---

371/93

/mb

IN THE SUPREME COURT OF SOUTH AFRICA (APPELLATE DIVISION)

In the appeal between

CITY COUNCIL OF JOHANNESBURG APPELLANT

and

TELEVISION & ELECTRICAL DISTRIBUTORS

(PTY) LIMITED FIRST RESPONDENT

and

TEDELEX PROPERTIES (ANDREA ROAD)

(PTY) LIMITED SECOND RESPONDENT

CORAM: VAN HEERDEN, VIVIER, KUMLEBEN, HARMS et

SCOTT JJA HEARD: 26, 27 & 28 AUGUST 1996 DELIVERED: 20 SEPTEMBER 1996

JUDGMENT

KUMLEBEN JA/..

2

KUMLEBEN JA:

In the Witwatersrand Local Division of the Supreme Court

Television and Electrical Distributors (Pty) Ltd ("Tedelex"), the first

respondent, sued the City Council of Johannesburg ("JCC"), the appellant,

for damages. Tedelex Properties (Andrea Road) (Pty) Ltd ("Tedelex'

Properties"), the second respondent, was joined by the JCC as a third

party. The court (Schutz J) ruled that the merits be first adjudicated. After

a lengthy trial they were decided in favour of Tedelex, no liability was

attributed to Tedelex Properties, and a special order of costs was made.

The order in all three respects is now the subject of appeal.

By way of introduction the background facts need be no more than sketched. Those particularly concerned with the outcome of this appeal are conversant with the details. They are comprehensively summarised in the judgment of the court a quo.

In the West Turffontein Valley just south of Johannesburg,

3 Klip River Road ran in a northerly direction to form a T-junction with

Booysens Road, which is one of the main arterial roads leading to the city.

The natural watercourse in the valley flowed in the same direction. By

about 1934 this had been channelled by the construction of an open canal.

The road surface of Booysens Road is at a level higher than its surrounds.

The opening to allow the canal to pass through the embankment, and under

the road at the T-junction, was described in evidence as a bridge and I shall

continue to refer to it as such. In 1978 the JCC decided to replace Klip

River Road with a new high-traffic capacity road, to be known as Klip

River Drive, and at the same time construct a new canal system. A

company, Piovesan S.A. (Pty) Ltd ("Piovesan"), was appointed to construct

the new drainage system and a firm called TAB Construction (Pty) Ltd

("TAB") the road. The work started early in 1980. When completed in

1981 this new drainage system viewed from south to north consisted of a

4

broad, open concrete canal running on the right hand aside of Klip River

Drive and adjacent to it. At a certain point about 230 metres from the bridge it converts to an underground conduit, a culvert, which at a certain stage runs below the surface of the new road. The culvert had to be tapered to tie in with the bridge. Over a length of 15 metres before the bridge, referred to in evidence as the transition section, the culvert was narrowed from 8 metres x 2 metres at the start of this section to 4 metres x 3.5 metres at the bridge. From the bridge the conduit remained an open canal as formerly and proceeds in a northerly direction.

Tedelex Properties are the owners of a building in the area known as Reuven. In September 1981 plans were passed by the JCC for the erection of this building. It was completed in 1983. From that date Tedelex was the lessee. The building is situated south of Booysens Road and east of Klip River Drive close to the inlet of the culvert. The land in

5

that area slopes markedly from east to west and the elevated level of Klip

River Drive impedes the natural flow of water from continuing westwards. On Sunday 7 December 1986 54 mm of rain fell over a period of about 56 minutes. In lay terms this rain storm could be described as a heavy, but not an exceptional, downpour. It was agreed that it had a Recurrence Interval of 13 years, the "RI" being the period expressed in years during which a storm of a particular intensity (measured by the height of precipitation x its duration) is likely to recur. The new canal system, by virtue of its design and construction, had a drainage capacity, also referred to as a discharge range, of no more than 62,5 cubic metres per second (62,5 cumecs). This proved inadequate to contain the storm water. It overflowed the system at a point where the open canal converts to a culvert, that is, at the inlet to the culvert section. Turbulent water forcibly entered the basement of the Tedelex property by breaking an outer concrete panel wall

6

and buckling the steel roller door of the basement. Thus, so it is alleged, property belonging to Tedelex stored in the basement came to be damaged.

The incapacity of the new system to cope with the storm water, and hence the flooding, was primarily due to the constriction of the water flow in the transition section. A subsidiary cause was that at the bend in the canal just before it converts to a culvert, close to the Tedelex property, the retaining wall on the outer side of the curve was not high enough to contain the flood water entering the culvert. Had the old system with its open canal still existed, any overflow would have gradually dammed up against Booysens Road and Klip River Drive and, unless of exceptional proportions, would not have reached the Tedelex property.

The claim as pleaded is a delictual one based on the lex Aquilia for damages with reliance upon negligence on the part of the JCC. Wrongfulness as an element of this cause of action was not expressly

7 alleged. It was, one presumes, taken to be inferred from the facts pleaded.

(Cf. Cornation Brick (Pty) Ltd v Strachan Constraction Company (Pty)

Ltd 1982(4) SA 371 (D) 378 D - F; Lillicrap, Wassenaar and Partners

Pilkinton Brother S A (Pty) Ltd 1985(1) SA 475 (A) 497 B - C.) The

main defence pleaded is a denial of negligence without reference to a defence of justification on the ground that the conduct complained of was authorised by statute. The issues as pleaded were canvassed in evidence. However, during argument this question of strict liability was raised not -one notes - as a rather belated alternative cause of action but in answer to certain defences raised and in reply to the JCCs alternative prayer for an apportionment of damages. This led to questions of strict liability being discussed at some length in the judgment of the court a quo

Any interference and increase in the natural flow of water over another's property is Prima facie an infringement of the right of undisturbed

8

possession which on the face of it attracts liability. Apart from express

authorisation, an incursion may be implicitly authorised by statute but only to the extent necessary for the clue exercise of the right conferred or duty imposed. In New Heriot Gold Mining Company Limited v Union

Government (Minister of Railways and Harbour) 1916 AD 415, the facts

of which closely resemble those in this case, the legal principle in this

regard is thus stated by Innes CJ at 421:

"The general rule that Statutory authority to construct a work excuses from liability for damage thereby caused to third persons is subject to the proviso that the work be not negligently executed or maintained. And as pointed out in Halliwell v Johannesburg Municipality (1912, A.D., at p. 669) if such damages could have been avoided by measures reasonably practicable, it would be negligence not to have taken the necessary steps."

Implicit in the above is that the "negligence" may be of two kinds: a

failure to select reasonably practicable measures for the exercise of the

right; and fault in the implementation of the measures chosen. In

9

1957(1) SA 312(A) 323 C - D Schreiner JA in reference to the first form

of "negligence" observed that

"in deciding what measures are reasonably practicable regard must be had to the total requirements and resources of the local authority and not merely to the means of providing protection to an individual landowner."

See too Johannesburg Municipality v African Realty Trust Ltd 1927 AD

163 at 177 and Diepsloot Residents' and Landowners' Association and

Another v Administrator, Transvaal 1994(3) SA 336(A) 346 F - H. In

Simon's Town Municipality V Dews and and Another 1993(1) SA 191 (A) the

limits of statutory immunity were considered in a somewhat different context. This court in that case was concerned with the interpretation of s 87 of the Forest Act, 122 of 1984. One of the two requirements for legal sanction of an otherwise unlawful act postulated in that section was that the act in question must have been done under a power or duty under the Act.

10

(It was referred to as "requirement (b)" in the judgment.) This requirement

was thus the counterpart of the statutory exemption with which we sure now concerned. In the course of discussing it the present Chief Justice said at

196 B - E:

"[W]here the statute does authorise interference with the rights of others, the person or authority vested with the power is under a duty, when exercising the power, to use due care and to take all reasonable precautions to avoid or minimise injury to others. Failure to carry out this duty has been described as 'negligence', but, as pointed out by Prof J C van der Walt in Joubert (ed) Law of South Africa vol 8 para 30, in this context the word is used in a special sense; and

'(t)he presence of 'negligence' in this special sense in the exercise of a statutory power is, however, a conclusive indication that the defendant has exceeded the bounds of his authority and has therefore acted wrongfully.'

See also Neethling, Potgieter and Visser The Law of Delict at 91-2; Van der Merwe en Olivier Die Onregmatige Daad in die Suid-Afrikaanse Reg 6th ed at 105-6; Boberg The Law of Delict vol 1 at 771-3. In my view, these writers all correctly state that jurisprudentially the consequences of the repository of the statutory power having exercised it without due care and without having taken reasonable precautions to avoid or minimise injury to others, are that

11

the repository must be taken to have exceeded the limits of his authority and accordingly to have acted unlawfully."

And it was pointed out that:

"The person sought to be held liable must show that he acted within the authority conferred by the power in question. It necessarily follows that if, owing to a failure to exercise due care or to take reasonable precautions, he exceeded the power and acted without authority, he will be unable to establish requirement (b) and his reliance on s 87 must fail." (196 I - 197A).

What plainly emerges from these decisions is that there are at

least two requirements to be satisfied before statutory immunity for harm

caused can be successfully claimed: one, that reasonably practicable

measures must have been decided upon for the carrying out of the

necessary works, and, another, that the work must have been executed

without negligence. To avoid any problems of nomenclature, and the use

of "negligence" in more than one sense, I shall refer to them as

requirements (a) and (b) respectively. A simple example will serve to

12 demonstrate that the latter is as significant as the former. If practical

considerations justified the construction of an elevated aqueduct of certain

dimensions, which in certain circumstances would or might have caused

harm to adjoining properties, but as a result of negligent construction its

walls collapse, any reliance on statutory authority as an excuse for the

resultant damage would be clearly misplaced. In short, statutory authority

that impliedly, and within limits, permits an infringement of rights does not

sanction negligent conduct causing harm.

On this question of statutory immunity there was some discussion on where the burden or burdens of proof should rest when this defence is pleaded or arises. In the course of the debate there was reference to Germiston City Council v Chubb and Sons Lock and Safe Co (SA) (Pty) Ltd (supra), Johannesburg Municipality v African Realty Trust Ltd (supra), Johannesburg City Council v Vucinovich 1940 AD 365 and

13 Bloemfontein Town Council v Richer 1938 AD 195. In the last-mentioned

decision at page 231 Stratford JA said in reference to the onus of proof as

regards requirement (a):

"It is for the defendant to prove that in whatever way the power is exercised there must result some interference (of the nature complained of) with the common law rights of others. This being proved, the onus is thus upon the plaintiff to prove that by the adoption of certain precautions (reasonably practicable) or by the adoption of another method (also reasonably practicable) to achieve the purpose of the power the extent of the interference will be lessened - not entirely avoided, for, if the defendant has discharged his onus, avoidance is impossible."

I find it unnecessary to comment on this approach or to express a view on

the discussion of this subject in the other decisions cited since I am

prepared to assume in favour of the JCC that the onus was upon Tedelex

to prove that the bounds of the statutory authority were transgressed, that

is, to prove that one or both of the requirements for immunity were not

satisfied. This conforms to the general rule that it is for the plaintiff to

14 prove inter alia wrongfulness in order to succeed in an action under the lex

Aquilia.

Reverting to the cause of action pleaded, Mr Osborn, who

appeared throughout for the JCC, submitted that Tedelex had failed to prove

negligence in its accepted delictual sense as formulated in Kruger v Coetzee

1966(2) SA 428 (A) 430 E - F. As I have said, this was the main defence

in the plea.

Mr Green was at the relevant time the design engineer. He

was not called as a witness. He planned the new canal system to

accommodate a flood R I of at least 1 : 50 years and a peak discharge

range of some 130 cumecs. It was common cause that to construct the

new canal system with this intended drainage capacity would have been the

reasonable and appropriate thing to have done having regard to the locality

and the interests of property owners in the vicinity. But the discharge

15 range of the system actually installed was between 52,5 and 62,5 cumecs.

This was because the design of the new canal made no provision for any

alteration and enlargement of the bridge. This fault arose from Green failing

to take the hydraulic effect and increased velocity of the storm water into

account when designing the culvert which replaced the open canal: he

overlooked the energy effect of the closed culvert. Mr Brooker, one of the

JCC's expert witnesses, conceded in cross-examination that this was "an

inexcusable mistake".

The concern that the drainage capacity was inadequate with

potentially adverse consequences was raised by officials of the JCC. On

12 June 1980 a minute was written by Mr Taitz, the Resident Contract

Engineer of the JCC, for Green's attention (Ex E62). With reference to

his, Green's, design it reads as follows:

"The downstream end of Piovesan's job is at the bridge below Booysens Rd. This bridge is apparently undersized and has flooded

16

several times in the past. Piovesan ties into this bridge, ie no alteration to the bridge takes place.

I spoke to I Green of Design branch today & he agrees that the bridge is undersized. Mr Green stated that in the design stage he approached Mr Roxburgh of RP &TE about this he Mr Roxburgh wanted the bridge left as it is.

If any alterations are to be made I feel they should be made now, for
two reasons:

1. Piovesan are on site.

2. A major intersection is to be built at this point by TAB Construction.

Should we carry on as per drawing ie tie into the bridge, or should we ask for the necessary money & design for alterations to the Booysens Rd. bridge."

This was followed by a further minute of 15/18 July 1980 (Exh E 63)

whilst the work was in progress. In it Mr Grant, the Chief Roads Engineer,

points out that:

"The downstream section of the culvert which is being constructed under this contract ends at the bridge crossing below Booysens Road. It appears that the stormwater drainage capacity of this bridge is inadequate for the anticipated flows.

17

If any alterations to this bridge are contemplated in order to accommodate the flows these should be carried out at this stage as the culvert contractor is presently on site and the road contract which includes the reconstruction of the intersection at the bridge referred to above has also commenced.

Will you please inform this Branch as soon as possible of your decision in this connection in order to avoid delays at a later stage."

The third minute dated 28 July 1980 (Ex E64), countersigned by inter alios

Green, was in similar vein. In its opening paragraph it is again stressed

that:

"The bridge crossing below Booysens Road is at present inadequate to pass the anticipated flows. This is mainly due to the fact that when the depth of flow in the bridge approaches 2,15m, the backwater effects and entrance losses cause the canal upstream to overflow into Bawden Park."

(In each of these minutes I have emphasised.)

The final decision not reconstruct the bridge was taken by Grant, as head

of the engineering section of the JCC.

On the evidence it is thus clear that the JCC, through the

18

employees responsible, was negligent in more than one respect. First, a

serious design mistake was made and, second, it was followed by a failure to remedy the defect or at least investigate the matter properly after the potential danger of not altering the bridge was drawn to the attention of the officials concerned.

Mr Osborn conceded that this amounted to negligence but it was, as he submitted, "negligence in the air" since it was not reasonably foreseeable that these mistakes could result in any damage. This submission was principally based on the fact that at the time of the decision not to enlarge the bridge the Tedelex building had not yet been erected and the area any flood water might have inundated was vacant or substantially vacant land. I do not agree with this submission. When the new system was designed and installed it was known and envisaged that the land adjacent to the canal and culvert was destined for commercial and other

19 development. The installation of the new system was not a short term

project but was plainly intended to serve its purpose after buildings stood

in its vicinity. It was next submitted that, though damage of some kind

could have been reasonably foreseen, the actual damage caused could not,

viz, damage to valuable goods stored in the basement of a building with a

basement floor level lower - 2 cm lower - than the lip of the canal. But no

such refinement of liability for damage caused is recognised in our law:

"In ons deliktereg word 'n persoon aanspreeklik gehou vir skade indien hy kon voorsien dat skade uit die daad kan ontstaan. Dis nie nodig dat hy inderdaad moet voorsien presies watter soort skade die daad kan veroorsaak of wat die omvang van die skade sal wees nie. Weens die aard van die menslike samelewing word dit geag dat indien hy skade kan voorsien hy ook moet voorsien dat die skade van 'a onverwagte aard en ongewoon kan wees."

S v Bennardus 1965(3) 287 (A) 302H. (See too Botes v Van Deventer

1966(3) SA 182 (A) 191 F - G and cf Kruger v Van der Merwe and

Another 1966(2) SA 266 (A) 272 F - G.)

20

As part of his submission on foreseeability counsel referred to

the Johannesburg Town Planning Scheme and particularly to clause 43 thereof. It circumscribed the use to which a "basement" as defined may be put in the absence of special permission. He submitted that no such consent to store goods in the basement was granted and for that reason damage to goods placed there was not foreseeable. There was some - at times perplexing - argument as to whether the basement of the Tedelex building was a "basement" as defined and whether clause 43 applied to it. It is, however, not necessary to decide this question since there is a fundamental flaw in the premise on which this; submission is based. The test of foreseeability is to be applied as at the time of the act or omission. At that stage, as one knows, there was no Tedelex building and no one could have had the prescience to know where the building would be erected, whether it would have a basement or, if so, whether permission to

21

store goods in it would have been sought or granted.

The conclusion that the JCC was guilty of negligent conduct is, or should be, the end of the enquiry as far as liability is concerned. It

may be readily assumed that the JCC intended reasonably practicable

measures to avoid flood damage by designing a system with a 130 cumec

capacity. It, however, failed to ensure that the work was properly, and not

negligently, done. The defence of statutory authority therefore fails in

respect of requirement (b) and two of the elements of delictual liability in

terms of lex Aquilia - wrongfulness and fault - have been proved.

However, a great deal of evidence was adduced in an attempt

to show that when the new system was designed and constructed certain

circumstances - financial constraints and other considerations - led to the

decision not to enlarge the bridge. I can understand the relevance of such

evidence if the JCC had in fact at the outset decided not to enlarge the

22

bridge in order to restrict the flow to a range between 52,5 and 62,5

cumecs. Such evidence would then be aimed at showing that requirement (a) relating to statutory authority was complied with. But how such evidence can relate to a result not intended still eludes me. The truth of the matter is that this evidence was adduced in an ex post facto attempt to explain an "inexcusable mistake". One is driven to thus conclusion for the following reasons.

The considerations relied upon were, as tabulated in the written heads of argument, "Financial Constraints"; the "Downstream problem"; the "Dislocation of traffic and services" and "Delay in the execution of the contracts". Grant was the main witness called by the JCC to deal with all of these alleged difficulties. I shall in the first place consider his evidence that financial constraints caused the canal to be constructed with the bridge unaltered or that such constraints contributed to the decision to let things

23

be. I have referred to the minutes, one signed by Grant himself and the

third one written by the Chief Design Engineer, Mr Dick, and addressed to him. These point out, as I have already stressed, that : "This bridge was undersized and has flooded several times in the past"; "the bridge crossing below Booysens Brigde is at present inadequate to pass the anticipated flow"; and that alterations to the bridge should be made promptly while Piovesan was on site and the intersection was still to be built by TAB Construction. In the circumstances one would have expected Grant to have reacted positively: to have taken steps to ascertain the extra costs involved and obtain, or attempt to obtain, the necessary funding. The first minute raises the question: "Should we ask for extra money?" He did not do so, instead he simply decided against the reconstruction of the bridge. Not surprisingly, Grant when giving evidence had difficulty in explaining this. He said that financial constraints existing at the time strongly influenced his

24

decision. But the facts, and his evidence under cross-examination, do not

bear this out. It was agreed at the trial that such costs in 1980 would have amounted to no more than R240 000,00. Grant conceded that if he had known of the problem before the contract with Piovesan had been

concluded or if the construction presented a "serious problem", defined by

him as one which could cause loss of life, the extra amount of money could

have been obtained. The combined estimates of the TAB and Piovesan

contracts was R3 680 000,00. An extra R240 000,00 on the contract as a

whole would not have amounted to an appreciable cost increase. In fact,

according to the JCC witness Mr Papendorff, in the 1980/1981 financial

year the need arose to increase the payment due in that year to TAB

construction from an allocation of Rl 000 000,00 to Rl 960 000,00. This

was done by the curtailment of certain other schemes. There was some

evidence that in July 1980 when the decision was taken not to enlarge the

25

bridge the particular budget for building operations was to an extent

stretched. But even if that were the case, the fact that the bridge was never

at any later stage enlarged confirms that financial constraints were not a

material consideration when the decision was taken. In so far as the real

reason for his decision is ascertainable from Grant's evidence, which was

far from explicit, it was that he did not consider that to leave the bridge

unaltered would present a problem. This emerges from the following

passages in his evidence:

"I was puzzled by the statement [that the bridge was inadequate] because we had rebuilt Booysens Road some time during the 1960s and as part of that process we had to satisfy ourselves that the existing bridge was not creating a problem and the, the engineer responsible for that particular design was one of the most thorough, competent engineers who has ever worked for me. He subsequently became a partner in a very well known firm of consulting engineers, but he, he never came back and said there is a problem with the Booysens Road bridge and he would have discussed this with the local district engineer and the local superintendent for that area."

and

26

"COURT: Well Mr Cohen is saying to you was it only a threat to life that would have led you to find money at that time or was there anything else that might have led you to find money? — Well I was dealing with a history of a bridge that had proved perfectly satisfactory from the time before I even joined the city engineer's department. It was built in 19) .. or improved in 1934. We had investigated it in the 1960's.

MR COHEN : Mr Grant you are only saying that you .. (intervenes) -There was no history of a problem at this bridge.

Yes, you just say you did not believe there was a problem? — No we got a detailed reply and, and the problem, my reading of the reply is that the problem was not serious. They had restored the capacity to almost what it was before.

You did not think there was a serious problem? — No."

Turning to what was called the "downstream problem", there

was flooding down-stream of the bridge before and after the construction

of the new canal system. It was particularly severe in the area of the

confluence of this canal and another one, the Robinson canal. The JCC

owned property in that vicinity and sustained substantial loss damage as a

27

result of the flooding down-stream in November 1979. It was therefore

fully aware of this problem when the canal system was installed. Yet it was designed for, and but for Green's oversight would have been constructed to cope with, a discharge of 130 cumecs. There is no suggestion that in designing a canal with this capacity it was unmindful of past flooding downstream, that it disregarded the interests of downstream property owners or that it had reason to think that the bridge as planned would aggravate the problem. As a matter of fact Green's mistake may -I stress may - have averted flooding downstream and caused the Tedelex property to be inundated instead. But this in no way supports the proposition that Grant, when deciding against widening the bridge to prevent overflowing, had the interests of down-stream property owners in mind. At the time the widening of the bridge was being considered it was not stated in any of the minutes that this consideration was a reason for the

28

decision to leave the bridge unaltered. There is in fact no mention of this

problem in the minutes.

The last two considerations need not detain one. In argument before us they were not touched upon. I need only mention that the "services" referred to are post office cables, water mains, electric cables and such like conduits which are normally under or adjacent to main roads. Had the enlargement been decided upon when the issue was raised in the minutes, the contractors were on site and such services were in any event disrupted. This is confirmed by what was suggested in regard to the minute of 18 July 1980. I cannot in the circumstances accept that this factor was taken into account when the decision against widening was taken.

In the result one must conclude that the construction of the canal system with an inadequate discharge range was by no means justified and, as I have said, the requirements of wrongfulness and negligence in the

29

delictual action pleaded were proved.

Mr Osborn: in the court a quo raised the question of causation in a somewhat unusual form. Though the design of the new canal system in fact and manifestly caused the flooding of the basement, counsel submitted that had the old system remained in place, such flooding would have in any event occurred with some resultant damage to the property of Tedelex. Schutz J described this argument as the "it would have happened anyway" defence. I shall simply call it the "causation defence".

At the outset something ought to be said about the burden of proof as regards this defence. Counsel submitted that the onus is upon a plaintiff in a claim under the lex Aquilia to forge the necessary causative link between misconduct and damage. This is generally speaking the position. However, when the unlawful and negligent act or omission as a matter of fact caused the damage this causation defence is in the nature of

30

confession and avoidance and the answer to what would have happened if

the new system is "thought away" speculative.

In the New Heriot decision (supra), in which the same

causation defence was put forward, the following comments, which may

aptly be applied in the present case, were made by Innes CJ at page 432:

"Turning now to the contention itself I express no opinion upon the question whether the fact (if it could be proved) that the appellant would have been damaged by the negligence of the Netherlands Company if the station had not been built, would be a good defence to a claim founded upon the negligence of the respondent in building it. Because I am satisfied that the facts necessary to found the contention have not been sufficiently made out. To ascertain with certainty the relevant particulars of the flooding which has occurred is no easy task. But to declare what would have happened under entirely different circumstances would be very much harder, and is impossible upon the evidence before us."

It is likewise unnecessary for me in this case to decide whether this is a

good defence or whether Innes CJ in the above passage had in mind a

defendant bearing the onus in the strict sense (cf. South Cape Corporatio

31 (Pty) Ltd v Engineering Management Service(Pty) Ltd 1977(3) SA 534

(A) 548G) to prove "that it would have happened anyway". What can be

said is that at the very least it was for the JCC, having raised this defence,

to place some reliable evidence before the court indicating that the flooding

would in any event have caused damage to the property in the basement.

As Prosser and Keeton on Torts (5th Ed) state at pages 268 and 269:

"On the issue of the fact of causation, as on other issues essential to the cause of action for negligence, the plaintiff, in general, has the burden of proof. The plaintiff must introduce evidence which affords a reasonable basis for the conclusion that it is more likely than not that the conduct of the defendant was the cause in fact of the result. ... The plaintiff need not negative entirely the possibility that the defendant's conduct was not a cause, and it is enough to introduce evidence from which reasonable persons may conclude that it is more probable that the event was caused by the defendant than that it was not."

In my view, this causation defence in the present case fails on

two scores. First, when determining foreseeability what is to be postulated

is the situation had there been no negligence: in this case had the design

32

mistake not been made. In that event the old system would not have

remained: the new system would have been in place with a discharge range as planned and adequate. There would consequently have been no Gooding of the Tedelex Property. Second, and bearing in mind the duty to adduce evidence when such a defence is raised, had the old system not been replaced when the Tedelex building came to be designed and built, it is as likely as not that a risk of flooding would have been taken into account and a different building designed or precautions taken on the Tedelex property to drain off any anticipated overflow from the open canal.

Although this determines the causation issue in favour of Tedelex, I must discuss the JCC's approach to this question since the evidence in this regard has a bearing on the special order for costs.

In support of the causation defence the JCC sought to establish a peak flow of the flood of the highest possible intensity at the inlet to the

33

culvert where the overflow occurred, the "peak flow" being the maximum

intensity of flow resulting from a particular storm in the catchment area. As this argument was presented by counsel, the higher this flow, the better

the prospects of the causation defence being decided in favour of the JCC.

At the outset of the trial the experts were agreed that the flood RI of this

particular storm fell within the range of 1 : 10 - 20 years, or in terms of

peak flow intensity within the range of 48 cumecs and 80 cumecs. (As

already stated the storm (rainfall) RI of the downpour of 7th December was

agreed upon as 1 : 13. The reasons why a storm RI and a flood RI need not

necessarily correspond are fully explained in the evidence.) During the

course of the trial the JCC attempted to determine a more accurate estimate

of the peak flood intensity. In fact it sought to prove that it was as high as

72,8 cumecs in which case, so it was contended, water would have in any

event entered the basement and caused some damage had the old system

34

been in place. This led to a protracted examination of this issue in the

course of the evidence of the three expert witnesses, all of whom were civil engineers experienced in this Geld. Mr Fourie, called by Tedelex was first

examined on this. He explained his use of certain methods and formulae

to reach his conclusion that the peak flood intensity at the outlet was about 50 cumecs. Mr Brooker, for the JCC, sought to substantiate the degree of intensity relied upon, the 72,8 cumecs, by applying other methods and calculations. By the use of one of the methods employed by Fourie - the Rational Method - the figure Brooker arrived at was 64 cumecs. Mr Annandale, the Tedelex Properties expert witness, on the information available to him accepted a range of between 52 and 67 cumecs. He explained generally, and with reference to Brooker's own calculations of 78,2 cumecs and 64 cumecs, that any attempt at an accurate figure was a fruitless exercise. As he expressed himself "hydrology is too inaccurate a

35

science to make statements about specifics, especially if it is so close to

the threshold value [of 80 cumecs]". All the evidence bears this out and the reason for the lack of anything like unanimity among the experts on this question is not far to seek. The calculation of flood (peak flow) intensity

is a far more complicated exercise than the determination of the intensity

of rainfall. The former involves taking into account a number of variables which are incapable of accurate assessment. These include the nature and topography of the catchment area concerned which influences the rate of run-off; the extent of pre-existing moisture or saturation; and the periods and degree of peak intensity of rainfall during a particular storm. It is therefore hardly surprising that there should be this large variance of opinion and one can fully appreciate why the initial agreement estimated the peak flow within a broad range of 48 to 80 cumecs.

Mr Osborn on the strength of Brooker's evidence submitted that

36

the JCC had proved on a balance of probabilities that the peak flow rate

was 72,8 cumecs. This submission is wholly untenable in the light of what has been said in the previous paragraph. As a matter of fact the evidence,

if anything, proves the contrary.

I turn to consider the plea of contributory negligence.

By March 1981 the new canal system was in place and virtually completed. In September 1981 the plans for the Tedelex building, which made provision for the basement subsequently built, were approved. The building was erected in accordance with the plan. The plea of contributory negligence, broadly stated, is that the risk of flooding ought to have been foreseen when the decision to build was taken. In the circumstances, so it is submitted, a basement should not have been included; or its floor should have been at a higher level in relation to the lip of the canal; or steps should have been taken to divert flood water from

37

its entrance. It is alleged that Tedelex Properties or Tedelex was remiss in

one or more of these respects. But what is important is that negligence on anyone's part must be related to the facts of this particular case, viz, that Tedelex, Tedelex Properties or the latter's architect ought to have foreseen

that the new, sophisticated canal system would not be able to cope with a

rain storm having an RI of no more than 1 in 15 years, which in lay terms no one would describe as an exceptional downpour. The parties agreed, as I have said, that it is reasonable to expect a drainage system in an urban area to cope with a storm having a RI of 1 : 50 years. There is other evidence that in parts of Johannesburg old existing drainage systems have a discharge range to contain a storm with a RI of 1 : 20 years. But it is not suggested, nor could it be suggested, that anyone would have expected a system recently installed to be incapable of coping with a RI 1 : 13 year downpour. This is really all that need be said in answer to the plea of

38

contributory negligence.

Mr Osborn nevertheless argued in the first place that the officials of Tedelex and Tedelex Properties, all layman in the Geld of hydraulics and engineering, ought to have foreseen that excavating the

ground for the erection of the building and including a basement could give

rise to flooding problems given the proximity of the canal. This is

unrealistic. A building owner appoints an architect and instructs him to

design the building he wants built. It is left to the architect to consider the

question of possible flood water levels and, if necessary, to call in other

specialists.

Counsel next submitted that the architect was at fault in that he

should have foreseen the risk of flooding and appointed an engineer, or

better still a hydraulics engineer, to investigate whether the new system had

an adequate discharge capacity: in effect, that he ought to have foreseen

39

that there might have been a latent design error. The answer to this

proposition must surely be "no". An architect too would have been entitled to assume that the newly constructed canal would conform to reasonable urban drainage standards (a storm with a RI 1 : 50) and be able to contain

a rainfall of far less intensity, in fact one having a recurrence interval of

about 4 times the frequency of the accepted norm.

Even if it could be said that the architect's failure to check the drainage capacity of the new system amounted to a negligent omission on his part, the JCC faced the insurmountable hurdle of attributing such negligence to Tedelex. In the exercise of his professional function of taking potential drainage problems into account, and designing accordingly, the architect was plainly operating as an independent contractor. But counsel argued that the work entrusted to the architect was per se dangerous and that on that account there should be an apportionment, Lanrley Fox

40 Building Partnership (Pty) Ltd v De Valence 1991(1) SA 1 (A) sets out the principles to be applied when a third party seeks to hold an employer liable for the negligence of an independent contractor. What was said at 13 A -C would apply equally when contributory negligence on the part of the employer is alleged, viz:

".. that the existence of a duty upon an employer of an independent contractor to take steps to prevent harm ... will depend in each case upon the facts. It would be relevant to consider the nature of the danger; the context in which the danger may arise; the degree of expertise available to the employer and the independent contractor respectively; and the means available to the employer to avert the danger. This list is in no way intended to be comprehensive. It does follow, however, that the duty of an owner of premises such as the present may not be the same as that of the building contractor employed by him to do the work. That question, too, must be answered with due regard to the facts."

In short, the facts must demonstrate that the employer should have realised

that the work was inherently dangerous. To erect this building, supervised

by an architect, near a recently built drainage system could not possibly

41

have been regarded by Tedelex as presenting, realistically speaking, a

danger to it for the reasons already stated.

The plea of contributory negligence cannot be upheld.

During the course of the trial the JCC amended its plea to rely

on what it termed an exclusionary clause. Before the Tedelex building was

erected application was made for the approval of the building plans in terms

of the building by-laws of the JCC, authorised by the National Building

Regulations and Building Standards Act, 103 of 1977. The application

form required the name and address of the owner to be filled in and the

form was required to be signed by the owner. "Owner" is defined in s 1

of the Act as the person in whose name the land is registered on which the

building is to be built. In the application form there is an instruction that

the applicant should "PLEASE READ THE NOTES ON PAGE 2 BEFORE

PREPARING AND SUBMITTING PLANS." These notes appear under a

42

letter forming part of the application, the letter being included in

anticipation of the grant of approval. It reads as follows:

"Sir/Madam,

The plans submitted by you as detailed overleaf have been approved

subject to the conditions stipulated in the notes printed below.

This approval is valid for twelve months from the above date.

B.L. LOFFELL

for City Engineer"

Note 5 of the "NOTES" which follow states that:

"The Council does not accept responsibility for any damage which might be caused to your property by storm water or by infiltration of water from any source through cellar walls."

On the strength of this note it was submitted that Tedelex had

contractually renounced its right to recover delictual damages from the JCC.

This contention was rejected by the court a quo on four grounds: (i) That

by submitting the application form in which this condition features in no

43

way amounted to a contractual forfeiture of rights; (ii) that, if this clause

had any such effect, it was for the JCC to show that the exemption sought in this case falls squarely within the terms of this clause and this it had

failed to do; (iii) that any exclusionary effect the clause might have, could

only relate to the building owner, Tedelex Properties; and (iv) that it was in any event not competent for the JCC, when carrying out the statutory function of approving or disapproving building plans, to impose a contractual term, which had the far-reaching effect of exempting it from liability for negligence.

To my mind, all four of the above answers are sound. I shall restrict my comments to the last two.

As appears from the application form and the letter of approval, it is implicit in the entire procedure for obtaining the necessary consent that it is a matter between the JCC and the prospective building

44

owner as applicant, whether the latter acts personally or through a

representative. Therefore, even if the note were to be elevated to the status of a contract or construed as a waiver, it can have no reference to the rights of a third party, in this case, the tenant of the building owner.

Section 7 of the Act peremptorily enjoins the local authority to grant its approval if it is satisfied that the application in question complies with the requirements of the Act. It manifestly follows that a local authority cannot impose a condition exempting it from liability, the acceptance of which is a pre-requisite to the grant of approval. (Cf Fernwood Estates, Ltd Cape Town Municipal Council 1933 CPD 399 at 403.)

It remains to discuss the special costs award in favour of both Tedelex and Tedelex properties for the time devoted to the unsuccessful causation defence. The court assessed this period to be 12 days of the trial

45 preceded by 2 days of preparation on this issue. This estimate was not challenged by Mr Osborn. THE court accordingly awarded attomey-and-client costs for what was regarded as wasted time.

The reasons for his doing so appear from this passage in the

judgment:

"[I]t seems to nie that the introduction of this defence wholly

inverted the trial and protracted it enormously. As I have shown

above once this defence is out of the way the facts are fairly

straightforward and are largely common cause. The experts reached

a large measure of agreement. But then it was insisted by the JCC

that proof of an even bigger flood than that accepted by Fourie and

Annandale would absolve it .... [T]he ultimate concession by

Brooker, ... which brought the whole structure tumbling down was

gained by asking a few simple, and if I may say so quite obvious

questions. Nor did they come as any surprise. At least in the

evidence of Annandale, and probably before, there was a clear

warning of the difficulty which would face the JCC. These simple

and obvious questions should, in my opinion, have been asked in

consultation before this extended and hard-fought issue was raised in

the trial.

There is more to be said. The JCCs case was presented always on the verge, based on the high figure of 72,8 cu. There was no real

46

attempt to prove that the Witwat system was intrinsically superior. Brooker did not have to be probed very hard to make the concession, the inevitable concession, that hydrology being the imperfect science that it is, one should not fall into the trap, as he put it, of taking one figure, but should have regard to a range of figures, and to the opinions of other competent experts. Once all this was put to him he conceded, quite apart from the question of impedances which had led to his main concession, that it was not more probable than not that the flood would have reached an elevation of 1 688 m. Coupled to this is the fact that he did not, in his own enquiries, address the important question of the level of incipient flooding. I think that there is substance in Mr Loxton's submission that if he had done so, and he no doubt realised it, the enormous edifice of hydrological data, as Mr Loxton described it, would come down." (I emphasise.)

In sum, it appears from what has been said that the learned judge concluded

that the JCC embarked upon this ambitious excursion when it ought to have

been appreciated that a flood intensity as high as 72,8 cumecs at the outlet

could never be proved and, were this to be proved, there was no realistic

prospect of establishing a water level in the basement likely to have caused

damage to the contents. Both considerations, but particularly the second

47

one, weighed with the court a quo. The judgment is correct in saying that

Annandale's evidence, which from the record seems to have been reliable and impressive, ought to have served as a warning.

The JCC nevertheless persisted in its attempt to prove an exact peak Good intensity. At the risk of some repetition, the reasons why the JCC ought to have realised that a rate of 72,8 cumecs could not be established are: (i) The experts agreed at the outset on a range, 48 to 80 cumecs. They did so, one presumes, because it was accepted all round that a precise rate could not be determined; (ii) Brooker's own estimate was 64 or 72,8 according to the two different methods he used - incidentally methods primarily intended for an estimation of hypothetical peak flow and not for a determination in respect of an actual flood; (iii) there are a number of components of any formula or method of calculation of peak flow which are subjective and incapable of accurate determination. The

48

failure to show a Good intensity of 72,8 cumecs, even as a reasonable

possibility, without more put paid to this causation defence. As regards the

second issue, Brooker's own evidence failed to indicate with any degree of

certainty that if the new system had not been installed a significant amount

of flood water would have reached the basement. After it was shown that

Brooker did not take into account any drains which may have been within

the basement to disperse water, the following was the exchange question

and answer:

"MR COHEN: Let us go on with the assumption. You have these roller shutter doors or photographs of them? - Yes I have, I have seen them. I was at the inspection.

The water would have been able to get through the edges is that right? The edge between the roller shutter door and the wall. - The edges and the bottom of the door would not seal completely no.

Would not seal completely? Now how do you see in your mind's eye the water entering the building at that, at that point? It would not spurt him, would it? — No it could not spurt him of course.

49

It would seep in? — Yes.

And it would be seeping in for at the most something like ten minutes? — That is what the calculations show yes.

But in fact the operative period would be six minutes, that is from ten centimetres or more? — Yes it would start at the bottom but there would be very little flow until the head rose.

Now you have seen how large that warehouse floor is? — Correct.

How much water do you say at your best guess, would get in? How much of that floor would even be wet, water seeping in through those roller shutter doors in a six or ten minute period? — My lord I have not done a calculation but I would say very little.

That is right. And would you say Mr Brooker, at your best guess as an expert, that for that period of ten minutes, water seeping in would not even cover the entire floor? — Yes I would, I would agree with that yes. COURT: That is to any depth at all?

MR COHEN : Yes, yes. - There would be parts, I would agree that there would be parts of the floor that would not get wet at all."

It was not disputed that in appropriate circumstances the

conduct of a litigant may be adjudged "vexatious" within the extended

50

meaning that has been placed upon this term in a number of decisions, that

is, when such conduct has resulted in "unnecessary trouble and expense which the other side ought not to bear." (In re Alluvial Creek, 1929 CPD 532 at 535.) See also Phase Electric(Pty) Ltd v Zinman's Electronic Sale (Pty) Ltd 1973(3) SA Ltd 1973(3) SA 914 (W) at 918H-919B, and Hyperchemicasls International (Pty) Ltd and Another v Maybaker Agrichem (Pty) Ltd and Another 1992(1) SA 89 (W) at 101G - 102D. Naturally one must guard against censuring a party by way of a special costs order when with the benefit of hindsight a course of action taken by a litigant turns out to have been a lost cause. That said, in the present case I am certainly not persuaded that the court failed to exercise a proper judicial discretion in deciding upon a punitive costs award.

In the result the appeal is dismissed with costs, these to include the cost of two counsel representing the first respondent, Tedelex, and two

51

counsel representing the second respondent, Tedelex Properties.

M E KUMLEBEN

JUDGE OF APPEAL

VAN HEERDEN JA

VIVIER JA - Concur

HARMS JA

SCOTT JA

---