---

CASE NO 537/92

IN THE SUPREME COURT OF SOUTH AFRICA (APPELLATE DIVISION)

In the matter between:

ARTONE PRESS (PTY) LTD Plaintiff - Appellant
and
HADDONS TRADING (PTY) LTD Defendant - Respondent

CORAM: BOTHA, SMALBERGER JJA et MAHOMED AJA

DATE OF HEARING: 8 MARCH 1994

DATE OF JUDGMENT: 29 MARCH 1994

JUDGMENT

MAHOMED, AJA

2

The appellant was the plaintiff in the Court a

quo in which it claimed damages in the sum of R330 757,
said to have been suffered by the appellant in the course
of certain repairs effected by the respondent's
employees to a certain Miller printing machine which the
appellant operates in its business. It was common cause
on the pleadings that an oral agreement was concluded by

the appellant and the respondent on or about the 1st or

2nd of August 1986, in terms whereof the appellant

engaged the defendant to effect repairs to this machine.

At a pre-trial conference between the legal

representatives of the parties damages were agreed in the

sum of R327 370.58 and it was further agreed that the

issue which fell to be determined by the Court was

whether or not these damages were caused by

"the defendant breaching the agreement by failing to exercise due skill and diligence or negligently causing damage to the printing machine; alternatively

3

the defendant's employees, acting in the course and scope of their employment, negligently damaging the printing machine".

The pre-trial minute records the plaintiff's belief that "its claim against the defendant is based on contract and that it need not concern itself with the delictual claim" which was pleaded in the alternative. Nothing, however, turns on the characterization of the appellant's cause of action because it was common cause on the pleadings that one of the terms of the agreement between the parties was that in effecting the repairs to the appellant's printing machine the respondent would not negligently cause damage to that machine. The real issue which had to be determined was whether the respondent's employees were indeed negligent.

The ground of negligence relied upon by the appellant in the pleadings was that in carrying out the relevant repairs to the machine the respondent had caused

4
damage to the machine by leaving a certain Allen key "so placed as to enable it to enter the printing machine and cause damage thereto". In terms of further particulars furnished by the appellant, it was alleged that the appellant believed that the Allen key "was placed in the vicinity of the area where the respondent's employees were carrying out work on the printing machine, namely where the paper feeds into the press". The appellant pleaded that it was unable to say who owned this Allen key but that it was a key which was at the material time used and under the control of one or both of the employees of the respondent who were authorized by it to effect the relevant repairs to the machine.

During the course of the trial it became common cause between the parties that certain cylinders in the printing machine of the appellant had indeed been damaged and that the cause of the damage was a 2.5 mm Allen key

5
which had entered into the printing part. This damage

had been done towards the left-hand side of the machine

(facing the direction in which paper flows through it).

The onus was clearly on the appellant to establish on a

balance of probabilities that the respondent's servants,

acting in the course and scope of their employment, had

negligently left this key in a position where it entered

the printing machine and caused damage thereto.

[Pillay v Krishna and another 1946 A D 946 at 952; South

Cape Corporation (Pty) Ltd v Engineering Management

Services (Pty) Ltd 1977 (3) SA 534 (A) at 548]. The

Court a quo held that the appellant had not discharged

that onus and granted absolution from the instance.

Leave to appeal was granted to this Court.

There was no direct testimony from any witness

as to who had negligently put the Allen key into the

machine or at some place which enabled the key to get

6
into the machine. It was contended on behalf of the
appellant however, that from the evidence which was in
fact lead the inference which could and should be made,
was that it was probably the employees of the respondent
who had negligently left the key in a position where it
got inside the machine and caused the damage and that
this act had occurred whilst these servants were acting
in the course of their duties and the scope of their
employment on behalf of the respondent. The respondent
contends on the other hand, that this was not the most
probable inference and that it was more probable
(alternately equally probable) that the key found its way
into the printing machine in consequence of the negligent
conduct of the appellant's own employees or some other
cause not attributable to the respondent's employees.

In support of its case the appellant called two witnesses. The first witness was Mr. Zonele Dlamini
7 ("Dlamini") who was at the relevant time employed as a machine assistant by the appellant. It was Dlamini who discovered the damage to the appellant's machine whilst he was cleaning it. There was the mark of an Allen key in the machine and he reported this to the "machine minder", Mr. Colin Hendricks ("Hendricks") the next morning. A search was then made for the Allen key, which was found under the machine by another employee called Stephen, The key which was found was a 2.5 mm Allen key. Dlamini said that he did not use that key for cleaning the machine on that occasion, but he conceded that he used an Allen key which was much larger, once a month when he took off the drums.

The second and last of the appellant's witnesses was Hendricks. He had been a printer for some fourteen years and had worked for the appellant since 1988. He testified that the appellant had experienced certain problems with its printing machine including a

8
problem with the "side lay" which required repairs. The respondent was engaged for this purpose and its technicians who were sent to the appellant's premises decided to dismantle the "side lay" and take it to the respondent's workshop. It is common cause that these technicians were Willy Fait ("Fait") and Ken Butz ("Butz").

Hendricks said that after the " side lay" had been repaired it was re-installed by the same technicians on the 2nd August 1988. He identified a time sheet which he had signed and which reflected that the job had been done on that date.

Hendricks stated that immediately after the reinstallation of the "side lay" a test run was conducted by feeding blank sheets of paper through the machine and the machine was thereafter used for a printing job for approximately half an hour. On completion thereof there was a clean-up of the machine. In the course of this

9
clean-up the damage caused by the Allen key was discovered and reported to him by Dlamini on the following morning. He confirmed that the Allen key was discovered underneath the machine by Stephen. It was a 2.5 mm black matt Allen key.
Hendricks conceded that he also had a set of Allen keys for use on the machine but he said that the colour of these keys was silver matt and that the smallest key in the set was a 3 mm Allen key. He stated that the cleaning process conducted after the repairs had been completed did not require the use of an Allen key at all and that when there was a general cleaning once a month the cleaners used a 5 or 6 mm Allen key.

On the evidence of Dlamini and Hendricks therefore the key which did the damage to the appellant's printing machine was not a key which belonged to the employees of the appellant and the inference which the appellant contends for was that it must have been left by one of the respondent's employees who had indeed been

10
working on the machine shortly before the damage to the
machine was discovered.

This inference is however resisted on behalf of

the respondent which called both Butz and Fait as well

as the respondent's service manager, Mr. Gabriel Perkins ("Perkins") in support of its case. Butz testified that

he first heard about the problem which the appellant was experiencing with the printing machine on the 1st August 1988. He went to the appellant's premises on that date and the machine minder showed him that the problem concerned the "side-lay" and the "ink-ducts". He checked the "side-lay" and found that it needed a part. On the next day which was the 2nd August he put in a "cam follower" but that did not solve the problem. He then dismantled the "side-lay" and took it to the respondent's workshop. He said that after the necessary repairs had been effected he returned to the premises of the

11

appellant in the company of Fait, for the purposes of reinstalling the "side-lay" into the machine on the 4th August 1988. He spent some four hours on the premises of the appellant on that day and returned again on the 5th August in order to complete the "settings". Fait helped him both on the 4th and 5th of August. In giving this testimony Butz relied on various entries in a "job cost sheet" of the respondent and the relevant "CTM". A CTM is a client's time check showing the amount of time spent on a particular job on a particular day by a workman of the respondent.

Butz conceded that the respondent's employees would have to use Allen keys when re-fitting the "side-lay" and re-setting the "side-lay" cam but he said that these operations would require the use of a 3 mm or a 4 mm or even probably a 6 mm Allen key. A 2.5 mm Allen key would be required only on the perfecting unit and this was not near the work that Butz had performed. He

12

produced a set of Allen keys which he said he had had for some years. He said that he had cut off the shank of each of these Allen keys in order to facilitate access to machines which needed repairs. He did not deny that he was equipped with a 2.5 mm Allen key but he said that he did not use it to effect any repairs to the appellant's machine. He claimed that he had checked all his tools before he left the premises of the appellant upon completing the job and none of his Allen keys were missing. He re-iterated that in any event the Allen key which had been found under the machine by the employees of the appellant was not his because his Allen key had its shank cut off. He also said that Fait would not have required any Allen keys in order to perform his function which was to turn gears in order to get the setting right.

Butz further testified that he and Perkins were called to the premises of the appellant on Monday, 8th

13
August 1988, and shown where an Allen key had gone
through the machine of the appellant and where it had
left a mark on the cylinders. Butz said that they asked
the appellant's representatives to show them the job that
had just been run through the machine and upon checking
this job it was found that approximately fifteen sheets
had been damaged. These fifteen damaged sheets were
found some 6" from the top of a stack of sheets
containing some three to four thousand sheets and the
damaged sheets were torn in the shape of an Allen key.
The inference sought to be drawn on behalf of the
respondent from this evidence was that the Allen key must
have been among the papers which were fed into the
machine and that it found its way into the cylinder for
at least the duration of time while those fifteen sheets
took to go through the machine, before falling out of it.
Fait sought to support the evidence of Butz.

14 He also repeated the version that the final completion of the job took place on the 5th August 1988 and he denied that he either used or had in his possession, at the premises of the appellant, any Allen key. He further supported the version of Butz that the plastic bag containing the Allen keys belonging to Butz would be opened away from the machine so as to avoid the danger of a tool falling into the machine. ("You walk away from the machine to get the tool that you require and walk back towards the machine".)

Perkins was finally called by the respondent to support the testimony of Butz that they had found on the premises of the appellant some fifteen sheets of paper which had been cut in the shape of an Allen key. Perkins supported the theory that the Allen key must have found its way into the machine through this paper which had been fed into the machine and he added that in his experience objects such as "screws, spanners, ball-
15 bearings, paper clips, rags etc." sometimes found their way into machines.
How did the 2.5 mm Allen key get into the printing machine?

From the evidence led at the trial there seem theoretically to be three ways in which the 2.5 mm Allen key could have found its way into the appellant's printing machine.

a)	If the evidence of Butz and Perkins was acceptable, the Allen key might have been in the paper which was fed into the machine by the employees of the appellant and it remained in the machine for a duration sufficient to tear approximately fifteen sheets and to damage the cylinder.
b)	One or other of the appellant's employees left the Allen key concerned in the machine or in a position where it was allowed to get into the machine.
c)	One or other of the workmen employed on behalf of

16 the respondent to repair the machine, left the key in the machine or in a position where it found its way into the machine.
It is necessary to examine the evidence and the probabilities in regard to each of these three possibilities.

The evidence pertaining to the discovery of the damaged sheets and the inference sought to be drawn on behalf of the Respondent in support of the first possibility.

I have several difficulties with this theory. In the first place the evidence to support the finding that the damaged sheets were discovered at all is unimpressive. It rests on the testimony of Perkins and Butz both of whom said that they considered the discovery of the damaged paper to be so significant, as to absolve the respondent from responsibility for the damage caused to the machine. Notwithstanding that fact, however, no
17 mention of any such discovery was ever made in any of the correspondence which preceded the litigation. This is particularly surprising, regard being had to the fact that according to Perkins, the view he took after this discovery was that the respondent could not conceivably be liable for the damage to the machine and the fact that he made this perfectly clear to the director of the respondent upon returning to the respondent's premises. I would have thought that there would have been at least some reference to this discovery and its significance in response to the letter of demand which emanated from the appellant's attorneys or at one or other opportunity which presented itself during the period between the date of this discovery and the commencement of formal litigation. Moreover, if the discovery was in fact made, it must have come as a relief to Butz because it exonerated him from blame and he could therefore have been expected there and then to say that it showed that

18
he was indeed blameless but the evidence of Butz was that none of this was discussed at the premises of the appellant at the time when the discovery was allegedly made. I would have thought that the significance of the discovery would immediately have been pointed out to the representatives of the appellant on the premises but Butz does not support the averment of Perkins that it was. It was also never suggested to Hendricks in cross-examination that Perkins had made any such claim at the time.

When was the discovery made? Both Butz and Perkins suggested that it was on the 8th of August 1988 but for the reasons which 1 will return to later, this could not possibly be true because the repairs to the machine had already been completed on the 2nd August 1988 and it was immediately after that, that the damage was discovered.

I also have some difficulty in accepting that a
19
2.5 mm Allen key would have been imbedded in the paper
which was fed into the machine. It is true that
according to Perkins he was aware of "incidents where
paper delivered from the mill had objects such as paper
clips, pins and stuff like that embedded in the paper, in
the pulp before it, when the paper was manufactured from
the mills", but he does not claim to have ever seen an
Allen key so embedded. Why should it? It is a special
key used by skilled mechanics to gain access to certain
parts of printing machinery, and, on all the evidence it
is regarded as a valuable part of the equipment of such a
mechanic. It is not in the same category as "paper
clips, pins, and stuff like that" which can easily become
mixed with paper. In my view the evidence tendered on
behalf of the respondent with respect to the discovery of
the damaged paper is so thin and unconvincing as to raise
serious doubts as to the credibility and reliability of
the witnesses who deposed to it.

20 The evidence and the probabilities pertaining to the two remaining possibilities.

It follows therefore that the Allen key concerned was negligently left in a position where it got into the machine either by the employees of the respondent while they were working on the machine or at some stage by one or other of the employees of the appellant. Which is more probable? In answering that question one must favour the conclusion which seems to be the more natural or more plausible conclusion from amongst the two competing possibilities. [Govan v Skidmore 1952 (1) SA 732 (N) at 734; Ocean Accident and Guarantee Corporation Ltd. v Koch 1963 (4) SA 147 (A) at 159 (C).]

There is, in this regard, an important objective probability which supports the case for the appellant: Since it was the employees of the respondent who were indeed working on the machine shortly before the

21
damage to the cylinder was discovered, it is more likely
that one or other of these employees left the Allen key
in a position where it got into the machine. The force
and the cogency of that inference must however depend, to
some degree at least, on the weight which can be attached
to the evidence of Butz that the only Allen keys used by
the employees of the respondent on the appellant's
machine were the Allen keys belonging to Butz with their
shanks cut off. If this evidence were accepted it would
destroy the inference that the Allen key which damaged
the machine had been left in or near the machine by the
respondent's employees because it is common cause that
the key which did damage the machine was not a key with
its shank cut off. I have considerable difficulty
however in accepting that evidence. Its cogency does not
rest on any objective probability. It is based simply on
a claim to that effect made by Butz. Neither Butz nor
Fait could however be regarded as reliable witnesses.

22
They both insisted that although they had been instructed to deal with the faults in the appellant's machine on the 1st August 1988, the task of removing the "side-lay", effecting the necessary repairs and finally re-installing the relevant parts in the machine on the premises of the respondent, was completed on the 5th August 1988 after they had spent some hours on those premises on that date. This evidence was patently untrue because it is clear from a letter written by Perkins himself on the 5th August 1988 that the repairs to the machine had been effected before that date and that the complaint of the appellant about the damage done to the machine as well as an inspection by Perkins of the machine itself had all taken place well before the 5th August 1988. Perkins, himself, was driven to concede this difficulty. What is worse is that both Butz and Fait sought to rely on a "job cost sheet" and daily "client's time checks" which were prepared by the service engineers themselves and

23
purported to reflect daily entries of what work had been done on the appellant' s machine from the 1st to the 5th August 1968. These appear to be fictitious entries but Butz and Fait persisted in defending them notwithstanding the fact that the letter from Perkins dated 5th August 1988 clearly destroyed their reliability.

The credibility of Butz as a witness was therefore subject to serious criticism both with respect to his claim that he found damaged sheets of paper on the premises of the respondent and to his claim that he completed the repairs to the machine on behalf of the respondent only on the 5th August 1988. Moreover these claims are not easily explicable on the basis of an honest but mistaken belief in their correctness. His further claim, that the Allen keys which he used had their shanks cut off must therefore be subject to careful scrutiny, unsupported as it is by any obvious objective probability.

24

Thus approached I find the quality of the evidence of Butz to be unimpressive. Perkins who says that he inspected the Allen keys of Butz after the damage to the machine had been reported to the respondent does not say that the shanks of these keys had been cut off. Butz does not support Perkin's' claim that he inspected the Allen keys belonging to Butz at all.

In material respects Butz proved himself to be a stubborn and inflexible witness desperately anxious to avoid the most reasonable concessions, lest they might impact upon his own interests. He insisted that because he "normally" checked his tools after a job had been completed, in order to see whether any were missing, he could not have made a mistake in the instant case. He was equally rigid in his description of the procedure he adopted when effecting repairs; the tool bag would always be away from the machine he was repairing and he would go back and forth from the machine to the tool bag each time

25
he needed a tool or each time he needed to put it back. So meticulously careful was he with his Allen keys that in thirty five years he had never lost one single key. Many of these responses create the impression that the relevant test he consciously or instinctively applied in answering a question was not what the truth was but what might hurt his own interests.

The cumulative effect of all these criticisms, is to satisfy me that the claim made by Butz to the effect that his Allen keys had all their shanks cut off is not sufficiently reliable to detract from the most probable and natural inference that the person or persons who were repairing the machine immediately before the damage was discovered were the persons most likely to have negligently left the Allen key in a position where it damaged the cylinder in the machine. In forming that impression I have had regard to the fact that Butz produced before the trial Court certain Allen keys which

26
apparently had their shanks cut off, but the crucial enquiry which must be made is whether at the time when he effected the relevant repairs on the appellant's machine, Butz used Allen keys which had their shanks cut off or whether he in fact used Allen keys which included the very key which did the damage to the appellant's machine. The answer to that enquiry depends substantially on the quality and credibility of Butz as a witness. For the reasons I have given I am unimpressed by that quality.

Fait who worked with Butz and who was with Butz on the premises of the appellant when the repair to the appellant's machine was completed was also not a very impressive witness. Like Butz, he insisted that these repairs were completed only on the 5th August 1988 although this was patently untrue having regard to the evidence of Perkins and the letter written by him on that date. Fait also stubbornly persisted in defending the fictitious entries on the "job cost sheet" and he was

27
also rigid in refusing to make any concession which could

in his perception have damaged his interests. I am
equally unimpressed by the quality of Fait as a witness,
although the possibility that it was his Allen key which

found its way in the machine of the appellant was more

remote, regard being had to the fact that unlike Butz,

Fait does not admit using any Allen keys on the

appellant's machine.

In the Court a quo Coetzee J found that both

Butz and Fait were "patently unreliable" in claiming that

they completed the repairs on the appellant's machine on

the 5th August 1988, but he concluded that "the evidence

of Butz as to his Allen key cannot be rejected out of

hand". The only reasons he gives for that finding are

expressed by him as follows -

" Although the documentation does not support his version about the repairs and the dates thereof it is undoubted fact that he had no occasion to use a 2.5 mm Allen key at the place where the exhibit entered the machine. Butz was vastly experienced and also on this specific machine. The probabilities are that he knew very

28

well that at the "side-lay" he did not require a 2.5 mm Allen key. All his Allen keys are still in his possession. The plastic pouch exhibits clear signs of having been used for a very long time".

I am not persuaded by this reasoning. The
relevant issue is not whether the evidence of Butz as to
his Allen key could be "rejected out of hand" or not, but
whether on a proper analysis of the evidence the
inference that the key which damaged the appellant's
machine emanated from the respondent's employees was more
probable than the inference that it emanated from a
different source. The claims made by Butz that it was
not his Allen key which damaged the appellant's machine
must be tested against his reliability and credibility as
a witness. That credibility was demonstrably suspect,
for the reasons which I have previously discussed. To
say that "all his Allen keys are still in his possession"
is to assume that the keys produced by him in court were
the keys which he had in his possession and which he

29
had used at the time he effected repairs to the appellant's machine. The correctness of that assumption has to be tested against his credibility as a witness generally. So tested it is unsatisfactory.
Against the probability that the Allen key which damaged the appellant's machine came from one of the workmen of the respondent who had been repairing the machine shortly before the damage was discovered, must be weighed, the probabilities which support the inference that that key came from one of the appellant's employees. The latter inference is certainly a permissible inference on the evidence but is it the more probable inference? I think not.

In the first place, unlike the employees of the respondent, none of the employees of the appellant would have had any reason to be in possession of any Allen key, in approaching the appellant's machine at any time during the period immediately preceding the discovery of the

30
damage. On the uncontradicted evidence of Hendricks and
Dlamini all that transpired after the service engineers
of the respondent's had repaired the machine, was that
the machine was used for a printing job for approximately
half an hour followed by a clean-up of the machine. None
of these operations would have required the use of an
Allen key at all. The only time that an Allen key would
be required by the employees of the appellant would be to
do a "general cleaning" once a month and no such general
cleaning was attempted after the repairs completed by the
respondent's employees. Moreover the Allen key which
would be used on such an occasion would be a 5 or 6 mm
Allen key which is of a substantially different size from
a 2.5 mm Allen key. Indeed on the evidence of Hendricks
the appellant never had access to a 2.5 mm Allen key such
as the one which did the damage to the appellant' s
machine, the smallest Allen key in the set which came
with the machine being a 3 mm Allen key. This set was

31
silver matt in colour in contrast to the black matt colour of the Allen key which damaged the machine.
It is of course possible that all this evidence emanating from Hendricks and from Dlamini is false and that in truth and in fact the appellant's employees always had access to a 2.5 mm Allen key, that this key was indeed black matt in colour, that for some or other reason one or other of the appellant's employees took possession of such an Allen key, that he for some unknown reason approached the appellant's machine whilst having possession of this key, that this took place after the machine had been repaired by the respondent's employees and that in some way or another one or other of these employees eventually put that key in a position where it entered the printing machine. But what are the probabilities in support of such a scenario on the evidence?

The evidence of Hendricks and to some degree

32
the evidence of Dlamini contradicted this scenario. Unlike the case of Butz and Fait, the credibility of Hendricks and Dlamini as witnesses was not damaged by any documentary evidence or any objective probabilities. There is nothing in the inherent circumstances which renders it more probable that the Allen key which damaged the appellant's machine emanated from Hendricks or Dlamini (or some other employee of the appellant), than that it came to find its way into the machine in consequence of some negligent act performed by one or other of the respondent's employees whilst they were busy repairing the machine during the period immediately preceding the discovery of the damage.

Having thus carefully analysed the objective probabilities and the quality of the respective witnesses for the appellant and the respondent who gave evidence at the trial, I am of the view that the appellant did, on a balance of probabilities, establish that it was the

33
negligent act of one or other of the respondent's employees whilst they were repairing the appellant's machine, which caused the Allen key concerned to become embedded into the machine and to cause damage to its cylinder and that the trial Court erred in granting absolution from the instance. It is ordered that -

1. The appeal is upheld with costs.

2. The order made by the Court a quo is set aside and

substituted by the following:

"Judgment is granted in favour of the Plaintiff for

a) payment of the sum of R327 370.58

b) interest thereon at the rate of 18.5% per annum
from the 21st February 1992 to date of payment

c) costs of suit".

I. MAHOMED

ACTING JUDGE OF APPEAL

BOTHA JA ) SMALBERGER JA ) CONCUR

---