
Occupiers from its land would be an eviction and that it therefore required an eviction
order under PIE.
[96] The parties also accepted that PIE’s procedural requirements for an eviction
have been met. There was also no argument from the Occupiers that they are
occupying the Properties lawfully. They accept they are unlawful occupiers as defined
by PIE. The precise reason their occupation is unlawful does not seem to matter much.
They do not have the City’s consent to occupy. Their occupation is also contrary to the
Roads Ordinance 19 of 1976.
[97] The real debate was whether eviction as just and equitable in light of the two
considerations I mentioned above: was there meaningful engagement, and will there
be suitable alternative accommodation?

20 This case is distinguishable from Ngomane and Others v City of Johannesburg Metropolitan
Municipality and Another [2019] ZASCA 57; [2019] 3 All SA 69 (SCA); 2020 (1) SA 52 (SCA) where the
SCA held there was no eviction because, while people were removed from land they were occupying,
there were no structures on the land, just loose wooden pallets, cardboard boxes and plastic. While the
Occupiers here used many of the same materials, they had constructed structures.
36


Did the City Meaningfully Engage?
[98] The requirement of meaningful engagement prior to an eviction was birthed in
Olivia Road.21 The case did not concern eviction under PIE, but an application under
the National Building Regulations and Building Standards Act 103 of 1977 to remove
the occupants on the basis that the building they occupied was unsafe. Yacoob J held
that the Constitution “ obliges every municipality to engage meaningfully with people
who would become homeless because it evicts them.”22
[99] But the Court was specific about what meaningful engagement requires. It did
not require a municipality “to make provision for housing beyond the extent to which
available resources allow. As long as the response of the municipality in the
engagement process is reasonable, that response complies with section 26(2)” of the
Constitution.23 At one point, the Court held that there must either be meaningful
engagement “or, at least, that the municipality has made reasonable efforts towards
meaningful engagement.”24
[100] In Joe Slovo, Ngcobo J (as he then was) emphasized that meaningful
engagement must be conducted “in good faith and with a willingness to listen and,

21 Olivia Road (n 17).
22 Ibid at para 18.
23 Ibid.
24 Ibid at para 21.
37

where possible, to accommodate one another.” 25 Still, the decision of what to offer
those facing eviction “ lies with the government” as long as it is “ informed by the
concerns raised by the residents during the process of engagement.”26
[101] Joe Slovo is an interesting example. The case produced multiple judgments,
with none commanding a clear majority. All of them saw defects in the engagement
process. But none of them refused eviction as a result.
[101.1] Yacoob J held that “t he state could and should have been more alive
to the human factor and that more intensive consultation could have
prevented the impasse that had resulted”.27 And he held that “[i]t would
have been ideal for the state to have engaged individually and carefully
with each of the thousands of the families involved. ”28 But ultimately,
he concluded meaningful engagement “involves realism and
practicality. There has been reasonable engagement.”29
[101.2] Similarly, O’Regan J held that despite “the failure to have a coherent
and meaningful strategy of engagement ” the state was entitled to an
eviction order.30 The engagement, she held, had “ not been coherent

25 Residents of Joe Slovo Community, Western Cape v Thubelisha Homes and Others [2009] ZACC
16; 2009 (9) BCLR 847 (CC) ; 2010 (3) SA 454 (CC) at para 243 ( Moseneke DCJ and Sachs J
concurring).
26 Ibid.
27 Ibid at para 113.
28 Ibid at para 117.
29 Ibid.
30 Ibid at para 302.
38

or comprehensive and that at times it has been misleading”.31 But still,
she concluded that given the size and novelty of the problem it was
unsurprising there were deficiencies in the engagement process. T he
state had engaged, even if the engagement was imperfect.32
[102] Ordinarily the topic of meaningful engagement is relatively straightforward –
what are the options to avoid eviction . That normally involves either allowing the
occupiers to stay where they are, or figuring out what alternative accommodation is
available. In Saratoga Avenue the Constitutional Court left open the question whether
meaningful engagement “would entitle all evictees to contest the quality of temporary
accommodation being provided to them”.33
[103] In light of these cases, what exactly does meaningful engagement require?
There is no comprehensive definition, and I do not intend to provide one . But in my
view it must include at least the following elements which are relevant in this case:
[103.1] The municipality must inform the occupiers about the consequences of
eviction, what options are available to them if they are evicted,
including what the municipality can or must do to assist them.
[103.2] The municipality must also afford the occupiers an opportunity to
express their views about those options, and to suggest alternatives.

31 Ibid.
32 Ibid.
33 Occupiers of Saratoga Avenue v City of Johannesburg Metropolitan Municipality and Another [2012]
ZACC 9; 2012 (9) BCLR 951 (CC).
39

[103.3] The municipality must listen to those opinions and see if it is possible
to accommodate the concerns.
[103.4] Both parties must engage in good faith with the hope of reaching an
agreed solution. They must be open to alternative proposals and
compromises.
[103.5] Meaningful engagement may require some back -and-forth; it should
be more than simply an offer which occupiers must take or leave.
[103.6] Engagement can happen individually, or at a collective level if the
municipality is satisfied that the occupiers have properly chosen people
to engage on their behalf.
[104] Having said all that, there are also limi ts on what meaningful engagement can
require. Courts must not impose a process that is so burdensome it unduly inhibits a
municipality from legitimat ely pursuing the eviction of those unlawfully occupying its
land. With that in mind, there are two important restrictions on the scope of meaningful
engagement:
[104.1] W hile an agreed resolution is ideal, meaningful engagement does not
require agreement . An eviction can be granted even though the
occupiers remain dissatisfied with the options the municipality has
presented.
[104.2] Meaningful engagement is about resolving a specific situation where
eviction is threatened. It is not a mechanism to resolve broader policy
disputes. A municipality conducts meaningful engagement within its
existing policy framework and budget. It is not required, in order to
40

show it has meaningfully engaged, to second-guess its policies or its
budget each time it seeks to evict people.
[105] What was the nature of the engagement in this case, and di d it meet these
requirements? There were three phases of engagement.
[106] First, b efore it brought the application the City alleges it interacted with the
Occupiers on multiple occasions, making offers of social assistance and alternative
accommodation. Some of the occupiers accepted those offers; others did not. The City
argues that its endeavours to engage were “entirely dependent on the co-operation of
the individuals concerned”. The high watermark of the interaction was collecting
information about the Occupiers, and making offers of alternative accommodation.
[107] At this stage, the Occupiers were neither represented, nor organized. There
was no person or group that could speak on behalf of the Occupiers. The City could
only interact with them individually.
[108] The Occupiers contend that the pre-litigation “engagement” consisted primarily
of the City’s law enforcement fining them, demolishing their homes, and chasing them
from one place to the next. At most, the Occupiers say the City conducted a “census”
of the Occupiers.
[109] The second phase occurred after the Occupiers filed their answering affidavit,
and before the City filed its reply. The engagement was specifically prompted by the
Occupiers’ allegation that the City had failed to meaningfully engage. While not
admitting any shortcomings, the City sought to do more. Following an exchange of
correspondence between the parties’ attorneys , the engagement was agreed to
between the parties and took place between 16 and 19 August 2023.
41

[110] The City’s version is that it engaged all the named respondents it could locate.
It sought information about each person, particularly: whether they had a partner,
whether they worked after closing hours, whether they understood the safe spaces’
rules (a copy of which was provided) and whether they had concerns about only being
able to stay there for six months. As I set out above, the process allowed it to provide
a better statement of the number of respondents, their circumstances, and what the
consequences of eviction would be.
[111] In the process, the City explained to the various Occupiers that: (a) they could
return to the Safe Spaces after hours if they were working; (b) there was some
provision for couples to live together; and (c) their stay would not be automatically
terminated after six months. The Occupiers’ attorneys were present throughout the
engagement.
[112] The City did not engage with the Occupiers about alternatives to the safe space.
It had already taken a decision that the only offer of alternative accommodation it could
make was for the safe spaces. Its engagement was about the details of that offer, and
seeking to understand the Occupiers’ possible concerns with that offer.
[113] But the City did change its position as a result of the engagement . It noted the
concerns about the lockout times and engaged with the management of the safe
spaces to emphasise that the rule had to be flexibly and reasonably enforced. It made
clear that, while the default rule was that people would vacate after six months, if they
had not found an alternative place to stay “they will remain at the Safe Space” subject
to ongoing assessment. It also recognized the need for couples’ accommodation,
although it did not (at this stage) commit to providing it for all who needed it.
42

[114] The Occupiers acknowledge that this second engagement was a step in the
right direction. But they still criticize it as “nothing but a box ticking exercise”. They
focus on the limited utility of seeking to ascertain the Occupiers’ personal details when
those could have been obtained from the Occupiers’ attorneys. They also say that the
engagement had a “pre- determined outcome” and that the City “has never been
prepared to seriously listen to the occupiers’ concerns and needs and to build the trust
necessary to find a workable solution”.
[115] The final phase of engagement occurred after the hearing in line with the order
I issued on 20 October 2023. The City again engaged with each of the persons present
on the seven Properties, trying to produce a more reliable list of the people occupying
the sites . It also obtained more information about people with disabilities, elderly
people, children, and those needing couples’ accommodation. It discussed the rules
of the safe spaces with the Occupiers – now including all the concessions it had made
through the litigation – and answered any questions. Depending on whose numbers
were used, either 37 of 99, or 67 of 140 people accepted the City’s offers of
accommodation at safe spaces. The City also allowed both the Occupiers and their
attorneys to view Safe Space 1.
[116] However, the City did not regard this as an opportunity to re- open issues that
had been addressed in the litigation or to re- argue whether the safe spaces were
suitable alternative accommodation. The purpose was limited to obtaining a better
sense of who occupied the properties and engaging with the occupiers about what
moving to the safe spaces would entail so they could decide whether to accept the
offer.
43

[117] The Occupiers’ view, after this engagement, was that it had “yielded partial
success”. But their objection remained that meaningful engagement was “meant to be
a two -way process where both parties actively participate in sharing ideas and
understanding”. It complained that the City had still taken too narrow an approach to
the engagement, that it was required to seek “mutually acceptable solutions”, and that
“both parties should buy into the solutions even if they have to compromise”.
[118] I conclude that, while imperfect, the City did meet its obligation to meaningfully
engage with the Occupiers. There are four factors that drive me to that conclusion.
[119] First, as I foreshadowed above, it is important to be clear about what the
purpose of meaningful engagement is, and what it is not. The purpose is to identify
alternatives to eviction for those occupiers. That could include avoiding relocation
altogether, or avoiding eviction because the occupiers are willing to accept an offer of
alternative accommodation. The goal is to discuss available options, and see if there
is something that can work for both parties.
[120] T
he purpose is not to challenge the municipality’s budget or policy decisions.
When it meaningfully engages with occupiers, a municipality does so within those
constraints. Occupiers are free to attack the legality and constitutionality of a
municipality’s policies that may not provide for what they think is a reasonable
alternative. But a municipality is not required to enter into those types of discussions
as part of the meaningful engagement prior to eviction.
[121] Meaningful engagement is also not required to find “mutually acceptable
solutions”. That is obviously the ideal, and parties should be open to the compromise
and agreement. But engagement that fails to reach that ideal outcome can still be
meaningful and reasonable. If a municipality gathers all the necessary information
44

about occupiers, provides them with all the options, listens to their concerns, and
responds to them, then it has meaningfully engaged, whether or not the occupiers like
the responses and the outcomes.
[122] Second, if there had been no further engagement after the City filed this
application, I may have found that there was not sufficient engagement. But there was
extensive post-application engagement. The first was prompted entirely by the City
and the second by this Court.
[123] To my mind, the issue for a court considering eviction is not whether there had
been meaningful engagement before the litigation is launched, but whether it had
occurred by the time it grants the eviction order. The purpose of requiring meaningful
engagement is not to offer a “defence” to eviction, but to ensure that occupiers are
treated with dignity and agency, that there has been an attempt to avoid eviction, and
that the parties have explored other plausible options.
[124] That should occur as early as possible and ideally before the case is launched.
But if it happens later and still fulfills the purpose, then the delay is not a basis to refuse
eviction. In Olivia Road the engagement happened only as a result of an order the
Constitutional Court granted after hearing the matter.
[125] Third, this is a novel and particularly difficult context for engagement. In most
other evictions, the occupiers are fixed not transient, and they are often organized.
That makes engagement far easier as the municipality knows with certainty with whom
it is dealing, and can engage collectively rather than individually.
[126] That was not the case here. The Occupiers have never been organized. Before
they were represented it was impossible for the City to engage them collectively. Yet
45

individual engagement was, in these circumstances, unlikely to be particularly fruitful.
My finding about what was reasonable in these circumstances may not apply in
different circumstances where occupiers are more permanent and organised.
[127] Fourth, the City clearly listened to the Occupiers. It has altered its stance based
on their complaints about the safe spaces’ rules. It has relaxed rules on lock out and
on the six-month residency rule. It has promised there will be couples’ accommodation
for all who need it. It has not closed its ears to the Occupiers complaints – it listened
and responded.
[128] It is true that it has not shifted on the basic position that safe spaces are suitable
alternative accommodation. But it was not required to. It was entitled to make an offer
and then defend that offer as reasonable alternative accommodation. If the safe
spaces are reasonable alternative accommodation, then the City can hardly be
criticized for not offering an alternative as part of the meaningful engagement process.
Any workable alternative in the City centre would require radical changes to the City’s
policies and budget – neither of which is a requisite part of meaningful engagement.
[129] In sum, the engagement was not perfect. The City can probably learn lessons
from this process and engage more effectively and respectively in similar situations in
the future. It should ideally do what it had achieved by the end of the litigation before
launching. But it was good enough. The City informed the Occupiers what options
were available, listened to their concerns about the alternative accommodation, and
responded to those concerns. The failure to reach agreement, and the failure to offer
different alternative accommodation does not mean the engagement was not
meaningful.

46

Was it Permissible to Offer Safe Spaces instead of TRAs?
[130] The Occupiers argue that the City wrongly did not offer the Occupiers
accommodation in TRAs built in terms of the E mergency Housing Programme. The
Emergency Housing P rogramme is part of the National Housing Code, which was
adopted under the Housing Act 107 of 1997. It provides for funding for municipalities
who need to offer emergency housing which used to include large scale evictions. 34
In practice municipalities have established s emi-permanent TRAs under this
programme where people who are evicted are often offered alternative
accommodation.
[131] The argument is that what the Occupiers really want is a house of their own ,
not temporary shelter in Safe Spaces . As I understood it, the argument was not that
the City was obliged to comply with the provisions for emergency housing under the
National Housing Code in the sense that it was legally precluded from offering anything
else. That argument would be inconsistent with the Constitutional Court’s holding in
Dladla
35 which I discuss below. It would also seem to me to be inconsistent with the
underlying constitutional obligation to act reasonably. The E mergency HP sets a
baseline or standard that can be accepted as reasonable unless and until it is
challenged. It does not preclude other measures that are also reasonable.

34 See Commando (n 3) at para 37.
35 Dladla and Another v City of Johannesburg and Others [2017] ZACC 42; 2018 (2) BCLR 119 (CC);
2018 (2) SA 327 (CC).
47

[132] Rather, the Occupiers contended that the offer of places in the Safe Spaces
alone was not reasonable because the City had not explained why the Occupiers could
not be accommodated in houses built in terms of the EHP.
[133] The City’s answer was that accommodating the Occupiers in TRAs was neither
possible nor desirable. The City argued, first, that there were no spaces available in
the TRAs. This on its own is not a compelling answer; an eviction could always be
delayed until those spaces became available as courts often do.
[134] But the City’s other answer is compelling. Accommodation under the EHP is
not desirable for these occupiers because TRAs are far from the CBD and
employment, and offer none of the additional material or developmental services
available at the Safe Spaces. There are no TRAs in the inner City, the City does not
plan to build any, and does not have the resources to do so. The City – following the
warning in Blue Moonlight 36 – provided details of its available funds to show why it
could not provide temporary accommodation in the CBD other than the safe spaces.
[135] The City’s po licy of not providing “ordinary” alternative accommodation in the
inner City was recently upheld by the Supreme Court of Appeal in Commando. 37 A
group of people who faced eviction from Woodstock challenged the reasonableness
of the City’s offer of alternative accommodation at Wolwerivier. The Supreme Court of

36 City of Johannesburg Metropolitan Municipality v Blue Moonlight Properties 39 (Pty) Ltd and Another
[2011] ZACC 33; 2012 (2) BCLR 150 (CC); 2012 (2) SA 104 (CC) at para 74.
37 Commando (n 3). That decision has in turn been appealed to the Constitutional Court and judgment
has been reserved. It is possible that Court will disagree with the SCA’s conclusions. But unless and
until Braamfontein reverses Bloemfontein, I am bound by the SCA’s judgment.
48

Appeal reversed a decision of this Court38 that the offer and the City’s underlying policy
was unreasonable. Mabindla- Boqwana JA held: “ The fact that no provision is made
for … emergency housing needs in the inner city, does not render the choices made
by the City irrational or unreasonable.”39
[136] The Occupiers’ position is somewhat ambiguous. On the one hand they claim
they just want to be offered a place in a TRA. But the deponent to the answering
affidavit makes it clear that what they really want is “well-located affordable housing”.
But that is simply not on the table. The Occupiers have not directly attacked the City’s
policy of not offering temporary EHP housing in the inner city, and I would in any event
be bound by the judgment in Commando that holds it is not obliged to do so.
[137] The question is whether the safe spaces are suitable alternative
accommodation or not. That must obviously be assessed while considering what else
might be available. But the only other option any of the parties identified is TRAs under
the EHP that are far from the City centre and which the Occupiers themselves accept
would not meet their needs.
[138] Even if it was offered, the Occupiers would be unlikely to remain so far from the
CBD. They would have no choice but to return, still with no accommodation other than
on the pavements. That outcome would serve nobody. The City’s goal of offering the
safe spaces is to avoid that cycle. They were entitled to make that offer. The question
is whether it was reasonable.

38 Commando and Others v Woodstock Hub (Pty) Ltd and Another [2021] ZAWCHC 179; [2021] 4 All
SA 408 (WCC).
39 Commando (n 3) at para 60.
49


Are the Safe Spaces Reasonable Alternative Accommodation?
[139] The City meaningfully engaged, and is not bound to offer only housing under
the EHP. But is the offer that it made – S afe Space 1 in Culemborg – reasonable?
That question raises the following subsidiary issues:
[139.1] What is he standard of adequate alternative accommodation?
[139.2] Is the standard different for people who are already homeless?
[139.3] Is it permissible for the occupiers to challenge the physical
infrastructure of the safe spaces?
[139.4] What did the Constitutional Court decide in Dladla?
[139.5] Are the Occupiers’ complaints about the rules imposed at the safe
spaces valid?

Adequate Alternative Accommodation
[140] Our courts consistently require “suitable” or “adequate” alternative
accommodation as a condition for eviction of people who face the risk of
homelessness. But what makes alternative accommodation “adequate”?
[141] The first point to make is that what is adequate accommodation extends beyond
merely the physical attributes of the structure. As the International Committee on
Economic, Cultural and Social Rights has held, the right to adequate housing should
not be equated with the “shelter provided by merely having a roof over one’s head …
50

. Rather it should be seen as the right to live somewhere in security, peace and
dignity.”40 The consequence is that what is adequate has a range of elements ,
including security of tenure, availability of services, habitability, accessibility and
location.41
[142] The second point is that there is a difference between temporary and
permanent accommodation. “[E] mergency accommodation by its very nature will
invariably fall short of the standards reasonably expected of permanent housing
accommodation”.42 The consequence is that “ those who need to occupy such
accommodation must accept less than what would ordinarily be acceptable. ” While
that may seem harsh, it must be “seen against the realities imposed by the vast scale
of the housing backlogs with which the state, in general, and the City, in particular, are
having to engage.”
43
[143] The third is that s uitable alternative accommodation need not be
accommodation that occupiers accept. Occupiers “ cannot delay their eviction each

40 ICESCR General Comment No. 4: The Right to Adequate Housing (1991). South Africa has ratified
the treaty the ICESCR was interpreting – the Covenant on Economic, Social and Cultural Rights – and
our courts have repeatedly referred to its General Comments to interpret socio- economic rights in our
Constitution (although our courts have not always agreed that the obligations under the CESCR should
be adopted in South African law). See, for example, Mazibuko and Others v City of Johannesburg and
Others [2009] ZACC 28; 2010 (3) BCLR 239 (CC); 2010 (4) SA 1 (CC) at para 52.
41 General Comment 4 (n 40) at para 8.
42 Commando (n 3) at para 62, quoting City of Johannesburg v Dladla and Others [2016] ZASCA 66;
2016 (6) SA 377 (SCA) para 20.
43 Ibid.
51

time by stating that they find the alternative accommodation offered by the City
unsuitable.”44
[144] The fourth is this: what is suitable is not absolute, but variable. A question which
has troubled me is whether, if this Court concludes that the safe spaces are adequate
alternative accommodation, will that be true for all evictions, or only evictions of
similarly situated homeless persons? Put differently, if someone was evicted from
established accommodation, would it be reasonable to offer that person a place in a
safe space? And would that be the necessary consequence of my judgment if I granted
the eviction?
[145] I do not read the case law to impose any universal standard for what constitutes
adequate alternative accommodation. What a court finds is reasonable for one
category of persons is not necessar ily reasonable for all other categories. To give
some obvious examples – accommodation that is reasonable for a physically abled
person may not be reasonable for a person with disabilities; accommodation that is far
from any schools may be reasonable for persons with no children, but unreasonable
for those with children.
[146] That does not mean that courts must re- evaluate from scratch the
reasonableness of alternative accommodation offers in each and every case. That
would be unworkable. Accommodation that has been held to be reasonable in one
case will be reasonable in other similar cases . And accommodation that meets the

44 Baron and others v Claytile (Pty) Limited and Another [2017] ZACC 24; 2017 (10) BCLR 1225 (CC);
2017 (5) SA 329 (CC) at para 50.
52

statutory standards in the Housing Code will ordinarily be reasonable absent a
challenge to that Code or the policy giving effect to it.
[147] What is reasonable will depend on occupiers’ needs, and municipalities’ means.
A key question may be whether it is better than their current accommodation. The
City’s case is that, for a person is living on the street, giving them access to a safe
space improves their conditions of living. That will not be the case for a person who
currently occupies a formal home, and who does not face the particular challenges
that homeless pe ople face. But the question for me is not whether safe spaces are
adequate alternative accommodation for all evictions, or even for all homeless people
in all circumstances. The question is whether it is adequate alternative accommodation
for homeless people like the Occupiers living in the City centre of Cape Town.
[148] The unusual feature of this case is that, ordinarily , the requirement to provide
alternative accommodation on eviction exists because of the risk the occupiers will
become homeless. What then is the role of alternative accommodation when the
people evicted are already homeless? To my mind, it must be to improve their situation
and seek – as best as the City can – to get them on the path to being housed. It need
not be to imme diately provide them with permanent or even semi -permanent
accommodation which is unlikely to assist them to off the streets.
[149] I say all this because I wish to emphasise the limits of this judgment. I ultimately
conclude that the safe spaces are suitabl e alternative accommodation for these
Occupiers. It follows that it will likely also be adequate alternative accommodation for
others that find themselves in similarly desperate circumstances. It does not follow
that safe spaces are suitable alternative accommodation for all occupiers in all
circumstances.
53


The Physical Conditions at the Safe Spaces
[150] In their heads of argument, the Occupiers did not attack the quality of the
accommodation offered at the safe spaces. They did not, for example, argue that it
offered insufficient protection from the elements. The attack was limited to the rules
imposed at the safe spaces, not the amenities.
[151] In oral argument, Mr Brickhill sought to argue that the Safe Spaces were not
adequate because they did not provide adequate physical shelter. The source for this
in the pleadings was a throwaway remark in the answering affidavit where Mr M quqa
said that, based on photographs attached to his affidavit, “the safe spaces is (sic)
nothing more than a roof for individuals to sleep overnight”. But the details of this
complaint are never expanded on. And t he Occupiers do not identify the physical
conditions at the Safe Spaces as the reason they did not want to return there, or did
not want to take up the City’s current offer. It was not stated as a defence to the
eviction.
[152] Instead, the gravamen of the complaint was always about the rules, not the
conditions. Mr Mquqa stated that “the safe spaces are worse than the places that we
currently occupy because of the rules imposed that infringe our constitutional rights”.
[153] In the engagement that followed my order of 20 October 2023, the Occupiers
and their attorneys visited Safe Space 1 . In the affidavit they filed following that
engagement, they complain about the conditions there. In particular, they object that
the occupants are required to sleep under “carport-like structures” that are open to the
54

elements on the side. They contend that in poor weather, this offers less protection
than their current structures.
[154] The City does not dispute the accuracy of this version. Instead, it maintains that
it would be inappropriate to consider an issue raised only after the hearing. They
accuse the Occupiers of seeking to make out an “entirely new case”.
[155] When they filed their answering affidavit, the Occupiers were legally
represented by attorneys with experience in eviction litigation. Many of the represented
Occupiers had previously lived in the safe spaces and were familiar with the
conditions. Yet they did not raise it plainly as a concern for why the safe spaces were
not an adequate alternative. It was prop erly put in dispute for the first time at the
hearing. The City was therefore denied a proper opportunity to address the issue, and
the Court was denied a full set of facts and the benefit of legal argument on the
question.
[156] I am not prepared to decide the issue on a basis that was not properly and fully
before me. In any event, I am satisfied that, notwithstanding the rudimentary nature of
the accommodation at Safe Space 1, it is not so deficient that they can not constitute
adequate temporary accommodation.

The Law in Dladla
[157] It is necessary to carefully consider the Constitutional Court’s decision in Dladla
because it is the source for the Occupiers’ attacks on the safe spaces’ rules.
55

[158] Dladla concerned a challenge to the constitutionality of rules that the City of
Johannesburg imposed on temporary alternative accommodation for occupiers
evicted from private property. The City did not provide housing in a temporary
residential area as provided for in the EHP. Instead, it placed them in a shelter, run by
a private service provider. The shelter was in downtown Johannesburg. It was a
“temporary place for destitute individuals looking for employment ”.
45 It consisted of
dormitories. Ordinarily, people would stay there for six months , which could be
extended to 12 months. The shelter also provided people with food.
[159] Like the City’s safe spaces, the shelter in Dladla had certain rules. Two of them
were in issue: the lockout rule and the family separation rule.
[159.1] The lockout rule required the applicants to leave the shelter by 8:00,
permitted them to return at 17:00, and locked them out if they did not
return by 20:00. It could be relaxed if a special arrangement was made.
However the rule was enforced so that persons working at night were
not permitted to sleep in the shelter during the day.
46
[159.2] The family separation rule was not specifically listed, but was enforced
through the provision of separate male and female dormitories. The
effect was that heterosexual couples could not live together. The rule
also separated parents from their children, as children under 16 had to
stay with their mothers, and boys over 16 with their fathers.47

45 Dladla (n 35) at para 7.
46 Ibid at para 13.
47 Ibid at para 12.
56

[160] There are four important findings from Dladla that guide my decision.
[161] First, the Court accepted that the shelters constituted temporary
accommodation “as required by section 26(2)” of the Constitution. 48 It also held that,
but for the shelter rules, “the resultant accommodation provided by the Shelter would
be satisfactory”. 49 Put differently, the shelters were temporary alternative
accommodation that avoided the risk of homelessness. This is important. It supports
my conclusion that temporary alternative accommodation is not limited to housing
offered under the EHP.50
[162] Second, Mhlantla J held that the shelter’s lockout rule limited the applicants’
rights to dignity, privacy and freedom and security of the person:
[162.1] The lockout rule “forces the applicants out onto the streets during the
day with no place whatsoever to call their own and to rest. As a result,
people seek refuge on the street while they wait for the Shelter to re-
open.”
51 For those who work at night, it meant they had no place to
stay – for them “the Shelter is no shelter at all .”52 The rule also

48 Ibid at para 41.
49 Ibid.
50 I accept immediately that the shelters in Dladla were different from the safe spaces. They were
permanent buildings; the safe spaces are not. But for the reasons I have given, it is not appropriate for
me to reject the safe spaces as suitable alternative accommodation based on their physical attributed
given the pleadings and (lack of) evidence before me.
51 Dladla (n 35) at para 48.
52 Ibid.
57

infantilised the applicants, undercutting their right and ability to manage
their own lives and to be “shepherded to and fro.”53
[162.2] Being forced onto the streets during the day denied them the privacy
the shelter offered them, at least at night. The lockout rule denied them
“a place they can call their own to which they can retreat at any time.”
54
Mhlantla J held: “One would think that people who have been evicted
from their homes in which they had some privacy would be provided a
substitute with a measure of the same. They were not.”
55
[162.3] The lockout rule forced the applicants onto the street both during the
day and – if they missed the 20:00 cut off to return – at night. Being on
the street placed them at risk of assault.
[163] Third, the Court also held that the family separation rule infringed the same
three rights – dignity, privacy and freedom and security of the person.56 The focus was
on the right to dignity which, since Dawood, 57 has included the right to family life and
to cohabit with your family. “The family separation rule creates a vast chasm – between
parents and children, between partners and between siblings – where there should be
only intimacy and love.”58

53 Ibid.
54 Ibid at para 50.
55 Ibid at para 50.
56 Ibid at para 48.
57 Dawood and Another v Minister of Home Affairs and Others [2000] ZACC 8; 2000 (3) SA 936 (CC);
2000 (8) BCLR 837 (CC).
58 Dladla (n 35) at para 49.
58

[164] Fourth, it was not possible for the City to justify these limitations of the
occupiers’ rights. The rules were imposed by a contract between the City and MES,
and not by a “law of general application” as required by s 36(1) of the Constitution. 59
The majority therefore did not consider the justifications for the rules. It did not
consider, for example, the reason why the shelter separated males and females. And
it did not decide whether those reasons justified the rules.
[165] Dladla naturally casts serious doubt over the similar rules that are imposed at
the City’s safe spaces. But, nonetheless, for the reasons that follow I find that the
existence of those rules do not prevent the safe spaces from constituting adequate
alternative accommodation.

The Safe Spaces’ Rules are Reasonable
[166] There are two preliminary issues to address before considering each of the
rules with which the Occupiers took issue.
[167] First, u nlike in Dladla , this case is not a challenge to the rules of the safe
spaces. The Occupiers have not brought a counter -application contending that those
rules are unconstitutional and invalid. The permissibility of the rules arises in a different
context – whether the safe spaces are adequate alternative accommodation.
[168] The Occupiers were not obliged to separately challenge the rules. But the fact
that they have not done so has an important doctrinal consequence that distinguishes
this case from Dladla – I am concerned not only with whether the rules limit rights, but

59 Ibid at paras 52-3.
59

whether the safe spaces’ rules mean that the accommodation is not adequate for
these Occupiers. As I have already explained, that assessment is a case -specific,
proportional enquiry that requires me to consider all relevant factors. I could notionally
conclude that the rules limit rights, but still determine that the safe spaces provide
adequate alternative accommodation.60
[169] It also means that I must assess the rules with the concessions that the City
has made for these Applicants. Other homeless people who voluntarily enter the safe
spaces may not get the benefit of guaranteed couples accommodation or guaranteed
accommodation beyond six months. But then, they enter the safe spaces voluntarily,
not under threat of eviction.
[170] Second, the City has – in the course of this litigation – significantly modified or
relaxed the rules it initially imposed. It has made express compromises to address
some of the Occupiers’ legitimate complaints. I detail the nature of those modifications
when I address each rule below.
[171] The Occupiers have, to some extent, been faced with a shifting target. But the
City should be praised for making concessions in response to the concerns raised by
the residents, even if it did so only in the course of litigation . In Mazibuko, the
Constitutional Court explained that, if “one of the key goals of the entrenchment of
social and economic rights is to ensure that government is responsive and

60 Even if there had been a challenge to the rules, I am not sure that Dladla would preclude a s 36(1)
enquiry. In Dladla, the rules were imposed through a contract between the City of Johannesburg and a
service provider. Here the City itself runs the safe spaces and it seems has enacted the rules to regulate
them. It is unclear to me whether , applying Dladla, these rules are a law of general application for the
purposes of s 36(1). I prefer not to decide the issue as it was not argued before me and, for the reasons
already given, is not determinative.
60

accountable to citizens through both the ballot box and litigation, then that goal will be
served when a government respondent takes steps in response to litigation to ensure
that the measures it adopts are reasonable, within the meaning of the Constitution.”61
[172] In the specific context of eviction litigation, t he purpose of meaningful
engagement in the context of an eviction is to listen. One of the key indications that
the City was listening is that it altered its rules. The fact that it did so during the litigation
rather than before, should not count against it in assessing the adequacy of the safe
spaces.
[173] That brings us to the specific rules.

Security of Tenure
[174] Initially, the safe space rules provided that the “ period of initial stay ” was six
months. The City explained in its replying affidavit and heads of argument that this
was applied flexibly. The Occupiers claimed that this was insufficient to guarantee that
they would not again be rendered homeless after six months.
[175] Following the request by this Court, the City stated that it would accept an order
in these terms: “The Respondents who take up the Safe Spaces shall be entitled to an
initial six -month stay, which shall be extended until they have acquired alternative
accommodation”. This offer is subject only to one proviso: that the occupier engages
“with the City and co-operating in meeting their Personal Development Plan.”

61 Mazibuko and Others v City of Johannesburg and Others [2009] ZACC 28; 2010 (3) BCLR 239 (CC);
2010 (4) SA 1 (CC) at para 96.
61

[176] In response, the Occupiers did not object to the proviso. Instead, they
contended that this would have the unintentional effect of making safe spaces – which
are intended to be temporary – into permanent or semi -permanent accommodation.
While the safe spaces may be adequate as a temporary measure, the Occupiers
argued, they could not be adequate for long term stays.
[177] As originally stated, the six -month rule was unsatisfactory. I agree with the
Occupiers that it would not be adequate to rely simply on “flexibility” in applying that
rule without some guidelines to indicate when people would be required to leave and
when they would be permitted to stay.
62
[178] But the City has now committed that – at least for these Occupiers – they can
stay for as long as they need; provided only that they cooperate with the City. In my
view the offer is reasonable. I accept the Occupiers’ concern that the s afe spaces
should not become permanent accommodation. But the City has no incentive to do
that; i ts goal is to get people out of safe spaces and into more permanent
accommodation. That not only achieves the goal of reintegrating those in the s afe
spaces, it frees up spots for other people.
[179] We also cannot lose sight of where the Occupiers are currently living; on the
pavement. Unlike in Dladla , the Safe Spaces are more dignified than their current
situation, not less.


62 This applies only to people who enter the safe spaces following an eviction, not those who freely
chose the safe spaces without any threat of eviction.
62

The Family Separation Rule
[180] The Cit y defended the Safe Space rules to keep single men and women
separately. This was necessary primarily to ensure that women are protected from
gender-based violence. The Occupiers did not take issue with this basic goal for single
people. But they argued that the safe spaces should make provision for couples and
families.
[181] Initially, the City made a limited offer for couples. It noted that couples’
accommodation was already available at Safe Space 1 , and that it could be made
available at Paint City. Eventually, through the process of engagement, it offered to
ensure that couples accommodation would be available for all couples . To my mind,
that answers the Occupiers’ complaint. The basis on which alternative accommodation
will be provided under my order will be on the condition that every couple is
accommodated as a couple.
[182] Th e Occupiers complain that the couples’ accommodation is inadequate. They
are permitted to erect “flimsy” separation that offers insufficient privacy. I accept that
the privacy may be limited. Ideally, couples would have separate rooms which they
could lock. If the City upgrades the safe spaces it may consider increasing the privacy
couples can enjoy. But in my view, the City has done enough to meet the minimum
requirements to accommodate couples.
[183] The City took the view – and the Occupiers did not disagree – that the safe
spaces were not suitable for children. It undertook that those children (and their
families) that were found to be present at the sites would be dealt with differently. I
return to that issue when I address the just and equitable remedy.
63


Lockout
[184] Initially, the lockout rule was strictly stated. But the City relaxed it so that it now
reads: “Residents will be encouraged to vacate the site between 8:30 and 17h00 every
day unless the personal circumstances of any resident makes this unreasonable on a
day or for a period of time.” The Occupiers expressly accepted in their counsels’ heads
of argument that this amended rule “is now consistent with the Constitution.”
[185] In my view, that concession was correctly made. As Dladla holds, a
strictly-enforced lockout rule is unconstitutional. But a rule which encourages people
to leave the safe space, rather than kicking them out, does not.

Conclusion
[186] As they originally stood, the rules of the safe spaces may well have been too
restrictive. But the rules have been relaxed to accommodate all the Occupiers’
legitimate concerns.
[187] The safe spaces are adequate alternative accommodation for these Occupiers.
They are better than the occupiers’ current conditions, and offer enough to meet the
standard of adequacy. Most importantly, Safe Space 1 is in the city centre where the
City acknowledges the Occupiers must stay to earn a living. Not only do they offer
accommodation, they offer meals, toiletries, safety, access to healthcare, drug
rehabilitation, work opportunities and, most importantly, a promise a plausible path out
of homelessness and to permanent accommodation.
64

[188] The Occupiers aver that the safe spaces create “a perpetual cycle of
homelessness”. That claim is not borne out by the evidence before me. The evidence
before me is that they provide the basic shelter people need, and at least some of the
resources to escape homelessness. The evidence shows that they have had at least
some success. Indeed, the report put up by the Occupiers themselves calls for the
expansion of the safe space model. Roughly a third of the Occupiers have indicated
they would be willing to move to the safe spaces if evicted.
[189] That does not mean the safe spaces are perfect. No doubt they can be
improved. The City should always be open to improvement. And it does not mean they
are the best or only solution to homelessness in the City. But they meet the standard
of reasonableness.
[190] Finally, the Occupiers raised concerns about whether the rules would in fact be
implemented as the C ity promised. It asked this C ourt to maintain supervision of the
eviction to ensure the C ity complies with its promises. For reasons I set out below, I
do intend to maintain supervision of this matter. That will accommodate this concern.

Is Eviction Just and Equitable?
[191] Whether eviction is just and equitable must “be decided not on generalities but
in the light of its own particular circumstances. Every situation has its own history, its
own dynamics, its own intractable elements that have to be lived with (at least, for the
time being) ”.
63 This fact specific enquiry also means that, “ when balancing the

63 PE Municipality (n 9) at para 31.
65

interests, compromises have to be made by both parties, in order to reach a just and
equitable outcome.”64
[192] Having carefully considered all the relevant factors, I conclude that eviction is
just and equitable. There are ultimately six reasons.
[193] First, the Occupiers’ occupation is not only unlawful, it is unsustainable. In other
cases, illegal occupation has grown, or people have become so entrenched that
eviction is no longer feasible and the only option is to live with the occupation.
65 That
is not the case here. The City has persuasively explained why the Occupiers cannot
remain indefinitely on the Properties. The Occupiers obstruct the pavements, create a
danger for the City’s infrastructure, and live in objectively unacceptable conditions.
The Occupiers do not contend that they may or should remain on the Properties
indefinitely.
[194] The value of pavements and the government’s legitimate interest in ensuring
people do not live on them was recognized by the Indian Supreme Court in Olga
Tellis.
66 Chandrachud CJ explained that “footpaths or pavements are public properties
which are intended to serve the convenience of the general public. They are not laid
for private use and indeed, their use for a private purpose frustrates the very object for
which they are carved out from portions of public streets.” It rejected an argument that
homeless people – that in India were called “pavement dwellers” – were entitled to live

64 Grobler v Phillips and Others [2022] ZACC 32; 2023 (1) SA 321 (CC); 2024 (1) BCLR 115 (CC) at
para 40.
65 See, for example, President of the Republic of South Africa and Another v Modderklip Boerdery (Pty)
Ltd [2005] ZACC 5; 2005 (5) SA 3 (CC); 2005 (8) BCLR 786 (CC).
66 Olga Tellis (n 7).
66

on pavements: “No one has the right to make use of a public property for a private
purpose without the requisite authorisation and, therefore, it is erroneous to contend
that the pavement dwellers have the right to encroach upon pavement [s] by
constructing dwellings thereon.”67
[195] The City is entitled to ensure that pavements and roads can be used for their
intended purpose. It is entitled to protect its infrastructure. And it is not only entitled,
but obliged, to provide suitable alternative accommodation for homeless people who
live on its pavements and roads. As the Occupiers refuse to vacate the Properties , it
can only do that if it evicts the Occupiers.
[196] Second, the City has engaged meaningfully with the Occupiers . The
engagement was not perfect. But the City did enough to inform the Occupiers of their
options, listen to their concerns, and make reasonable compromises.
[197] T
hird , the City has recognized that the Occupiers must remain in the City centre
and has offered alternative accommodation that will allow them to do so. Again, the
Supreme Court of India described the futility of seeking to move homeless people
away from where they can make a living as “ a game of hide and seek. ”
68 The
government “removes the ramshackle shelters on the pavements … , the pavement
dwellers flee to less conspicuous pavements in by -lanes and, when the officials are
gone, they return to their old habitats. Their main attachment to those places is the
nearness thereof to their place of work.”
69 The safe spaces at least try to avoid that

67 Ibid.
68 Ibid.
69 Ibid.
67

inevitable result by ensuring that the Occupiers remain close to employment
opportunities.
[198] Fourth, the City does not seek the Occupiers’ eviction without a plan. It seeks
their eviction to safe spaces that, on the evidence before me, offer not only suitable
alternative accommodation, but a plausible path out of homelessness. I do not suggest
that moving to the safe spaces will ensure that none of the Occupiers return to the
streets. Inevitably some of them will (which is why the City seeks an interdict). But the
facts show that there is a reasonable possibility that the safe spaces will help some of
them to escape homelessness.
[199] F
ifth, the City has shown that there is no other better, reasonably available
option. TRAs are far from the City centre and are not a realistic solution for the
Occupiers. It does not have the funds or the land to build temporary accommodation
other than safe spaces in the City Centre. Its decision not to do so has been upheld
by the Supreme Court of Appeal.
[200] S
ixth , I accept that mos t of the Occupiers would prefer to remain where they
are, rather than move to the safe spaces. They have built homes, some of them over
many years. They have formed communities and become used to where they are and
how they live. T he safe spaces come with advantages, but do somewhat limit their
freedom. But “the wishes or personal preferences of the unlawful occupier are not
relevant. An unlawful occupier … does not have a right to refuse to be evicted on the
basis that she prefers or wishes to remain in the property that she is occupying
unlawfully.”
70 While that was said in a very different context, it seems to me it applies

70 Grobler (n 64) at para 36.
68

equally here. The Occupiers’ desire to continue occupying land unlawfully cannot be
a reason to refuse eviction.
[201] In the end, on these facts, eviction is just and equitable. But to ensure the
eviction does not impact unduly harshly on the Occupiers, the conditions under which
it will be carried out must be carefully regulated.

When and How should the Occupiers be Evicted?
[202] I have concluded that the City is entitled to an eviction order. But there remain
several thorny problems about how exactly to implement this order. Some of those
problems are inherent in this type of eviction – that of a transient population – where
neither the City nor the Occupiers’ attorneys can say with certainty who occupies the
Properties. It has, unfortunately, been exacerbated by my own delay in delivering this
judgment. The result of that regrettable delay is that the information carefully gathered
by both the City and the Occupiers after the hearing is likely no longer accurate.
[203] To my mind, there are six practical issues to address:
[203.1] Ensuring that the City has enough places at Safe Space 1 in
Culemborg for all the Occupiers (including couples) who wish to take
advantage of them;
[203.2] Providing special protection for any children that are occupying the
Properties;
[203.3] Addressing the needs of the elderly and disabled;
69

[203.4] Dealing with the Occupiers’ possessions;
[203.5] Service of the Order on the Occupiers;
[203.6] The timing of the eviction;
[203.7] To whom the eviction applies; and
[203.8] Supervision.

Adequate Places at the Safe Spaces
[204] In the affidavits filed after the hearing, the City informed this Court that it could
accommodate 184 persons at the Safe Spaces. The high estimate of how many
people were occupying the Properties was 140. Between 37 and 64 indicated they
wanted to move to the Safe Spaces. Similarly, the maximum number of couples that
could be accommodated was 17. The City stated it could accommodate all these
couples.
[205] At the hearing, the City had sought a staggered order that would require people
to vacate the Properties in stages to ensure it had sufficient capacity to accommodate
them all. However, my understanding is that – as things stood on 10 December 2023
– it could immediately accommodate all the Occupiers and did not require a staggered
process.
[206] However, things may have changed in the months since then. It may not be
possible for the City to immediately accommodate all those who want to take
70

advantage of the offer of alternative accommodation at the Safe Spaces. There may
be more Occupiers, or fewer spaces.
[207] Given my delay in delivering judgment, I do not believe it is in the interests of
justice to require the parties to conduct yet another count of the number of Occupiers
before I deliver judgment. Rather, the best way to deal with the uncertainty is as
follows:
[207.1] The Order will be structured on the default assumption that the City
can immediately accommodate all the Occupiers who wish to take up
the offer of Safe Spaces.
[207.2] However, it will build in flexibility in case the City is unable to meet the
demand at the time of eviction. In my view, the best way to achieve
that is threefold:
[207.2.1] Require the Occupiers to indicate within one week of the
date of the Order whether they intend to take advantage of
the Safe Spaces. This should give the City enough
information to determine whether it has enough capacity.
However, the Order will not prevent Occupiers who
subsequently wish to take up a space at the Safe Spaces
from doing so.
[207.2.2] Permit the City itself to stagger the eviction to meet its
capacity constraints; and
71

[207.2.3] If this approach proves unworkable, p ermit the parties to
approach this Court to vary or supplement the Order if
necessary.
[208] The same principle applies to couples’ accommodation. I do not think it is
necessary to provide a separate process for that particular capacity issue. The Order
will simply make clear that the City is obliged to provide couples’ accommodation to
those Occupiers that require it.

Protecting Children
[209] The latest evidence available is that in December 2023, there were no children
living on the Properties. There had been children, but they had all been moved from
the Properties. If that is still the case, then there is no need to address the issue.
However, if minor children have since moved to the Properties, then the City accepted
that they cannot simply be evicted and required to fend for themselves. Nor can they
responsibly be accommodated at the Safe Spaces.
[210] Children’s best interests must be paramount in every matter concerning that
child.
71 That includes where there is a risk that children will be evicted, particularly
when everyone accepted that the safe spaces were not appropriate for children.
Special provision for any children is needed.

71 Constitution s 28(2).
72

[211] The City proposed a mechanism to address the possible needs of any children
at the Properties. With some minor changes, I have adopted that proposal. In short, it
will operate as follows:
[211.1] Children and their parents or caregivers will not be evicted together
with the other Occupiers.
[211.2] The City will ensure that the Provincial Department of Social
Development conducts an assessment of any child and produces a
report for the City.
[211.3] The City will engage with the parents or caregivers (and if appropriate
in light of their age, also the child) about the alternative accommodation
options that are available.
[211.4] The City will file an affidavit with the Court setting out the report of the
DSD, the outcomes of its engagement, and its proposals.
[211.5] The affected Occupiers will be entitled to file a responding affidavit.
[211.6] The Court will then make a specific order for how any children should
be dealt with in light of that information.

73

The Elderly and Disabled
[212] PIE requires courts to have specific regard for the elderly and the disabled when
granting eviction orders.72 The Court asked the parties to report on whether there were
any elderly or disabled people amongst the Occupiers.
[213] The post-hearing engagement revealed one person living at the Helen Suzman
site who is disabled. Ashton Jumat has impaired mobility and struggles to walk . The
City proposed that he be accommodated with his mother at Safe Space 1.
Alternatively, he can be accommodated at Safe Space 2, which has disabled
bathrooms. The Occupiers did not directly respond to this offer.
[214] There was only one person who the Occupiers described as elderly – Nceba
Manyela who is 57 or 58. I agree with the City that this does not qualify as elderly as
contemplated in PIE.
[215] In my view, the City’s approach seems reasonable. If once the Order is granted,
it emerges that there are other disabled persons living on the Properties, they can be
dealt with in two ways:
[215.1] The Order will include a general requirement that the City reasonably
accommodate any disabled person who wishes to take up a place at
the Safe Spaces.
[215.2] If there is a dispute, about whether the City has done so, it can be dealt
with under this Court’s general ongoing supervision.

72 PIE s 4(6).
74


The Occupiers’ Possessions
[216] The Occupiers all have possessions – their clothes, their shelters, and all the
other things that make up a life. They may seem meagre to some, but they are part of
a dignified life. The Occupiers have collected them over a period of time, and should
not be required to give them up. There are three categories of issues.
[217] First, there may be Occupiers who do not wish to go to the Safe Spaces, but
have somewhere else that they wish to go. The City has agreed to transport those
Occupiers’ belongings to their new home provided that: (a) it is within 40km of the city
centre; and (b) the person has permission to occupy that property or store their
belongings there. Those conditions seem reasonable to me. The City cannot be
complicit in facilitating a fr esh unlawful occupation of its own or another’s property. I
accept that there may be few if any Occupiers who can demonstrate consent to occupy
alternative land, but if they can, the City should assist them to transport their
possessions.
[218] S
econd, for those who are evicted , the City has agreed to store their
possessions in its facility in Maitland for six months. The Occupier can regather their
possessions later if they require them. This offer is well made. However, I intend to
impose two additional requirements on the City:
[218.1] I f an Occupier subsequently collects their possessions, the City will
transport them to the person’s new place of living, subject to the
same conditions that apply to those who choose to relocate at the
time of eviction.
75

[218.2] An occupier may request a once-off six month extension of the period
for which the City will store their possessions.
[219] Third, what of the possessions of those who do take up the Safe Space offer ?
Some of them can be taken to the Safe Space. The City should transport those limited
possessions for the Occupiers. But the storage capacity there is limited. It may not
accommodate all an Occupier’s possessions. The Occupier may not need the things
she used to construct a shelter while she is in the Safe Space. But she may need them
later. An Occupier should not be required to sacrifice all her possessions as a condition
for entering a Safe Space. I therefore intend to make the same regime applicable to
those who are evicted available for Occupiers who take up the offer of a Safe Space
and indicate they wish to have their possessions stored.

Service
[220] The City has proposed a two- pronged method of service – reading the Order
by loudhailer at each of the sites, and affixing it to a notice board or permanent
structure at each site. This seems appropriate. I have made some minor amendments
to the proposal to make it clear when and how often the Order must be read.
[221] In addition, I require that the Occupiers’ attorneys themselves must use their
best efforts to ensure all the Occupiers are aware of the Order. I have no doubt that
they would have done so in any event. But I include it in the Order because it seems
to me that may be the most effective way to ensure that the Occupiers understand the
content and effect of the Order.

76

Timing of the Eviction
[222] How long should the Occupiers be granted before they are required to vacate
the Properties? The City initially asked for an order that they vacate the Properties
within 30 days failing which they could be forcibly evicted. The draft order provided at
the hearing would have afforded the Occupiers between two weeks and four months.
That staggered approach was designed to meet the City’s capacity at the Safe
Spaces.
[223] Six weeks – thirty court days – seems reasonable to me in these circumstances.
It will provide enough time for the City and the Occupiers to perform the various tasks
that need to be done before the eviction can take place. It will also provide sufficient
time for the Occupiers to consider their position and, if they wish, to make alternative
arrangements.

To Whom Does the Eviction Order Apply?
[224] As I have set out in detail above, there has been ongoing uncertainty about who
lives on the Properties. Both the City and the Occupiers accepted that it was
impossible to obtain a single, comprehensive list of all those living on the Properties.
There are three groups of people:
[224.1] Those whom the City and/or the Occupiers’ attorneys have confirmed
have at some point between the launching of the application and the
post-hearing engagement, lived on the Properties; and
77

[224.2] Unnamed people who lived on the Properties previously, but do not do
so currently; and
[224.3] Unnamed people who live on the Property when this Order is granted
and/or when the eviction is sought to be carried out.
[225] For the purposes of the eviction order, it does not matter. All those who are
occupying the Properties when the eviction order is effected may be evicted. Both
those who have already indicated they wish to move to Safe Space 1, and those who
indicate that intention after this order must be accommodated.

Supervision
[226] This is an unusual eviction. T he order is, necess arily, lengthy and somewhat
complicated. I am also concerned that the position on the ground may have c hanged
substantially since the last affidavits were filed. As mentioned earlier, the Occupiers
are concerned about how the rules at Safe Space 1 will be implemented.
[227] I have therefore decided to maintain light supervision of the eviction. I do not
think it is necessary to have ongoing reports (save as concerns any children). But I do
think it is appropriate to require the City to file an affidavit at the conclusion of the
eviction, and to permit the Occupiers to respond. The purpose is twofold – to ensure
that the City has complied with the order, and to deal with any outstanding issues that
may arise. If the parties want further supervision at that stage, they can explain why
they think it is needed. I will then decide whether to continue supervision and, if so, on
what terms.
78

[228] In addition, in order to address any eventualities that occur during the process,
I will permit any party to approach this court for a variation of the order on notice to the
other parties.
[229] I emphasise that the purpose of this supervision is not punishment , not is it
motivated by suspicion. It is about trying to ensure that the different branches of state
work together to achieve constitutional outcomes in difficult circumstances.73

INTERDICT
[230] In addition to the eviction, the City seeks an interdict. Initially, the interdict it
sought was against all the Occupiers – named and unnamed – and sough to prohibit
them “from occupying the properties or engaging in prohibited conduct on the
properties and any other properties owned and/or controlled by the” City . In the draft
order handed up at the hearing of the matter, the relief was altered slightly. It applied
only to the named Occupiers, and it sought to interdict them “from occupying the
properties, and taking any action that denies the Applicant or other members of the
public from accessing or using the properties and any other properties owned and/or
controlled by the Applicant.”

73 Mwelase and Others v Director-General for the Department of Rural Development and Land Reform
and Another [2019] ZACC 30; 2019 (11) BCLR 1358 (CC); 2019 (6) SA 597 (CC) at para 46.
79

[231] Limiting the interdict to named applicants was proper. Courts should not
generally grant interdicts against unnamed persons. 74 In this case, it would be
impossible to know who had been interdicted and who had not.
[232] But the question still is whether the interdict is justified. There are two parts to
what was sought. First, an interdict against the named Occupiers from re -occupying
the specific properties from which they were evicted, or other Occupiers were evicted.
I have no difficulty with this part of the interdict. In my view, it is implicit in any eviction
order that the person cannot re-occupy the same property with impunity. I am not sure
the interdict is even necessary; the original eviction order could be used to “re-evict”
someone who re-occupied the property.
[233] The difference here is that the City seeks to interdict re- occupation not only of
the properties from which a particular occupier w ill be evicted, but from all the other
properties from which other occupiers will be evicted. I am satisfied that is just and
equitable. I have determined that an eviction order is appropriate for all the Occupiers
on all the Properties. I am willing to interdict them all from re- occupying any of the
Properties, not only the particular one that they occupied.
[234] That leaves the question of which Occupiers should be covered by the interdict.
In my view, it should apply to all those Occupiers whose names have been provided
to the Court during this litigation.
[235] The second part of the interdict the City seeks is more problematic. It seeks to
interdict the named Occupiers from stopping the City or the public from using any other
property owned by the City. That relief is wide-ranging. It would impose a specific legal

74 City of Cape Town v Yawa and others [2004] 2 All SA 281 (C).
80

regime on the Occupiers that would not apply to other homeless people in the City. It
would prevent them from living on any other City -owned land in a manner that
interfered with its ordinary use.
[236] The City explained that without the interdict, its eviction order would be futile.
The Occupiers could simply move across the street to another pavement owned by
the City, but not listed in this case. The City would then have to apply for a new eviction
order. It would, in its view, be back to square one.
[237] The Occupiers saw the interdict differently. In their view, it would authorize the
City to evict the Occupiers from different properties in the future without following the
requirements of PIE. Th at, it argues, is unconstitutional. Moreover, the City has
alternative remedies available to it – an eviction application or its own By- law that
permits it to remove people from public places in certain circumstances.
[238] I am not inclined to grant the broader part of the interdict. In my view, while it
may have some short -term utility for the City, it is unlawful , unnecessary and
ineffective.

The Interdict is Unlawful
[239] I agree with the Occupiers that the interdict will – and is intended to – permit
their eviction in the future from City land without an eviction order that is compliant with
either PIE or s 26(3) of the Constitution. That is the whole point of the interdict – to
81

avoid the cost of eviction applications in the future. In Zulu,75 the Constitutional Court
held that a similar interdict that was used to remove people from informal settlements
constituted an eviction order. I do not see how the interdict the City seeks is different.
[240] But s 26(3) is clear: “ No one may be evicted from their home, or have their
home demolished, without an order of court made after considering all the relevant
circumstances.” PIE is the legislation enacted to give effect to s 26(3) and sets out the
process for an eviction order and the factors that must be considered. The
enforcement of the interdict would result in an eviction without following that process.
[241] I accept that the interdict the City seeks would not only apply to situations where
the Occupiers had made a new home on City land – it would also apply to other
conduct that obstructed access or use and enforcing it then would not violate s 26(3).
But its purpose is primarily directed at preventing them from re-occupying City-owned
land, and allowing the City to remove them through a process other than an eviction.
That is the purpose of the interdict. In my view, this Court cannot do that.
[242] If the interdict were granted, t he Occupiers would be evicted without an order
of court that considered all relevant circumstances , an d without an eviction order
granted in terms of PIE . They could be evicted without meaningful engagement, and
without a consideration of whether alternative accommodation is available. That, as I
understand it, is the entire point of the interdict.
[243] The majority in Zulu declined to make a finding on whether the interdict was
unlawful or not for procedural reasons. But in a concurring judgment, Van der

75 Zulu and Others v eThekwini Municipality and Others [2014] ZACC 17; 2014 (4) SA 590 (CC); 2014
(8) BCLR 971 (CC).
82

Westhuizen J (joined by Froneman J) held that it was unconstitutional and unlawful to
grant interdicts that effectively permitted evictions . He held: “ An order of this nature
deprives unlawful occupiers of rights enshrined in the Constitution and recalls a time
when the destitute and landless were considered unworthy of a hearing before they
were unceremoniously removed from the land where they had tried to make their
homes.”
76
[244] Wilson AJ (as he then was) reached a similar conclusion in Johannesburg City
v K2016498847 (Pty) Ltd.77 He held that a municipality was not entitled to an interdict
to enforce a land use scheme that would have the effect of evicting people from their
home unless it met the requirements for an eviction under s 26(3) of the Constitution.78
That would include meaningful engagement and the offer of alternative
accommodation if there was a risk of homelessness. The same principle, to my mind,
applies here.
[245] In short, while evictions may be expensive and time-consuming that is because
they have serious consequences. People should not lose their homes unless the
proper process has been followed and a court has considered all the relevant
circumstances. I do not believe the City has a clear right to bypass those requirements.


76 Ibid at para 44.
77 2022 (3) SA 497 (GJ) . See also Ekurhuleni Metropolitan Municipality v Harmse and Others [2023]
ZAGPJHC 860 and Ekurhuleni Metropolitan Municipality v Sibanda [2022] ZAGPJHC 286.
78 Wilson AJ concluded that PIE may not apply because the people facing eviction in that case were
not “unlawful occupiers” as defined in PIE. The implication is that if they were unlawful occupiers, the
Municipality would be required to comply not only with s 26(3), but also with PIE.
83

The Eviction is Unnecessary
[246] T he City claims that without the interdict the costs of addressing new
occupations by the Occupiers will be too high. It will be hamstrung from clearing its
pavements and other public places. As I have explained, that is the price s 26(3)
requires the state to pay. But there is another reason to refuse relief. The City has an
alternative remedy – its own By-Law.
[247] The City’s By -Law permits the City to, in effect, evict people who have made
their homes in public places without a court order, at least in certain circumstances .
Section 2 of the By-Law prohibits a range of conduct in public. In particular, s 2(3)(m)
prohibits sleeping overnight or erecting any shelter in a public place. Section 22A –
which was introduced in 2022 – permits the City’s officials to direct people to comply
with the By-Law and, if they refuse, to arrest the person. That power is subject to an
important proviso. Section 22A(2) provides that the power to arrest “may only be
exercised in respect of a contravention of section 2(3)(m) if the person refuses to
accept an offer of alternative shelter.”
79 The same proviso appears to apply to the
power to impound a person’s possessions in s 22A(1)(e).
[248] In short, the By-Law permits the City to, in effect, evict (and even arrest) people
who are living in public places, provided only that the person refuses an offer of
alternative shelter. If there is no alternative shelter available the City cannot, under its
own By-Law, remove/evict the person.

79 Section 22A(2) must be read with s 22(1)(d) which expressly makes the power to arrest “subject to
subsection (2)”.
84

[249] When I heard this matter, there was a pending challenge to the constitutionality
of the By -Law; that was part of the reason the Occupiers did not proceed with their
counter-application.80 T o the best of my knowledge, it is still pending. Until it is
decided, the By-law remains valid. I express no view on whether that challenge or any
other constitutional challenge to the By-Law is good or bad. Either way, the City should
not get its interdict.
[250] If the challenge is bad, then the City has an alternative remedy and need not
enlist the Court to provide a remedy through an interdict that it already has through its
own laws. If the challenge is good, then the City would be seeking to do through court
orders what it cannot constitutionally do through its by-laws.
[251] T he Constitution binds all branches of government – the legislature, the
executive and the judiciary .
81 If the removal of homeless people from the streets
without an eviction order is unconstitutional when done through a By -Law, it is also
unconstitutional when done through a court order.82 And if it can be done through the
By-law, the City does not need its interdict.
[252] The City argues that this case is different because it involves an interdict
against people who have already been evicted. I do not see why that makes a
difference. Section 26(3) applies to all people, whether they have previously been
evicted or not. It is not a once- off guarantee. In any event, the City could amend its

80 I was initially allocated to hear that application as part of a Full Court shortly after this application was
heard, but it was postponed by agreement.
81 Constitution s 8(1).
82 There may be circumstances where removal without a court order, or without an eviction order, is
justified. For example, in emergency situations, or to prevent imminent safety risks. But then that
limitation of s 26(3) will be justifiable whether enacted in By-Law or court order.
85

By-Law so that it would only permit removal without court order of people who have
already been evicted from City -owned land. If it believes that justifies a limitation of
s 26(3), it should enact it into law and defend it.
[253] The City is not an ordinary litigant that must live within the existing legislative
universe. It is a law -maker. It can enact new by -laws and amend existing by -laws. It
should do so, rather than seeking to legislate through court orders against specific
groups of people.
[254] I accept that the scope for by-laws that permit evictions outside of PIE may be
narrow, not only because of s 26(3), but also because of s 156(3) which provides that
by-laws contrary to national legislation are ordinarily invalid.
83 But that is the
constitutional scheme. I do not see why the City should be able to escape it through
interdicts.

The Interdict will be Ineffective
[255] Th is interdict on its own will not solve the City’s “problem”. It would only permit
to evict without court order those named Occupiers who re-occupy City land. But it
would still need to obtain an eviction order for all the other homeless people who will
likely be occupying the same land. It seems inherently unlikely that there will be a
future occupation solely of the named Occupiers. But if it must obtain an eviction order
for some people occupying the same portion of land, why not all of them?

83 The relevant part of s 156(3) reads: “Subject to section 151(4), a by-law that conflicts with national or
provincial legislation is invalid.” Section 151(4) reads: “The national or a provincial government may not
compromise or impede a municipality’s ability or right to exercise its powers or perform its functions. ”
86

[256] This interdict will not avoid the need for an eviction order. It will – at most –
mean some people on a particular property must be evicted under PIE and some can
be evicted outside of PIE. That will not significantly reduce the City’s costs.
[257] All t he interdict would achieve is to create a special category of homeless
people subject to a different regime than all other homeless people in the City. The
only reason would be that they had previously been evicted from different City-owned
land. I do not see the value in that outcome.
[258] I was informed from the bar that, depending on the outcome of this application,
the City would seek to bring further similar eviction- plus-interdict applications. The
plan, it seems, is to capture as many homeless people as possible within a net of
interdicts. To my mind, that just shows this is an attempt to legislate through courts,
rather than through the legislative process. In any event, I cannot grant an interdict
that will be ineffective now because it may become effective only once numerous
future interdicts are granted. Put simply, the interdict will not prevent the injury.

Conclusion on the Interdict
[259] For these reasons, I take the view that I cannot grant the broad interdict the City
seeks. The interdict would be unlawful and unconstitutional. If the City is permitted to
limit s 26(3) and circumvent PIE, then it must do so through its by -laws. And the
interdict would only aid the City if it obtained a similar interdict for all or most homeless
people in the City.
[260] In refusing the interdict I do not mean to undermine the City’s goals. I have
accepted that the City has a right and an obligation to ensure that all public places are
87

accessible for all its residents. It has a right and an obligation to assist homeless
people to leave the streets and reintegrate and rehabilitate. It has a right and an
obligation to protect its infrastructure from damage and degradation. But the
Constitution imposes limits on how it can achieve those goals when people are living
on the streets. The City must work within those limits because they protect the
fundamental rights of all its residents, including those experiencing homelessness.
[261] That may require more carrot than stick – incentivizing homeless people to
leave the streets rather than threatening them. The expansion of the safe space model
seeks to achieve exactly that. Or i t may require new, imaginative solutions and
collaborations. Or it may be that the law as it stands makes it impossible for the City
to effectively achieve its goals. But the law is the law and this Court cannot ignore it.

CONCLUSION AND COSTS
[262] The City has achieved substantial success in this case. It has obtained its
eviction, and an interdict – albeit on much narrower terms than it sought. However, the
Occupiers have also had some success. Through the litigation they have extracted
substantial concessions from the City. If the City had not engaged further with the
Occupiers after the application was launched, and adjusted the rules for the safe
spaces, I would likely have refused the eviction. On the other hand, the Occupiers
continued to oppose the eviction even after the further engagement and the
adjustments to the rules.
88

[263] This litigation is covered by the principles for costs in non-frivolous
constitutional litigation between private parties and the state .84 The City has no right
to recover costs from the Occupiers even if it successful. The Occupiers, on the other
hand would be entitled to their costs against the City if they had been successful.
[264] In this case, where the Occupiers have been partially successful, the
appropriate order is to award them part of their costs. In my view 30 % fairly reflects
the extent of their success. As this case was resolved prior to 12 April 2024, Rule 67A
has no application.
[265] For all the above reasons, I make the following order:

1. In terms of section 4, read with section 6 of the Prevention of Illegal Eviction
and Unlawful Occupation of Land Act 19 of 1998, the Respondents are evicted
from properties described in paragraph 1 of Part A of the Notice of Motion, and
graphically depicted in Annexure B to the Notice of Motion (the Properties).
2. The Respondents shall vacate the Properties by 30 July 2024.
3. The Applicant (the City) shall provide alternative accommodation in the form of
a place at its Safe Space 1 located in Culemborg (the Safe Space) to any of
the Respondents who has informed the City, or at any point prior to eviction
informs the City, that they wish to be accommodated at a Safe Space. To give
effect to this order:

84 Biowatch Trust v Registrar Genetic Resources and Other s [2009] ZACC 14; 2009 (6) SA 232 (CC);
2009 (10) BCLR 1014 (CC).
89

3.1. The R espondents’ attorneys shall, within ten days of the date of thi s
order, provide a list of R espondents who wish to move to the Safe
Space; and
3.2. The City may, depending on the number of Respondents who indicate
an intention to move to the Safe Space, stagger the eviction to ensure it
has enough space to accommodate all the Respondents.
4. The Respondents who take up the alternative accommodation at the Safe
Space shall be entitled to an initial six-month stay, which shall be extended until
they have acquired alternative accommodation, subject to such individuals
engaging with the City and co- operating in meeting their respective Personal
Development Plans.
5. The City shall ensure that those Respondents who take up the alternative
accommodation at the Safe Spaces , who are in a partnership with another
Respondent, and who require accommodation with their partner , are provided
with accommodation for couples.
6. The City shall provide storage for the belongings of the Respondents who take
up the offer of accommodation at the Safe Space for six months. In addition:
6.1. That period may be extended for a single period of six months at the
Respondent’s request;
6.2. If the Respondent wishes to collect their possessions, the City shall
transfer them on the same conditions set out in paragraphs 9.1 and 9.2,
save that the information in 9.2 need only be provided when the
Respondent wishes to collect their possessions; and
90

6.3. If the Respondent does not collect their possessions, the City shall be
permitted to dispose of those possessions.
7. The City shall ensure that if there are any minor children occupying the
Properties as at the date of this Order:
7.1. Those children are referred to the Provincial Department of Social
Development (DSD) for an assessment of the minor child and their
parents to be undertaken, including options for the alternative
accommodation of the minor children and their parents or caregivers;
7.2. The City shall consider the DSD’s report, engage with the children’s
parents or caregivers and, if appropriate in light of the child’s age, the
child, about alternative accommodation options;
7.3. The City shall, as soon as possible, and within 10 days of the date of this
order file an affidavit with the Court setting out the assessment by DSD,
its engagement with the child and the parents or caregivers, and the offer
of alternative accommodation;
7.4. The Respondents shall, within five days of the date the City files its
affidavit, file an affidavit in response;
7.5. The Court shall determine the date of the eviction of any minor child, and
the alternative accommodation that shall be offered; and
7.6. Those children and their parents or caregivers shall not be evicted from
the Properties until the above process is complete.
91

8. The City shall reasonably accommodate any Respondent who chooses to move
to the Safe Space and is elderly or disabled.
9. If any Respondent elects to voluntarily relocate to a place other than the Safe
Spaces, the City shall assist that Respondent to transport their possessions to
their chosen relocation destination, subject to the following conditions:
9.1. The relocation destination is within a 40km radius of the Cape Town
Central Business District; and
9.2. The Respondents’ attorneys shall, within two weeks of the date of this
order, provide the City’s attorneys with:
9.2.1. The full names of the Respondents who require the City’s
assistance;
9.2.2. The destination to which the possessions must be transported;
and
9.2.3. Evidence that the owner or person in charge of the property has
given their consent for the Respondent to occupy their property
and/or store their possessions at their property.
10. In the event that the Respondents fail and/or refuse to vacate the Properties in
terms of this Order, subject to the special provision for children in paragraph 7,
the City and/or any person appointed by the City, duly assisted by the South
African Police Services to the extent necessary, is authorised to:
92

10.1. Eject any Respondent who has not indicated an intention to accept an
offer of alternative accommodation at the City’s Safe Spaces from the
Properties;
10.2. Demolish and or remove any structure unlawfully occupied by the
Respondents on the Properties; and
10.3. Remove any possessions found at the properties belonging to the
Respondents, save that:
10.3.1. Those possessions shall be kept in safe custody by the
Applicant for a period of six months, which may be renewed for
one further period of six months at the Respondent’s request;
10.3.2. If the Respondent wishes to collect their possessions, the City
shall transfer them on the same conditions set out in paragraphs
9.1 and 9.2, save that the information in 9.2 need only be
provided when the R espondent wishes to collect their
possessions; and
10.3.3. If the Respondent does not collect their possessions, the City
shall be permitted to dispose of those possessions.
11. The following Respondents, once they are evicted or voluntarily vacate the
Properties, are interdicted and restrained from re-occupying the Properties:
11.1. The Respondents listed in Annexure A to the Notice of Motion;
11.2. The Respondents listed in Annexure KB1 to the Supplementary Affidavit
of Khululiwe Bhengu dated 9 October 2023;
93

11.3. The Respondents listed in the Affida vit of Nazlie Du Toit dated 10
November 2023;
11.4. Annexure PK9 to the Affidavit of Portia Dyantyi dated 28 November
2023; and
11.5. The Respondents listed in the affidavit of Megan Pangeni filed on 8
December 2023.
12. This Order shall be served on the Respondents in the following manner:
12.1. By delivery of the Judgment by the Registrar to the Respondents’
attorneys by email;
12.2. By the City’s officials reading aloud at the Properties, the contents of this
Order by loudhailer in English, Afrikaans and isiXhosa before 9 AM and
after 6 PM for three consecutive days, beginning two days after the date
of this Order;
12.3. By the City’s officials, within two days of the date of this order:
12.3.1. Erecting two notice boards at each of the Properties, where the
erection of such notice boards is possible, and affixing thereto
copies of this Order in English, Afrikaans and isiXhosa; or
12.3.2. Where it is not possible to erect a notice board, by affixing three
copies of this O rder to any wall, fence and/or permanent
structure on each of the Properties; and
94

12.4. By the Respondents’ attorneys communicating to the Respondents in
the way they deem most appropriate, the content of this Order.
13. This Court shall retain supervision of the implementation of this O rder on the
following terms:
13.1. The City shall, within ten days of the final eviction, file an affidavit with
this Court specifying which Respondents accepted offers of
accommodation at the Safe Space, and including any other informa tion
the City believes relevant;
13.2. T he Respondents shall be entitled, within ten days thereafter, to file an
affidavit in response;
13.3. The Court shall consider the affidavits and make any further order it
deems just, including terminating its supervision; and
13.4. Any party may, at any stage, and on notice to the other parties, approach
the Court for a variation of this order.
14. T he City shall pay 30% of the Respondents’ costs.



______________
M J BISHOP
Acting Judge of the High Court

95

Counsel for Applicant: Adv K Pillay SC, Adv M Ad hikari, Adv M
Ebrahim
Attorneys for Applicant: Fairbridges Wertheim Becker


Counsel for Respondents: Adv J Brickhill, Adv N Simmons
Attorneys for Respondents: Socio-Economic Rights I nstitute of South
Africa