---

THE HIGH COURT OF SOUTH AFRICA

(WESTERN CAPE DIVISION)

JUDGMENT

	Case No: 11256/2019
In the matter between
THE CAMPS BAY & CLIFTON RATEPAYERS’ ASSOCIATION	FIRST APPLICANT
SPRING LIGHTS 1130 CC	SECOND APPLICANT
MICHAEL WILLIAM SHIEL	THIRD APPLICANT
and
AL KHALIFA FAMILY TRUST	FIRST RESPONDENT
THE CITY OF CAPE TOWN	SECOND RESPONDENT

  

Coram: Rogers J

Heard: 23 November 2020

Delivered: 15 December 2020 (by email to the parties and same-day release to SAFLII)

 

JUDGMENT

 

Rogers J

[1] The applicants seek the review and setting aside of the approval of a set of original building plans and of two sets of rider plans relating to the construction of a dwelling on a Camps Bay property belonging to the first respondent trust (‘KFT’). They also allege that the dwelling as built deviates from the approved plans. The first applicant (‘CRA’) is a ratepayers’ association which plays a watchdog role in safeguarding the property interests of residents in the area. The second applicant (‘Spring Lights’, a corporation belonging to the Stern family) and third applicant (‘Mr Shiel’) are the neighbours on each side of KFT’s property. The applicants also seek various declaratory orders. The second respondent (‘CCT’) is the local authority which approved the plans.

[2] KFT opposes the review of the original plans on the grounds of delay. It actively contests one ground of review which is common to all three approvals. KFT did not, in its opposing papers or in written or oral argument, address the other grounds of review, but in response to a question from the court its lead counsel stated that he did not have instructions to concede the review on any of the grounds.

[3] The CCT abides the court’s decision. Its view is that the approvals of all three sets of plans are reviewable. In an explanatory affidavit it has explained the grounds on which it makes this concession. To the extent that the applicants need an extension of time to pursue the review of the original set of plans, the CCT does not oppose an extension.

[4] The application was launched on 1 July 2019 in two parts, Part A being for urgent interdictory relief pendente lite. By that time construction was far advanced. When KFT filed its preliminary answering affidavit on 15 July 2019, its deponent claimed that the building was 99% complete. The interdictory relief was resolved by an agreed order.

[5] The original set of plans was approved in November 2015 (‘the 2015 plans’). The first and second sets of rider plans were approved in October 2017 (‘the 2017 plans’) and December 2018 (‘the 2018 plans’) respectively. The application was launched on 1 July 2019 for hearing on 16 July 2019. The notice of motion was directed at the approval of rider plans. In terms of an order taken by agreement on 17 July 2019, the review application was set down for 6 November 2019. CCT was to deliver its record by 31 July 2019, and the applicants their supplementary founding papers by 30 August 2019.

[6] The CCT filed a record at the end of July but in the applicants’ view it was seriously deficient. The CCT supplemented the record on 6 September 2019. The matter was not ripe for hearing in November 2019. By a further agreed order granted on 9 January 2020, the review was set down for hearing on 17 June 2020, the applicants to file their supplementary founding papers by 31 January 2020.

[7] The applicants delivered an amended notice of motion and supplementary founding papers on 31 January 2020. The amended notice of motion not only attacked the rider plans (now identified with reference to file numbers and approval dates) but also the original plans. In subsequent negotiations, it appeared that in principle the respondents were willing to concede the review of all three approvals. However, the parties could not reach agreement on the precise terms of an order. Because no opposing papers were filed while negotiations were ongoing, the case could not be heard in June 2020. By way of a further order, the application was postponed to 27 October 2020, later changed to 23 November 2020, which is when I heard it.

[8] KFT filed an answering affidavit in September 2020. It resisted the review of the 2015 and 2017 plans on grounds of delay. In argument, the delay point was, as I have said, confined to the 2015 plans. The CCT filed its explanatory affidavit in October 2020.

[9] Since the CCT has only conceded certain grounds of review, and because KFT has not formally conceded any of the grounds, I shall have to deal with all of them. This is in any event necessary because the grounds on which I uphold the review will or may bind the parties going forward. It is unfortunate that I should have to do so without the benefit of argument from the CCT and KFT. Because there are multiple grounds on which each of the approvals is attacked, and because there are also complaints about building work which is not authorised by the latest approved plans, the judgment is much longer than I would have liked.

[10] It is important to be clear about what the CCT decided in each round of approvals. The effect of Camps Bay Ratepayers’ and Residents’ Association & another v Harrison & another 2011 (4) SA 42 (CC) is that where a feature in earlier plans appears without alteration in later plans, municipal officials will not have reason to revisit that feature. A review attack on that feature should thus be directed at the approval of the original plans, because that is when in substance the impugned decision was taken.  

[11] In Harrison the applicants sought to attack a feature of this kind by impugning rider plans approved in September 2007, despite the fact that the feature had been present in the original plans approved in February 2005. The late attack came in a replying affidavit filed in May 2008. On grounds of delay, the Supreme Court of Appeal (‘SCA’) refused to entertain the merits of the attack.[1] I do not understand the applicants in Harrison to have failed because they did not, as a matter of form, seek to have the original approval set aside. Rather, their attack, in so far as this feature is concerned, was treated by the SCA as being in substance an attack on the February 2005 approval, and the Constitutional Court agreed. The SCA held that, having regard to all the relevant circumstances, there were no grounds to overlook the delay of more than three years, and this exercise of its discretion was not attacked in the further appeal.

[12] The approach in Harrison accords with the practice followed by the CCT. When original plans are submitted for approval, the plans in their entirety are assessed. When rider plans are submitted, the submitting party is required to mark up the alterations and their type in specified colours, and it is only the marked-up features that the officials assess. It may be that the approval of rider plans is an approval of those plans in their entirety, but the approval of unchanged features does not entail any fresh application of the mind, so that in substance the relevant approval is the earlier one.

[13] One of the CCT’s complaints is that not all alterations were marked up when the 2017 and 2018 rider plans were submitted, and that alterations which the CCT would not have approved if they had been marked up slipped through by default. In such a case, one might say that the unmarked alterations were not the subject of any decision at all. Alternatively, if the approval of rider plans is in law an approval of the plans in their entirety, one would be dealing with an approval in part to which, due to the fault of the submitting party, the municipality was precluded from applying its mind. In the former case, the court could make the absence of approval the subject of a declaratory order; in the latter case, the approval of the unmarked alterations could be set aside on review. Since there is no practical difference, I shall treat unmarked alterations as having been unwittingly approved and thus the proper subject of review.

[14] In view of Harrison, it is necessary to consider the approvals in sequence, in order to see to what extent complaints about the rider plans were present in the original plans. Since prospects of success are relevant in deciding whether to condone delay, I shall at the same time state my views on the grounds of review. I number the grounds of review for ease of reference. The CCT concedes grounds 1 and 2 in relation to the 2015 plans, ground 2 in relation to the 2017 plans and ground 2 in relation to the 2018 plans.

The 2015 plans

[15] Buildings may not be erected before the building plans are approved by the relevant local authority. Section 4 of the National Building Regulations and Building Standards Act 103 of 1977 (‘the Building Act’) makes provision for applications for such approval. Section 7 governs the grant and refusal of approval. In terms of s 7(1)(a), the decision-maker must be satisfied that the application complies with the requirements of the Building Act and with ‘any other applicable law’. Included in the latter expression are the local authority’s legislated development rules and any development restrictions contained in the property’s title deed.

[16] In regard to the development rules applicable to the 2015 plans, the CCT’s Municipal Planning by Law 2015 came into force on 1 July 2015. The By-Law incorporates Schedule 3, the Development Management Scheme (‘DMS’). The 2015 plans were approved in November 2015. The review was presented and argued on the footing that the By-Law and DMS applied to all three sets of plans. Neither the CCT nor KFT contended otherwise, and the common assumption seems to me to be correct.[2]

Ground 1: Mr Joseph not a ‘competent person’

[17] The CCT concedes this ground, in regard to which the applicants contend as follows. In terms of the regulations promulgated under the Building Act, plans submitted for approval in terms of s 7 must be signed by an ‘appointed competent person’. Mr Joseph signed the plans, declaring that he was an ‘appointed competent person’.

[18] Section 26 of the Architectural Profession Act 44 of 2008 (‘the Architects Act’) provides for the regulation of the types of work which may be performed by different categories of persons registered under that Act. Section 18(1) specifies the four classes of architectural professionals who may be registered. And ‘architectural technologist’ is one such class. In terms of Board Notice 154 of 2011, promulgated by the South African Council for the Architectural Profession (‘SACAP’) under the Architects Act, an architectural technologist was not authorised to sign building plans for residential buildings comprising more than two storeys. Board Notice 154 was in force until its withdrawal on 30 November 2015.

[19] Mr Joseph has at all material times been an architectural technologist. The dwelling which is the subject of the impugned plans consists of a basement and three storeys. Accordingly, so the applicants allege, the 2015 plans did not qualify for approval. The CCT agrees, stating that if Mr Joseph had not misrepresented his authorisation to sign the 2015 plans, the CCT would not have approved them.

[20] To anticipate, the applicants advance the same point in respect of the 2017 and 2018 plans (the CCT does not concede it). On 3 February 2016 the SACAP published Board Notice 10 of 2016. This notice stated that a registered person undertaking work for a client had to do so in compliance with rule 2.1 of the Code of Professional Conduct. Rule 2.1 of the Code required a registered person only to undertake work identified for the category of his or her registration. This regime was in force when the 2017 and 2018 plans were approved.

[21] KFT contends that this ground should be rejected in relation to all three sets of plans. Its argument on the 2015 plans is that the 2011 notice was withdrawn because it had not been approved by the Council for the Built Environment (‘CBE’) or the Competition Commission (‘CC’). The absence of these approvals meant that the 2011 notice did not meet the statutory requirements of s 26 of the Architects Act and was void. In relation to the 2017 and 2018 plans, KFT argues that the 2016 notice did not cure the problem, because rule 2.1 of the Code did not define the work which different categories of professionals may perform. This was first done on 4 October 2019 when the CBE, pursuant to s 20(2) of the Council for the Built Environment Act 43 of 2000 (‘the CBE Act’), issued GN 1274. In terms of the 2019 notice, architectural technologists are confined to drawing plans for ‘simple double storey buildings’.

[22] In relation to the 2015 plans, KFT’s opposition misconstrues the nature of the 2011 notice. The SACAP was not purporting to act in terms of s 26 of the Architects Act or to perform a function which only the CBE could perform in terms of s 20(2) of the CBE Act. The introductory paragraph and preamble to the notice make clear that although the interim regime set out in the notice accorded with a recommendation which the SACAP had made to the CBE (but on which the latter had not yet acted), the SACAP was in the meanwhile promulgating the interim regime as a rule in terms of s 36(3) of the Architects Act.

[23] Section 36(1) empowers the SACAP to promulgate rules in the Gazette with regard to

‘any matter that is required or permitted to be prescribed in terms of this Act and any other matter for the better execution of this Act or in relation to any power granted or duty imposed by this Act.’

In terms of s 36(2), the SACAP must first publish the draft rule for comment. Section 36(3) provides, however, that the SACAP may, if circumstances necessitate the immediate publication of a rule, publish that rule without consultation, subject to subsequent comment and possible appeal to the CBE or the court.

[24] The position is thus that with effect from 1 October 2011 the interim policy set out in the 2011 notice operated as a rule of the SACAP. Rule 2.1 of the Code, which in its current form was promulgated in 2009, stipulates that a registered person may only undertake architectural work which is identified in the category of registration in which he or she is registered in terms of s 18. As from 1 October 2011, the architectural work for the four categories was determined by a rule promulgated in terms of s 36 of the Architects Act. KFT has not sought to have the 2011 notice reviewed and set aside. A rule of this kind could plausibly have been enacted pursuant to the SACAP’s powers under ss 14(g) to (j) of the Act.

[25] The position changed with the withdrawal of the 2011 notice on 30 November 2015. I agree with KFT’s argument that the 2016 notice did not reinstate a regime which precluded Mr Joseph from signing the plans. The 2016 notice stated that a new identification of work would be drawn up and submitted to the CBE and CC. In the meanwhile, so the notice stated, registered professionals undertaking work for a client ‘shall do so in compliance with rule 2.1 of the Code of Professional Conduct’ which meant, so the notice continued, that ‘registered persons may only perform such work as they are professionally qualified and competent to undertake.’ Rule 2.1 does not take the matter further, because it does not delineate the scope of work of the various categories of registered professionals.

[26] In terms of the 2016 notice, a registered professional is simply enjoined not to do work which he or she is not professionally qualified and competent to undertake. In the absence of evidence, I cannot say that it is always beyond the qualifications and competence of an architectural technologist to draw plans for a dwelling with more than two storeys. The applicants did not seek to make the case that Mr Joseph was not, by training and experience, competent to do so; they relied on bright lines supposedly drawn by rules of law.

[27] The regulations promulgated under the Building Act do not take the matter further. They require the involvement of a ‘competent person’ who is registered in the appropriate category of registration in terms of the Architects Act.[3] The regulations do not themselves incorporate by reference any particular notice or rule promulgated under the Architects Act or other legislation.

Ground 2: Wrong test applied by BCO

[28] The second ground, which the CCT also concedes, is that the BCO, in his recommendation that the plans be approved, applied the wrong test. He stated in his recommendation that he the was ‘unable to find satisfying factors indicating that’ the proposed building would have any of the disqualifying features specified in s 7(1)(b)(ii) of the Building Act.

[29] In Turnbull-Jackson v Hibiscus Court Municipality & others [2014] ZACC 24; 2014 (6) SA 592 (CC) the Constitutional Court confirmed the interpretation of s 7(1)(b)(ii) given in Walele v City of Cape Town & others [2008] ZACC 11; 2008 (6) SA 129 (CC), viz that the decision-maker must be positively satisfied that the disqualifying factors do not exist. Walele also holds that although the decision-maker must him- or herself be positively satisfied that the disqualifying factors do not exist, the BCO’s recommendation is the proper means by which information on disqualifying factors should be placed before the decision-maker.

[30] In the present case, the BCO’s recommendation did not serve its intended function, because it was stated in negative rather than positive form. There is nothing to show that the decision-maker had any additional relevant information to reach the positive conclusion required by Walele and Turnbull-Jackson.

Ground 3: The building violates s 7(1)(b)(ii)

[31] The applicants allege that the decision-maker could not be positively satisfied that the building would not disfigure the area, or be unsightly or objectionable, or derogate from the value of neighbouring properties. They say that even if the building were set back 3.5 m from the street in accordance with the DMS, it would still be closer to the street than the other houses in Meadway, which were built at a time when greater street setback rules applied.

[32] The historical perspective is something which the applicants allege should have been taken into account by the decision-maker. It was not enough for the decision-maker to find that the proposed dwelling was within the restrictions of the recently enacted DMS. The earlier limitations, which gave the streetscape its character, should have been considered. According to the applicants, KFT’s ‘multi-storey and over-bulked building’, stretching from one common boundary to the other ‘brutally interrupted’ the previous uniform street line. The Shiel property would lose much of its view towards the sea while the Spring Lights property’s view towards the mountain would be largely obliterated.

[33] Qualitative considerations of this kind are always hard to assess. Since there are other valid grounds to impugn all three sets of plans, it seems preferable to leave this ground open, particularly since it has not yet been adjudicated by the mandated decision-maker in accordance with the correct legal test.

Ground 4: Absence of environmental authorisation

[34] During the CCT’s assessment of the plans, an official from the Environmental and Heritage Management department noted that the proposed basement fell within activity 19 of List 1 of the Environmental Impact Assessment Regulations promulgated on 4 December 2014 pursuant to the National Environment Management Act 107 of 1997. Item 19 includes the excavation, removal or moving of soil of more than 5 m³ within a distance of 100 m from the high watermark of the sea.

[35] The applicants allege, and it has not been denied, that the excavation of the basement parking area of 240 m² exceeded the 5 m³ limit and that the property is situated within the 100 m margin. An environmental authorisation from the Western Cape Provincial Department of Environmental Affairs was thus required. There is no explanation from the CCT as to why the comment from its environmental and heritage official was disregarded.

[36] In the absence of contrary evidence, I must perhaps accept that the property at its western-most point is less than 100 m from the high watermark. I think, though, that better evidence of this should have been given. According to Google Earth tools, the western-most point of the property is about 90 m from the beginning of the sand of the beach. It is not self-evident, to one unfamiliar with Camps Bay, that the high watermark is within the next 10 m.

[37] The applicants’ counsel accepted that even if this ground of review were made out, the excavation has long since taken place and that the property in this respect could not realistically be restored to its pre-excavation state.

Ground 5: Height to wallplate and top of roof

[38] In terms of the By-Law, KFT’s property is zoned General Residential (‘GR’) 2. Item 42 of the DMS, which relates to properties zoned GR2-6, makes item 22 (which relates to properties zoned GR1) applicable to a building that is a ‘dwelling’.

[39] If building plans exceed the limitations imposed by the DMS, the submitting party must apply for and obtain a departure before the plans can be approved. Although two departure applications were made subsequent to the approval of the 2015 plans, neither has been granted. It follows that if the various plans violate applicable DMS restrictions, they were unlawfully approved.

[40] In terms of the table forming part of item 22, the following height restrictions apply to a property the size of KFT’s property: (a) 8 m from base level to wallplate; (b) 10 m from base level to top of roof. Where a building is permitted within 3 m of a common boundary, the height is limited to 4 m measured from base level to top of roof, save that this further limitation is not applicable to any part of the building within the first 12 m along a common boundary measured from the street boundary.

[41] The following expressions, as defined when the 2015 plans were assessed, are relevant:

‘base level’ – ‘an imaginary plane drawn horizontally at the average ground level of the building, or vertical division of the building, is at the bottom of the ground floor and is directly above or on top of the ceiling of any basement;

‘average ground level’ (‘AGL’) – ‘the average of the highest and lowest existing ground levels immediately abutting the external elevational plane or wall cutting into the ground of a building or vertical division of a building…’;

‘basement’ – ‘that space in a building between a floor and ceiling, including such floor or ceiling, which is not intended as habitable space and which is completely below the existing ground level and remains underground, except that it may include vehicular access from a road, provided that such access may only be obtained at a downward or level angle’;

‘wallplate’ – ‘the lowest point of a longitudinal member, truss, bracket, pillar, post, structure or any other similar device as determined by the City, supporting a roof’;

‘top of the roof’ – ‘the top of the roof ridge in the case of a pitched roof, or the top of the parapet where a parapet extends above the roof’.

[42] By way of an amendment promulgated on 1 July 2016, the definition of ‘base level’ was amended to delete the words ‘is at the bottom…’ to the end. ‘Basement’ was redefined to mean ‘that space in a building between a floor and ceiling, including such floor or ceiling, which protrudes not more than 1.5 m above any point on the existing ground level’

[43] KFT’s property is located in the Camps Bay and Bakoven overlay area and is thus subject to item 190 of the DMS. Sub-item 2 provides that no building shall exceed three storeys in height. Sub-item 3 states that ‘no point on the facade of any building … shall be more than 10 m above the level of the ground abutting the facade immediately below such point.’

[44] The alleged height of the building was shown in section AA of the 2015 plans. The plans did not reflect anything which could be described as the ‘base level’ defined in the DMS. Measurements were taken from something marked as ‘NGL’, presumably ‘natural ground level’. NGL was stated to be at 5800 millimetres (5.8 m), presumably above mean sea level (‘MSL’). The basement was shown to be entirely below NGL, and one can deduce that the upper side of the basement floor (‘TOC’ – top of concrete) was at 3400 mm above MSL.[4] On the assumption that NGL was meant to be the defined ‘base level’, it was necessary – as at 2015 – for the basement to be entirely underground. If it were not, it would no longer qualify as a ‘basement’ and would instead be a ‘storey’. Since the plans reflected three storeys above the basement, item 190(2) of the DMS would have been infringed unless the basement qualified as such.

[45] According to the plans, the height from NGL to wallplate was 7800 mm, just within the 8 m restriction.[5] Because the roof was a flat slab with modest parapets (labelled ‘upstands’), the height from NGL to top of roof was 8400 mm, comfortably within the 10 m restriction.

[46] An architect, Mr Wilkinson, who submitted an objection on behalf of Mr Shiel, pointed out to the CCT that the ramp to access the basement would be impossibly steep if it were wholly underground. He believed that the basement would have to protrude above NGL in order to be functional, meaning it would no longer be compliant as a ‘basement’. The submitted plans suggested a flat site (there were no contour lines), but Mr Wilkinson estimated that there was a slope of 1 m – 1.5 m over the length of the site. He urged the CCT to require the owner to show MSL heights.

[47] In addition to adopting Mr Wilkinson’s points, the applicants draw attention to the fact that the south elevation in the 2015 plans shows the property to be significantly stepped as one looks at it from Meadway. The south elevation shows the beginning of the ramp down to the ‘basement’ to be at a level 1500 mm lower than the parts of the building to the right of the ramp (the entrance hall and kitchen). The applicants say, without contradiction, that the property is not in fact stepped in this way, and no such stepping is reflected in section AA.  

[48] The CCT approved the plans without obtaining the information it needed to make an informed decision. The CCT needed to know what the building’s AGL was. The DMS authorises the CCT to adopt three different methods to determine the AGL of a building:[6] it may (a) determine the AGL from measurements supplied on a building plan; or (b) deem a level to be the AGL, based on measurements interpolated from a contour plan, local height benchmark or other information held by the CCT; or (c) require the owner to get a surveyor to measure the levels.

[49] The 2015 plans did not contain measurements which allowed the CCT to determine the building’s AGL. The plans did not even purport to specify a ‘base level’ as defined. The CCT did not adopt either of the other two methods to clarify the matter.[7] There is no explanation as to how they believed the basement could be accessed if it were completely below the AGL.

[50] Subsequent events show that Mr Wilkinson’s concerns were well-founded. The approval of the plans was irrational. It is not rational to approve plans which show features that cannot sensibly be used. In any event, armed with the approved plans, KFT might contend that the CCT had accepted 5800 mm as the base level, even if later investigation showed that the true base level was lower. (In the event, the true base level has been ascertained to be higher. The CCT has accepted the measurement of the applicants’ land surveyor, Mr Trevor Stander, who determined the base level to be 6060 mm.)

The 2017 plans

[51] The only marked-up features of the 2017 plans were: (a) the introduction of a lift, with lift shaft features in the basement and three storeys; and (b) the creation of a store room and toilet in the basement. The CCT has stated that only these features were examined. However, other important changes were not marked up and thus did not receive attention.

[52] The 2017 plans did not remedy the failure in the 2015 plans to specify the ‘base level’ of the building. The ‘NGL’ of 5800 mm does not feature in the 2017 plans. Instead, there is a level indicated as ‘street level’. Its height above MSL is not stated. There is no indication as to how ‘street level’ relates to ‘base level’. Significantly, and as Mr Wilkinson predicted, the basement is now shown to protrude 650 mm above ‘street level’. (It so happens that by virtue of the 2016 amendment, the basement could protrude up to 1.5 m above ‘existing ground level’ without contravening the ‘basement’ definition.)

[53] On the assumption that Mr Joseph was treating ‘street level’ as the equivalent of his 2015 ‘NGL’, the 2017 plans show that the basement floor TOC was now at 3750 mm above MSL,[8] ie 350 mm higher than the 2015 TOC of 3400 mm. The new building was significantly higher than the previous one, not only because it was starting out at a higher level but because the basement and storey heights were increased (the basement from 2400 mm to 2700 mm, the storeys from 2400 mm to 2800 mm).

[54] The CCT complains that none of these changes were marked up. The CCT does not explain how changes in height should be marked up. In fairness to Mr Joseph, I should mention that the heights of the three storeys in section AA (2800 mm) are circled and initialled, apparently by Mr Joseph, and that the accompanying application for approval stated that the rider plans concerned ‘new lift, heights’, so the CCT cannot so easily be absolved.

[55] I agree with the applicants’ contention that, having regard to the significant changes in height and the other features that were marked-up, the 2017 plans were not rider plans; they were plans for a different house, starting at a different level, with the basement and each storey being higher than previously shown. None of the height differences were de minimis.

Ground 1: Mr Joseph not a ‘competent person’

[56] This ground of relief fails in relation to the 2017 plans.

Ground 2: The top of roof height

[57] The only explicit concession made by the CCT is in respect of the height restriction to top of roof. The CCT makes the point that the relationship between ‘street level’ and ‘base level’ is not apparent from the plans. The CCT accepts Mr Stander’s base level measurement of 6060 mm above MSL.[9] The CCT says that section AA shows the height to the top of the roof (ie to the top of the parapet) to be at the 10-metre limit. The CCT says, however, that this measurement has been taken from the soffit of the ground floor slab,[10] which is at 6530 mm above MSL. If the measurements had been taken from the true base level of 6060 mm, the top of the roof would have exceeded the 10-metre restriction by 470 mm.[11] (The CCT appears to me to be mistaken as to the true level of the ground floor soffit. The deponent, Mr Gerber, does not give his source. According to Mr Stander’s certificate, the ground floor soffit was at 6700 mm, so on Mr Gerber’s logic the 10-meter restriction was exceeded by 640 mm.)

[58] On the assumption that Mr Joseph intended his 2017 ‘street level’ to be the same as his previous ‘NGL’, the height to top of roof was 10,650 mm.[12] By unjustifiably measuring the top of roof height from the ground floor soffit, Mr Joseph was able to reflect the top of roof as exactly 10 m.

[59] The plans should not have been approved without an explicit statement as to the building’s base level. There was no justification for Mr Joseph to treat the ground floor soffit as the base level. To the extent that the ‘street level’ could be regarded as a synonym for the base level, the plans on their face violated the 10 m roof height restriction. And objectively, having regard to the true base level as subsequently measured, the 10 m roof height restriction was likewise violated.

[60] Moreover, the 2017 building was significantly higher than the 2015 building. This was a combination of various factors: the basement started at a higher level, and the height of the basement and of each of the three storeys was increased. Since there had previously been objections in terms of s 7(1)(b)(ii), the CCT may have revised its assessment if the changes in height had been marked up.

Ground 3: Wallplate height (terrace roofs)

[61] Section AA of the 2017 plans showed that the second storey terraces would now have roofs slanting down to the boundary lines. This change was not marked up. The height of the wall on which the slanting roofs rest at their lowest point was annotated as the wallplate height. There is a printed measurement on the left-hand side purporting to show the wallplate height as 8000 mm.

[62] A analysis of the plans shows that this was not the true position. If one measures from the street level, the height was 8809 mm.[13] If one measures from the ground floor soffit (as Mr Joseph did in the case of the roof height), the height was 8159.[14] Only if one measures from the ground floor TOC would one come within the 8 m limit (7959 mm).[15]  Using the ex post facto information about the true base level, the wallplate height was 8629 mm.[16]

Ground 4: Wallplate height (roof slab)

[63] The applicants make the further point that wallplate height for the slanting roofs of the terraces is not the only relevant wallplate height. The slanting terrace roofs only featured in the front (southern) half of the building’s second storey. On the back (northern) half there were no side terraces and thus no slanting roofs. The wallplate for the roof above the northern half of the second storey was thus higher than for the slanted terrace roofs. The applicants also point out that Mr Joseph only mentioned a wallplate height in section AA, where he could utilise the slanted terrace roofs to show a lower wallplate height. In the elevations, particularly the north elevation, the drawings are silent as to the location of the wallplate.

[64] Although the applicants only make this point with reference to the roof above the back half of the second storey, it is more fundamental. The wallplate is the lowest point of a longitudinal member, structure or other device ‘supporting a roof’. If different roofs of a building are supported by different longitudinal members or structures, each wallplate needs to be within the 8-m limit. The roofs of the terraces are specified as IBR[17] sheeting. The main roof of the house, which covers the interior rooms, is a 200-mm-thick cement roof slab to engineer’s design and specifications. The wallplate for this roof is thus at the height of the slab’s soffit. It is perfectly obvious that the walls on which the IBR roofing rests at its lowest point provide no support for the cement slab. No downward pressure from the cement slab is distributed along the IBR sheeting. The plans do not show any structural connection between the roof slab’s soffit and the IBR roofing. The clearest practical proof of this fact is that the roof slab was cast in June 2018, while the slanting IBR roofs have not yet been installed.[18]

[65] The wallplate height for the roof slab is 9650 mm measured from street level,[19] 9000 mm measured from the ground floor soffit,[20] and 8800 mm measured from the ground floor TOC.[21]  The wallplate height is 9470 mm measured from the true base level.[22] This applies not only to the roof above the northern half of the second storey; it applies to the entire cement slab roof above the second storey.

Ground 5: Boundary-building

[66] The applicants’ boundary-building complaint is directed at the 2017 and 2018 plans, though it is equally applicable to the 2015 plans. The table in item 22 of the DMS provides for a building line of 3 m from common boundaries. However, there is a 0.0 m building line (ie the owner can build on the common boundary) ‘for the first 12.0 m measured perpendicular from street boundary and 0.0 m for 60% of the total remaining linear distance along all common boundaries around land unit and 3,0 m for remainder’. This is subject to the further proviso (item 22(e)) that any portion of the building which contains an external window or door facing onto a common boundary shall be set back at least 1.5 m from such boundary. The length of the portion to be set back must be not less than 3 m. A building must also be set back 3.5 m from the street boundary.

[67] KFT’s property is 495 m². It has three common boundaries. The measurements on the ground floor plans show that the linear length of the three common boundaries is 78.81 m.[23] There is a question as to whether the 12-metre allowance in item 22’s table can apply to more than one common boundary. If, as the applicants contend, it is limited to one common boundary, the linear length of the three common boundaries after subtracting the 12-meter allowance is 66.81 m. Item 22 allows boundary-building on 60% of this remainder, ie 40.09 m, with 26.72 m having to be stepped back 3 m. In total, therefore, KFT could build flush on the boundary to the extent of 52.09 m.[24]

[68] If the 12-metre allowance can apply to two common boundaries, the linear length of the three common boundaries after subtracting the 24-metre allowance is 54.81 m, of which 60%, viz 32.89 m, would be available for boundary-building, with 21.92 m having to be stepped back 3 m. In total, KFT could build flush on the boundary to the extent of 56.69 m.[25]

[69] The 2015 plans showed that on each of the two common boundaries running from Meadway the house would be built on the common boundary for 12 m, whereafter it was stepped back by 3 m.[26] If the 12-metre allowance were limited to one common boundary, the owner could still build on the other common boundary from the street, provided it fell within the 40.09 m allowance, which it comfortably did. There was no material alteration in this respect in the 2017 plans.

[70] However, in terms of item 22(c), where a building is permitted within 3 m of a common boundary, the height is limited to 4 m from base level to top of roof. This is subject to the proviso in item 22(c)(iii) that ‘within the first 12 m along a common boundary measured perpendicular from the street line’ the heights in the table (8 m and 10 m to wallplate and top of roof) apply. The applicants contend that only one common boundary may benefit from this proviso. All three sets of plans reflect a building which was to be constructed on both common boundaries running off Meadway at heights exceeding 4 m. To the extent that this attack was raised with reference to the rider plans, it was a feature already present in the 2015 plans. The applicants’ counsel told me that their argument in this respect was at odds with the CCT’s view, which the latter has presumably applied in many cases.

[71] In the case of the building lines specified in the table, I am inclined to think that the initial allowance for building on common boundaries is 12 m along one boundary from the street, not 24 m along two boundaries from the street. The formula for determining the extent to which there can be building on common boundaries proceeds with reference to a 12-meter initial allowance.

[72] It does not follow that the same interpretation applies to item 22(c)(iii). The one restriction is concerned with the linear extent of boundary building, the other with height. Item 22(c)(iii) refers to a height restriction within the first 12 m ‘along a common boundary measured perpendicular from the street boundary’. This is not the first step in a formula setting a limit for all common boundaries. The use of the indefinite article suggests that any common boundary running from the street boundary qualifies. I thus reject this ground of review.

Ground 6: Windows on boundary wall

[73] The applicants raise this ground with reference to the rider plans, but it is equally applicable to the 2015 plans. Boundary-building is subject to the limitation (item 22(e)) that stretches of wall containing ‘external windows’ facing the common boundary must be set back at least 1.5 m. The east and west elevations of all three sets of plans show that the bedrooms on the west and east side of the first and second storeys have doors and windows giving out on to balconies. The balconies, which stretch to the common boundaries, have large openings facing the common boundaries, while on the north and south sides there are balustrades.[27] The openings on the common boundaries do not contain glass. The openings as actually constructed appear to differ dimensionally from the plans.[28]

[74] The ordinary meaning of ‘window’ is an opening in a wall to admit light or air and to allow occupants to see out. Although the opening usually incorporates fixed or moving glass panels, this is not an essential feature of a window.[29] In the context of item 22(e), the obvious purpose is to limit intrusion into neighbours’ privacy, though the hearsay evidence in this case suggests that openings of this kind may also increase the fire hazard for neighbours. In my view, therefore, the boundary walls of the balconies contain ‘external windows’ facing onto the common boundary.

[75] In an email of 21 May 2018, the BCO told Mr Shiel that the balconies ‘were approved with open sides on the boundary’. They were also open to the front and back of the property, so ‘these open porches can therefore not be regarded as enclosed spaces/rooms’. In justifying this, the BCO seems to have had in mind item 121(1)(a)(ii) of the DMS. Item 121(1) lists various structures which may encroach upon the building lines prescribed earlier in the DMS. Item 121(1)(a)(ii) allows ‘open and uncovered stoeps’. The BCO ignored the requirement that such balconies must not only be open but uncovered. The balconies here were not uncovered. The plans show that the floors of the second-storey balconies form roofs over the first-storey balconies, and that the second-storey balconies are to have IBR roofs.[30] The openings onto the common boundaries were thus impermissible.

Ground 7: Covered balconies

[76] The applicants claim, quite apart from the window issue, that item 121(1)(a)(ii)[31] under no circumstances permits covered balconies to encroach the building line. Although this complaint is directed at the rider plans, it is again equally applicable to the 2015 plans.

[77] I reject this ground, and I understood the applicants’ counsel not pursue it. Item 121(1)(a)(iii) does not state that any balconies constructed within the applicable building line must be open and uncovered; it says, in effect, that a balcony within the applicable building line will not be regarded as encroaching if it is open and uncovered. If, as here, the balconies are covered, they would not benefit from the exemption in item 121 but they might nevertheless be permitted as ordinary boundary structures within the linear extent permitted by item 22. Of course, such balconies may not have windows facing towards the common boundaries but that is a separate issue.

Ground 8: The building violates s 7(1)(b)(ii)

[78] Again I will leave this question open. I observe that this objection is not merely a repeat of an objection to the 2015 plans. Because of the height alterations in the 2017 plans, the factors in s 7(1)(b)(ii) had to be adjudicated with reference to the new features of the building, including height. Notionally a decision-maker could find that the 2015 building complied while the 2017 building did not.

[79] I should mention, here, an additional issue raised by the applicants with reference to s 7(1)(b)(ii). They say that according to an official in the CCT’s fire chief’s department, the approved plans were not circulated to that department, because the fire department would never have approved plans showing boundary walls with openings facing onto the common boundary. The applicants allege that without an assessment from the fire department, the decision-maker could not be positively satisfied that the building would not be ‘dangerous to life or property’. The BCO’s letter to Mr Shiel of 21 May 2018 seems to confirm that the plans were not considered by the fire department. This particular complaint may be well-founded. However, I only have hearsay evidence as to the views of the fire department. This aspect may warrant careful consideration if and when new plans are submitted.

The 2018 plans

[80] The main marked-up features of the 2018 plans were: (a) on the ground floor, the construction of a covered patio at the back of the house, a store in the north-east corner, a sauna and shower-room in the north-west corner and a jacuzzi near the western common boundary; (b) on the first and second storeys, patios at the back of the house leading off the bedrooms; and (c)  a small room on top of the roof slab, described as a ‘lift motor room’. The roof top is shown as ‘non-trafficable’, ie not for ordinary use.[32]

[81] Unlike the previous plans, the 2018 plans record a ‘base level’, namely 6100 mm above MSL. The source of this figure is not stated. In the event, it is only 40 mm higher than Mr Stander’s level of 6060 mm, which the CCT has accepted as correct. In assessing height compliance, one would – on the basis of Mr Stander’s figure – have to add 40 mm to any heights based on the 2018 plans.

[82] As I have said, one can deduce from the 2017 plans that the basement floor TOC was at 3750 mm above MSL. The 2018 plans show the basement floor TOC at 4730 mm above MSL, nearly one metre higher than the 2015 plans. Once again, one may legitimately query whether such a building, taking into account also the other marked-up changes, was in truth the same building as was approved in the 2017 plans, ie whether the 2018 plans qualified as rider plans (nothing turns on this).

Ground 1: Mr Joseph not a ‘competent person’

[83] This ground of relief cannot be upheld in relation to the 2018 plans.

Ground 2: Wallplate height (terrace roofs)

[84] The CCT complains that changes in height were not marked up. The CCT alleges that when section CC is analysed, it is apparent that the wallplate height is 8.7 m, which exceeds the permissible 8 m. Although the CCT does not say so, it is apparent that the CCT is referring to the wallplate height of the slanting terrace roofs.[33] This is the only ground explicitly conceded by the CCT.

[85] In section CC, the dotted line which marks the terrace roof wallplate height (ie at about 14,800 mm above MSL) was annotated by Mr Joseph as ‘8 m wallplate height’. Self-evidently this was a misrepresentation, since the line is in fact at 8.7 m above base level. One could only have arrived at a figure of (about) 8 m by taking the measurement not from the base level of 6100 mm but from the level of the ground floor TOC, viz 6780 mm. What makes the misrepresentation all the more striking is that in section AA the dotted ‘8 m wallplate height’ line (ie the maximum permissible height) is correctly located at 14,100 mm. In section AA the dotted line is well below the terrace roof wallplate height; indeed, it runs through the third-storey rooms at less than half the height of those rooms.

Ground 3: Wallplate height (slab).

[86] As I have already explained, the wallplate height for the terrace roofs is not the only relevant wallplate. Although the CCT does not make this point, it is apparent from the 2018 plans that the wallplate height for the cement roof slab is 9700 mm above base level.[34] This height, which again exceeds the 8 m limit, is nowhere reflected in the 2018 plans.

Ground 4: Top of roof height  

[87] Disregarding the roof room, section CC shows that the top of the roof height (ie the top of the parapets) is at 16,600 mm above MSL.[35] If one subtracts the base level of 6100 mm, one reaches a roof height of 10,500 mm, which exceeds the permitted 10 m.

[88] Section AA correctly shows that the 10-metre permissible roof height limit is at 16,100 mm (this is the dotted line just above the top of the roof slab, which is at 16,000 mm). None of the sections show anything marked ‘top of roof’ as defined in the DMS. One is left to wonder whether Mr Joseph wanted the CCT’s officials to assume that the ‘top of roof slab’ was also the ‘top of roof’ for DMS purposes. The truth is, though, that there are 600 mm upstands (parapets) above the slab, and these must be included to arrive at the DMS’ ‘top of roof’. Mr Joseph included the upstands in the maximum permissible roof height in the 2017 plans but failed to do so in the 2018 plans.

[89] In regard to the roof room (marked as a ‘lift motor room’), section AA gives its dimensions as 2 m x 2 m. A separate drawing of the room shows that it has a door, presumably to give access to lift technicians. Next to the roof room, at a low height (not given), is a 900 mm-wide ‘sliding hatch’ to give access to the roof.[36] The definition of ‘height’ in the DMS states that certain structures must be left out of account for purposes of height control. One such structure is a lift shaft with maximum dimensions of 2.5 m (horizontal) and 2 m (vertical). The roof room’s dimensions as approved are within this exclusion.

Ground 5: Facade height

[90] In terms of s 190(3) of the DMS, no point on the building’s facade may be more than 10 m above the level of the ground abutting the facade immediately below such point. The violation of item 22 of the DMS shows that item 190(3) has also been contravened. Because the restriction in item 190(3) is measured from the level of the ground abutting the relevant facade, and not from the defined ‘base level’, the violation of item 190 will be greater than the violation of item 22 towards the front of the building, where the ground level is lower. At the front of the property, the facade is 11,110 mm above ground level.[37]

Ground 6: Boundary-building

[91] This has already been discussed in relation to the 2017 plans. In the 2018 plans, the proposed storeroom in the north-east corner and the new sauna and jacuzzi in the north-west corner would have added some additional construction on or near the common boundaries, but still not nearly enough to violate item 22, even on the applicants’ interpretation.

Ground 7: Windows in boundary walls

[92] I have already dealt with this in relation to the 2017 plans.

Ground 8: covered balconies

[93] I have already dealt with this in relation to the 2017 plans.

Ground 9: Outbuildings

[94] As with other properties in the area, KFT’s property is subject to a title deed restriction (clause C8) that an owner may not build any ‘outhouses or adjuncts’. A building plan which shows features violating title deed restrictions is a plan that does not comply with ‘any other applicable law’ within the meaning of s 7(1)(a) of the Building Act. KFT has not followed any procedure for obtaining a relaxation of the title deed restriction.

[95] The storeroom and the sauna and shower-room proposed for the north-east and north-west corners of the property are roofed structures and are plainly ‘outhouses or adjuncts’.[38]

Ground 10: BCO applied wrong test

[96] The applicants claim that the recommendation of the BCO formulated the wrong test in regard to disfigurement. They allege that he considered only whether the proposed changes were within the permitted development parameters. I reject this ground. Although the BCO said that compliance was an important factor, he added that even if plans are planning-compliant, one must consider whether the alterations and additions ‘will be so unattractive or intrusive that it exceeds the legitimate expectations of the surrounding property owners’. This is in accordance with the test approved by the Constitutional Court in Harrison.

Ground 11: The building violates s 7(1)(b)(ii)

[97] Again, I leave this question open. And again, this objection is not merely a repeat of an objection to the 2015 and 2017 plans. Because of the height alterations in the 2018 plans and the new outhouses/adjuncts, the factors in s 7(1)(b)(ii) had to be adjudicated with reference to the new features of the building.

The building as actually constructed

[98] Section 4 of the Building Act prohibits the construction of buildings without approved plans. In terms of regulation A25(5) it is a criminal offence to deviate to a material degree from approved plans unless the deviation has been approved. The applicants allege that the building as constructed differs from the 2018 plans in the respects set out hereunder.

Complaint 1: Height

[99] According to Mr Stander’s survey certificate, the top of the roof slab is 15,910 mm above MSL. Since the 2018 plans reflect the top of the roof slab as 16,000 mm, the heights shown in the 2018 plans have not been exceeded.

[100] Mr Stander’s inferred wallplate height for the cement slab is 15,710 mm. KFT’s land surveyor, Mr M E Jakins, issued a certificate in May 2018 stating that the wallplate height was 15,700 mm, so there is no material difference between him and Mr Stander. Mr Jakins evidently appreciated that the cement roof slab’s wallplate was not dictated by the wallplate of the slanted terrace roofs.

Complaint 2: Trafficable roof deck

[101] The applicants allege that, contrary to the approved 2018 plans, KFT has constructed a trafficable roof deck on top of the concrete roof slab. There is photographic evidence that the entire slab has been covered with slatted timbers.[39] Quite clearly this expensive exercise has nothing to do with the slab’s function as a roof. The intention must be to use it as a roof deck, and this is confirmed by what KFT’s representative said at the meeting of 7 February 2019 (below) and by the second departure application.

[102] A roof deck has not been approved. Furthermore, such a deck would need balustrading around the perimeter. Such balustrading would protrude above the defined ‘top of roof’. This would be impermissible for the reasons I explained in paras 28-37 of Epstein v City of Cape Town & another [2019] ZAWCHC 84 .

Complaint 3: Roof room

[103] It is apparent from photographs attached to the applicants’ founding and the CCT’s explanatory affidavit[40] that the horizontal length of the roof house (running from south-north) comfortably exceeds 2.5 m. My rough assessment is that the length is about 4 m. This is a deviation from the approved plans.

[104] I notice from these photographs that the 600 mm parapets (upstands) shown in the plans are not visible. It is unclear whether this is because the decking has been placed at the height of the top of the parapets or because the parapets have simply not been built. The latter seems unlikely, since the waterproofing would need to be stepped up against the base of parapets.

Complaint 4: Boundary-building beyond the 12-metre mark

[105] The applicants allege that KFT has engaged in boundary-building beyond the 12-metre mark measured from the street boundary. The applicant’s principal deponent says that he has measured the distance himself and can confirm that it far exceeds the 12-metre mark. To the extent that this allegation is aimed at showing a violation of the DMS, it must fail, because the DMS does not prevent boundary-building beyond the 12-metre mark. Provided the total linear extent of permitted boundary-building is not exceeded, the portion of boundary building measured from the street boundary could legitimately exceed 12 m, though beyond the 12-metre mark a height restriction of 4 m would be applicable.

[106] However, the approved 2018 plans do not authorise boundary-building beyond 12 m from the street boundary except in respect of the new features in the north-east and north-west corners. KFT has not denied that it has engaged in boundary-building beyond 12 m. Such building is thus in violation of the approved plans.

Complaint 5: No approved plans for boundary wall

[107] KFT constructed a new boundary wall between its property and Mr Shiel’s property. This was not the subject of any approved plans. The applicants allege, furthermore, that the wall did not have a proper foundation. Mr Shiel obtained an engineering report which indicated that the wall was unstable and could collapse. This subsequently happened, causing damage to Mr Shiel’s property.

[108] The applicants also allege that the construction of the perimeter wall violates clause C9 of the title deed, which provides that the owner may not enclose the ground ‘with any unsightly fence’ and that ‘before erecting any fence or enclosure shall be bound to obtain the written approval of the company as owners of the estate as a whole, to the materials and mode of erection’. This is something which the CCT will need to consider if and when plans for the perimeter wall are submitted.

The 2017 and 2018 plans – conclusion

[109] It will be apparent that the approval of the 2017 and 2018 plans must be set aside on multiple grounds. The applicants are also entitled to declarations as to the respects in which the building deviates from the approved plans (the 2018 plans), though with the setting aside of the 2017 and 2018 plans, the declaration is perhaps not of great practical significance.

The 2015 plans – delay

[110] In assessing the question of delay, it may helpful to consider at the outset what is practically at stake. The building as actually constructed accords most closely with the 2018 plans, particularly in regard to height, though there have been material deviations. Once the 2017 and 2018 plans are set aside, allowing the 2015 plans to stand would not be of practical use to KFT. In order to build in accordance with the 2015 plans, construction would have to start at the levels indicated by those plans, and the height of the basement and of each of the storeys would have to comply with those plans. The entire basement would have to be underground. This would require the whole existing structure to be demolished.

[111] Although KFT opposes the review of the 2015 approval, it does not say that if the 2015 plans are allowed to stand it has any intention of re-building the dwelling in line with those plans. There is unrefuted evidence that the basement would be impossible to access.

[112] The differences between the 2015 and 2018 plans are so substantial, particularly in regard to the level of the basement floor and the heights of the basement and three storeys, that it is impossible to say that the dwelling which KFT began to build in 2016 was the dwelling approved in the 2015 plans. In terms of s 7(4) of the Building Act, approval lapses a year after the approval date unless the erection of the building is started within the 12-month period or unless the local authority extends the period on written request. It is thus fairly arguable that the approval of the 2015 plans has elapsed.

[113] However, since the implications of s 7(4) were not raised in argument, I shall assume that the approval of the 2015 plans would still be in force if I did not set it aside on review. In regard to delay, applications for building plan approval do not need to be advertised or notified to neighbours. The first intimation that building plans have been approved will usually be when building work starts. The early stages of the building project may not reveal features which a neighbour regards as impermissible or objectionable. Neighbours are not obliged to assume that local authorities will approve plans in violation of the law (cf Van der Westhuizen & others v Butler & others 2009 (6) SA 174 (C) at 184I-185C). Departures from the DMS or title deeds would require notification to neighbours.

[114] Mr Shiel knew about the 2015 plans before they were approved. Although Mr Wilkinson lodged an objection on his behalf, there is no evidence that the CCT told Mr Shiel that the plans had been approved in November 2015. There was an hiatus until October 2016, when KFT’s predecessor-in-title, Ms Veltman, notified the CCT that work was beginning. Neighbours would have thus have seen the first signs of building activity in late October 2016, from which they could have inferred that building plans had probably been approved. Even Mr Shiel, though, would not have known that the building was taking place in terms of the plans to which he had objected in 2015.

[115] Apart from the perimeter wall (a matter which was not the subject of any building plans), the building under construction first attracted Mr Shiel’s adverse notice in early May 2018, when he complained to the CCT about windows on his boundary. Not long afterwards he complained also about the excessive heights of the basement and three storeys, which exceeded the dimensions in the approved 2015 plans. He was evidently unaware of the 2017 plans which had been submitted in June 2017 and approved in October 2017.[41]

[116] In late May 2018 Mr Shiel engaged Mr Willem Buhrmann, an architect, to advise him. Mr Shiel must have made contact with the CRA as well, because we find Mr Willemse being copied on their correspondence. Mr Shiel had evidently now obtained a set of the approved 2017 plans, and I am satisfied that by late May 2018 Messrs Shiel, Buhrmann and Willemse had access to the approved 2015 and 2017 plans.[42]

[117] By letter dated 28 May 2018, the CCT notified Mr Shiel that KFT had applied for a departure from the DMS so as to allow the ‘height of the stairwell and covered roof patio’ to be 12 m rather than 10 m from the base level to the top of the roof. He was afforded until 3 July 2018 to lodge an objection. In all likelihood a similar notice was addressed to Spring Lights and other neighbours. Attached to the notification was an early version of the 2018 plans which, contrary to the plans subsequently submitted for approval, reflected a trafficable roof garden.[43]

[118] On 2 and 3 July 2018 Mr Shiel and the CRA lodged detailed objections to the departure application. It is clear from the objections that they were aware of the approved 2015 and 2017 plans. Mr Willemse set out a number of reasons why in his view the approval of the 2015 plans had been irregular. In his replying affidavit file on 3 November 2020, Mr Willemse says that he has no recollection of having been involved in any discussions, or of having examined the 2015 plans, in mid-2018. That may be so, but he has evidently forgotten that he indeed studied the plans in mid-2018. It appears from the replying papers that Mr Buhrmann only ever had them as email attachments, not in hardcopy. It seems probable that Mr Willemse saw them in the same format. By the second half of 2020 these emails may well have been automatically or manually deleted from their email folders.

[119] In the meanwhile, and notwithstanding the fact that the departure application had not yet been approved (it never was), KFT cast the roof slab at a height we know to have been in breach of the DMS and in breach of the approved 2017 plans.

[120] On 28 August 2018 KFT submitted the 2018 plans, and the building inspector notified Mr Shiel of this on 17 September 2018. Having previously issued a cease-work order, the CCT inexplicably allowed KFT to resume work because these plans had been submitted.

[121] On 17 September 2018 the CCT notified Mr Shiel of a second departure application, this time to increase the height of the stairwell and roof patio to 12.88 m from base level and to permit covered patios on the west and east boundaries beyond the 12 m mark and within 3 m of the common boundaries. Affected neighbours were given until 26 October 2018 to lodge objections. Mr Buhrmann on behalf of Mr Shiel lodged a detailed objection on 22 October 2018, which he copied to Mr Willemse.

[122] The 2018 plans, having gone through several iterations, were approved in final form on 11 December 2018. The building inspector informed Mr Shiel of the approval on 15 January 2019. Mr Buhrmann and Mr Willemse had not yet seen the 2018 plans in final form, and thus did not know that they included the roof room.

[123] On 7 February 2019 Mr Willemse and the Sterns (of Spring Lights) met with Mr Zubair Edwards of KFT to try to resolve the neighbours’ concerns. Mr Edwards claimed that KFT was building in accordance with approved plans. He also mentioned that there would be another departure application to approve a roof garden. A request that Mr Buhrmann be given unfettered access to the building plans and the site to establish the true position ‘on the ground’ was refused.

[124] After several failed attempts, Mr Buhrmann and Mr Willemse were finally permitted to inspect the approved 2018 plans at the CCT’s offices on 22 May 2019 but they were not allowed to make copies. They also saw some of the 2015 plans though apparently not the entire set.

[125] From the applicants’ perspective, the focus during the balance of May and June 2019 was on building work which they regarded as not being authorised by the approved plans. It is this which led them to move urgently by launching their application on 1 July 2019. The review focused on the most recently approved plans, ie the 2018 plans, though the notice of motion referred to rider plans in general. Mr Willemse explained that he and Mr Buhrmann, having not been permitted to take copies, had to rely on their memories.

[126] The CCT delivered an incomplete record on 31 July 2019, eventually supplemented on 6 September 2019. Later in September 2019 the CCT’s attorneys first raised a settlement involving a concession of the review of the 2017 and 2018 plans (at this stage there was no formal review directed at the 2015 plans). This, and the delay in furnishing the record, meant that the timetable contained in the order of 17 July 2019 fell away, and eventually a further order was made on 9 January 2020 in terms whereof, inter alia, the applicants’ supplementary founding papers were to be delivered by 31 January 2020. The supplementary papers were filed on that date, and this is when the 2015 plans were made part of the review.

[127] By late May 2018 Mr Shiel and the CRA knew that the 2015 plans had been approved and had seen the plans. In my view, the 180 day-period in s 7(1) of the Promotion of Administrative Justice Act 3 of 2000 (‘PAJA’) began to run from that time insofar as a review of the 2015 plans is concerned. The running of time was not delayed until the applicants became aware of the respects in which the 2015 approval was unlawful. The 180-day period expired around the end of November 2018. This means that Mr Shiel and the CRA require an extension of time in order to include the 2015 plans in their review.

[128] In regard to Spring Lights, the members of the corporation over the relevant period were Mr Erik Stern and his father Mr Joel Stern. Mr Stern Snr passed away shortly before the application was launched. According to the founding affidavit, Mr Stern Snr resided permanently in New York while his son spent much of the year working abroad. Neither of them was present regularly at the Camps Bay property. I do not think, however, that regular or continuous presence in Camps Bay was necessary. The scale of KFT’s building became apparent during 2018, and the applicants do not say that no member of the Stern family was at Spring Lights’ property during the course of that year. They do not say that Mr Stern Jnr’s wife, Michelle, was not ordinarily resident at the property. Notice of the to departure applications would have been given to Spring Lights as a matter of course. If they had been reasonably attentive and made reasonable enquiries, they could have acquired the same level of knowledge as Mr Shiel and Mr Willemse by mid-2018. Even if the period of Spring Lights’ delay were treated as less than that of the other two applicants, the 180-day period was nevertheless substantially exceeded.

[129] The explicit explanation given by the applicants for the delay is not particularly compelling or complete. However, if one has regard to all the facts set out in their founding and supplementary founding papers, it is not difficult to see why the 2015 approval was at first overlooked as a target for the review. The applicants were confronted with a constantly moving target.

[130] When KFT’s building began to display disturbing features in May 2018, the applicants came into possession of the 2017 plans. It was understandable that they would then focus on those plans as the current approved plans. There was no point in reviewing the 2015 plans if they could not successfully attack the 2017 plans. Just as Mr Buhrmann and Mr Willemse were getting up to speed, they learnt of KFT’s first departure application. If this had been granted, it might have legitimised height features which they regarded as objectionable in the 2017 plans. They thus switched their attention to the departure application and lodged objections in early July 2018.

[131] In mid-September 2018, and before any outcome on the first departure application, they learnt that KFT had submitted the 2018 plans. At the same time they were notified of KFT’s second departure application. Since this might again have legitimised objectionable features in the approved 2017 and/or pending 2018 plans, it needed to be dealt with, and Mr Buhrmann lodged an objection in October 2018.

[132] In mid-January 2019 the applicants learnt that the 2018 plans had been approved though they did not know in precisely what form. The second departure application, which superseded the first, was still pending. There was no purpose in reviewing either the 2015 or 2017 plans if the 2018 plans could not be reviewed. The applicants only got to see the 2018 plans on 22 May 2019, and even then were not allowed to take copies.

[133] Throughout this period (from May 2018 to July 2019) the applicants were also having to deal with construction which appeared not to accord with any of the approved plans. They were lodging complaints and pressing building inspectors to take action. Overall, the impression one gets is that the applicants – private individuals and a voluntary organisation to whose affairs its executive members devote private time – have been left, at their own cost, to tackle a steady flow of DMS violations and unauthorised building work which they were entitled to expect the CCT to address, particularly since the CCT bore a large measure of responsibility by approving non-compliant plans

[134] It is understandable that in launching their review in July 2019, the applicants focused attention on the rider plans, and in particular the approved 2018 plans, since a review of earlier plans would serve little purpose if the 2018 plans stood.

[135] Although the 180-day period in respect of the 2015 approval lapsed in late 2018, I do not think that there can be any substantial criticism of the applicants’ failure to launch a review of the 2015 approval prior to July 2019. Because of the constantly changing circumstances, the timing of the launch of proceedings – 1 July 2019 – was reasonable, and the primary attack by that stage had to be the 2018 approval. A review of the 2015 and 2017 plans independently of a review of the 2018 plans would have been futile, and quite possibly moot.

[136] The applicants can be criticised for not having included the 2015 plans within the scope of the July 2019 review, because a setting aside of the two sets of rider plans potentially revived the 2015 approval as a matter of practical significance. However, the application was launched as one of urgency (because of the need to address unlawful building work) and foreshadowed amplification upon production of the record. Mr Willemse said in the founding affidavit that as far as he had been able to ascertain from the inspection of plans at the CCT’s offices in May 2020, an original set of plans ‘that appear to have been compliant with applicable legislation’ had been lawfully approved some years previously, adding that his conclusions in this regard were subject to correction once the CCT produced the rule 53 record.[44]

[137] It appears that by July 2020 Mr Willemse had forgotten that he had seen the 2015 plans in May 2018 and that he had queried their approval in his objection to the first departure application. I do not believe that Mr Willemse was deliberately misleading the court. He had no reason not to include the 2015 plans in the initial scope of the review if he still had access to them and recalled his objections to them.

[138] Although the amended notice of motion was eventually only filed on 31 January 2020, this was in accordance with an agreed procedural timetable. Although it was only at this time that the applicants, as a matter of form, impugned the 2015 approval, some of the substantive grounds of review raised in the founding papers with reference to the rider plans were also applicable to the 2015 plans. Confining attention to those review grounds which I would uphold, the following can be noted:

(a) The boundary-windows ground, which I have upheld in relation to the 2017 and 2018 plans, is equally applicable to the 2015 plans, and in terms of Harrison this ground should be regarded as in substance an attack on the 2015 approval. Although the applicants failed in their initial founding papers to formulate it with reference to the 2015 approval, the attack was part of the review from the beginning. As I previously explained, the applicants in Harrison failed not because of their failure to formulate the ground in question with reference to the earlier approval but because the substantive attack came too late.

(b) The irrationality of the 2015 approval in relation to height was already discernible in the founding papers. Height has always been a major concern for the applicants, and the importance of accurately establishing the defined ‘base level’ was already a feature of the founding papers. KFT could not have been taken by surprise when the deficiencies in the 2015 plans, on matters of height, were raised in January 2020. Indeed, it knew of the CRA’s attitude to the 2015 approval upon receiving the objection of 3 July 2018 to the first departure application. That objection was attached to the original founding affidavit.

[139] There are two successful grounds in relation to the 2015 approval which were not foreshadowed in the initial founding papers and which are not applicable to the 2017 and 2018 plans. However, one can understand why these grounds, both conceded by the CCT, were not raised earlier:

(a) The fact that the BCO adopted the wrong test in recommending approval was not apparent from the 2015 plans. The applicants only learnt of this when the CCT produced the recommendation as part of its rule 53 record, either on 31 July 2019 or 6 September 2019.[45] Even if the applicants had launched a review of the 2015 approval on other grounds at an earlier time, it could not have been criticised for only raising this particular ground in January 2020.

(b) That Mr Joseph signed the 2015 plans was apparent from the plans themselves, but the fact that he was not a ‘competent person’ was not. It seems that the applicants were only prompted to investigate this question by documents in the record in which the owner appointed Mr Joseph, and he accepted appointment, as the ‘competent person’. While this irregularity was not something of which they had to have knowledge before time began to run, the applicants can nonetheless be forgiven for not having hitherto questioned this aspect.

[140] There is only one review ground in regard to which the attack of 31 January 2020 was unexpected and obviously late. This is the absence of an environmental authorisation for excavation. Excavation dominated activity on the site until mid-2017. The applicants seem to have been alerted to this review ground by the concern raised by an environmental official in a document forming part of the rule 53 record. However, they now say that on the facts an environmental authorisation was clearly required. In this instance, their ignorance or overlooking of the law cannot excuse their delay. It should have been raised at the time the excavations were taking place. I intend to deal with this by way of a carve-out from the review of the 2015 approval.

[141] Prospects of success are an important aspect of condonation. It will be apparent that the review of the 2015 approval enjoys strong prospects; indeed, the CCT has conceded two grounds, and two others are also good.

[142] Finally, there is the question of prejudice to KFT. I do not think that KFT will be materially prejudiced if the 2015 plans are set aside. The simple fact is that from the outset KFT began building a house which was materially different to the one shown in the 2015 plans. The 2015 plans will not allow the dwelling which has been built to be preserved to any material extent. KFT pressed ahead with its building work, knowing about Mr Shiel’s height concerns. It cast its roof slab in June 2018 at a time when it first departure application was still pending and the time for lodging objections had not expired. KFT carried on building in the second half of 2018, knowing that neither of its departure applications had been approved and that there were objections to the departures. And its building work at the stage was in any event far removed from the 2015 plans. It was not building in reliance on the 2015 approval but in reliance on the 2017 approval and the anticipated 2018 approval. The prejudice it may suffer is from the setting aside of the 2017 and 2018 approvals, not the setting aside of the 2015 approval.

[143] I have thus come to the conclusion that the applicants’ delay should, in the interests of justice, be condoned. The present case is distinguishable from Harrison on multiple grounds. In that case the late point (a street setback issue) introduced a completely new ground of review. It was the only point on which a review might have succeeded, since all the other grounds failed. The SCA took into account that there had been no obvious flouting of applicable restrictions by the owner. From the applicants’ perspective, the late ground was viewed as opportunistic, because the street setback had never been a real concern. Here, by contrast, there are multiple solid grounds of review, and those which were readily apparent were raised, albeit in the context of the rider plans, from the outset.

[144] I should mention, in conclusion on delay, that KFT’s counsel indicated a desire to cross-examine the applicants’ deponents. On my understanding, this was with a view to showing that by mid-2018 Messrs Buhrmann, Shiel and Willemse had knowledge of the 2015 plans and of their approval. I have accepted that they indeed had such knowledge, and I have concluded that Spring Lights should be treated on substantially the same basis. To the extent that the applicants’ explanation for delay is that they only saw the approved 2015 plans when they saw the rule 53 record, I have rejected that explanation.

Conclusion

[145] It follows that the review of the approvals of the three sets of plans succeeds and such approvals will be set aside:

(a) in the case of the 2015 plans, on grounds 1, 2 and 5;

(b) in the case of the 2017 plans, on grounds 2, 3, 4 and 6;

(c) in the case of the 2018 plans, on grounds 2, 3, 4, 5, 7 and 9  

I record that ground 6 in relation to the 2017 plans and the identical ground 7 in relation to the 2018 plans (windows in boundary walls) should, in terms of Harrison, be regarded as in substance a ground of review in relation to the 2015 plans, where this feature already existed.

[146] The applicants are also entitled to a declaration that the building as actually constructed violates the 2018 plans in the respects identified as complaints 2 – 5 of the judgment, since such deviations are a violation of the Building Act and its regulations. I am not prepared to give a general and unqualified declaration of the kind sought in para 3 of the amended notice of motion. The respects in which the approved plans are inconsistent with the DMS and title deed restrictions is apparent from my judgment, and it is obvious that a building constructed in accordance with the approved plans will thus be inconsistent with the DMS in these respects.

[147] Paras 4 and 5 of the amended notice of motion seek a demolition order and ancillary relief. The applicants’ counsel proposed that I should postpone these prayers sine die to allow KFT to consider its options for regularising matters. I shall follow this course, and it seems to me in the circumstances that the interim interdict in para 7 of the agreed order of 17 July 2019 should remain in force until Part B is fully and finally determined.

[148] Para 7 of the amended notice of motion is obscure but appears to be concerned with the period following demolition, and thus likewise will be postponed. In any event, I do not believe is necessary for the court to interdict the occupation of the building until an occupation certificate has been issued in terms of s 14(1)(a) of the Building Act. It is clear from the Building Act that a building may not be occupied without an occupation certificate, and KFT has not threatened to act in violation of the law in this respect.

[149] The applicants and the CCT have agreed that the latter will pay the applicants’ costs of the review up to 17 June 2020, including the costs of two counsel, and that this should be recorded in my order. This agreement relates to both Part A and Part B.

[150] KFT’s counsel submitted that, regardless of the outcome of the review, the applicants should be ordered to pay the Part A costs, because they delayed in arresting what they alleged to be unlawful building work, and then subjected KFT to unreasonably short time periods. On KFT’s version, the construction work was almost complete by the time the applicants moved. Although the applicants, in their Part A replying affidavit, take issue with the claim that the building was 99% complete, it nevertheless seems to be true that by 1 July 2019 all the features about which the applicants had been complaining were in place. It was mainly interior finishing that had to be done.

[151] There is merit in KFT’s submissions. The applicants had been complaining about unlawful building work for many months. There was a threat of court proceedings as early as 7 February 2019. They only issued letters of demand on 14 June 2019, requiring a same-day response. KFT’s attorneys replied fully on 17 June, stating inter alia that no structural work was taking place, that only internal finishing work such as painting and carpentry was being done, and furnishing an undertaking that no structural work would commence (ie resume) while KFT was engaging with the CCT. If the applicants thought that this undertaking was insufficient, they did not seek to clarify it. Instead they waited until 1 July to launch their application, which they set down for 16 July and served on the early afternoon of 2 July, with a timetable which gave KFT only four clear court days (from date of service) to file opposing papers. A supplementary founding affidavit containing Mr Stander’s survey results was served on 10 July.

[152] The applicants attempted to justify urgent action on the basis that the construction of a wooden deck on top of the roof was underway. However, according to KFT the deck was installed and completed in April 2019. The activity which the applicants observed on 14 June 2019 was the boxing-in of the roof hatch in anticipation of a large winter storm.

[153] In the circumstances, I think it was unreasonable for the applicants to seek urgent interdictory relief. Such an application may well have been justified some months previously, but by 1 July 2019 the proverbial horse had bolted, and the undertaking of 17 June should substantially have assuaged the applicants’ concerns. Even if proceedings were required to secure greater provisional relief than KFT was offering, a more reasonable time period may have enabled this to be achieved without KFT having to incur the costs of engaging counsel for the day. I thus consider that the applicant should pay the costs of KFT’s preliminary affidavit of 15 July 2019 and of the appearance on 16 July 2019.

[154] In regard to the Part B proceedings, KFT’s counsel argued that their client was unnecessarily required to engage counsel for 17 June 2020 because the applicants were alleging that there was an enforceable settlement entitling them to obtain a review of the three sets of plans and were intent on asking the judge for an order to this effect with a punitive costs order. I do not intend to burden an already long judgment by tracing the correspondence and affidavits bearing on this question. Although I can understand the frustration felt by the applicants and their attorneys, and although KFT may have been prevaricating, I am satisfied that no enforceable settlement existed which justified the taking of an order on the terms proposed by the applicants at that time.

[155] The stance which the applicants adopted required KFT to engage counsel for the day. KFT’s attorneys had previously proposed that the matter should be postponed to the next available date on the semi-urgent roll. In the event, on 17 June 2020 the applicants abandoned their insistence on the draft order, and all the Part B relief was postponed. I shall thus order the applicants to pay KFT’s costs of the day.

[156] For the rest, there is no reason why costs should not follow the result. KFT’s liability for costs in the review will, in respect of the period up to 17 June 2020, be joint and several with the CCT. The applicants sought a punitive costs order on the basis that KFT had shown itself to be a serial transgressor. I am not inclined to give such an order. KFT will already suffer a heavy blow with the setting aside of the approvals. Moreover, there has been nothing improper in its (limited) opposition to the application.

[157] I make the following order:

(a) In terms of s 9(2) of the Promotion of Administrative Justice Act 3 of 2000, the applicants’ non-compliance with the 180-day period for launching the review proceedings specified in s 7(1) of the said Act is condoned and the said period is extended so as to allow the bringing of the review application on 1 July 2019 and the amplification of the review on 31 January 2020.

(b) The following decisions by the second respondent concerning the building plans submitted by the first respondent from time to time in respect of Erf 196 Camps Bay, commonly known as 5 The Meadway, Camps Bay, are reviewed and set aside:

(i) the approval of building plans on 24 November 2015 under file number 70243013, save for the approval of such excavations as would have to precede the construction of the basement reflected in the said plans;

(ii) the approval of building plans on 18 October 2017 under file number 70362051;

(iii) the approval of building plans on 11 December 2018 under file number 70422662.

(c) It is declared that the building actually erected on the said Erf 196 Camps Bay is in conflict with the National Building Regulations and Building Standards Act 103 of 1977 in that it has deviated from the most recently approved plans in the respects identified in this judgment as complaints 2 – 5.

(d) The determination of the relief claimed in paras 4-6 of the amended notice of motion dated 31 January 2020 is postponed sine die.

(e) Pending the final determination of the said paras 4-6, the interim interdict contained in para 7 of the order of 17 July 2019, and the recordal in para 8 thereof, shall remain operative.

(f) In accordance with an agreement between the applicants and the second respondent, the latter is ordered to pay the applicants’ costs until and including 17 June 2020, such order relating to both Part A and Part B of the notice of motion and amended notice of motion, and to include the costs of two counsel.

(g) The applicants, jointly and severally, must pay the costs of the first respondent relating to the latter’s preliminary opposing affidavit of 15 July 2019 and of the appearance on 16 July 2019, including the costs of two counsel if engaged.

(h) The applicants, jointly and severally, must pay the first respondent’s costs in respect of the appearance on 17 June 2020, including the costs of two counsel if engaged.

(i) Save as set out in (g) and (h) above, the first respondent must pay the applicants’ costs in the review (Part B of the original notice of motion dated 1 July 2019 and the whole of the amended notice of motion dated 31 January 2020), including the costs of two counsel, such liability to be joint and several with that of the second respondent up to and including 17 June 2020.

 

____________________

O L Rogers

Judge of the High Court

Western Cape Division

APPEARANCES

For Applicant	D F Irish SC (with him D M Lubbe)
	Instructed by
	Van Rensburg & Co
	127 Main Road
	Bergvliet
For Respondent	G Elliot SC (with him S Rapaport)
	Instructed by
	Reid Attorneys Inc
	47 Belvedere Road
	Claremont

 

---