South Africa: Western Cape High Court, Cape Town

You are here: SAFLII >> Databases >> South Africa: Western Cape High Court, Cape Town >> 2024 >> [2024] ZAWCHC 439

| Noteup | LawCite

Walser Wine Projects (Pty) Ltd v Wine Machinery Group (Pty) Ltd (18693/2020) [2024] ZAWCHC 439 (6 December 2024)

Download original files

PDF format

RTF format

SAFLII Note: Page 32 [para 87.1] image is not available in html and rtf versions, please refer to the PDF attachment for images.

IN THE HIGH COURT OF SOUTH AFRICA

(WESTERN CAPE DIVISION, CAPE TOWN)

Case No: 18693/2020

In the matter between:

WALSER WINE PROJECTS (PTY) LTD Plaintiff

and

WINE MACHINERY GROUP (PTY) LTD Defendant

Heard: 13 to 16, 20 to 22, 27 and 28 May 2024

6, 12, 13 and 19 to 22 August 2024

3 to 6 September 2024

5, 26 and 29 November 2024

Judgment: 5 December 2024

JUDGMENT

Handed down by email to the parties

on 5 December 2024, that date being the date of the Judgment

KANTOR, AJ:

1. The plaintiff conducts the business of a winemaker. Its sole director and general manager is Pieter Walser (“Walser”) who has been making wine for over 20 years.

2. In 2019, the plaintiff purchased from the defendant a Zambelli six head filler which is equipment used to bottle wine (“the Filler”). The plaintiff avers that subsequent to it using the Filler, certain of the wines bottled using it (“the Wines”) were spoiled. The plaintiff contends that:

2.1. This was caused by the Filler being corroded as a result of which iron was released into the wine.

2.2. The plaintiff pleaded in paragraph 29.2 of the Particulars of Claim as follows: “The iron particles acted as an oxidation catalyst which, in turn, altered the qualities, characteristics and chemical composition of the wines.”

2.3. “As a result of the aforegoing, the plaintiff suffered damages in the amount of R31,814,035.99 which amount comprises the following:

30.1 R3,923,232.78 being the difference between the fair and reasonable value of the wine withdrawn from the market, in the amount of R4,175,341.28, and the saving of costs which the plaintiff would have but did not incur in the process of selling the wines in the amount of R252,108.50 …

30.2 R27,769,603.21 being the loss of profit the plaintiff will suffer in respect of the wine harvests in 2021 to 2030 …

30.3 R121,200.00 being the actual costs incurred by the plaintiff in having to rent bins and storage space it would not have incurred but for the defendant’s breaches pleaded above and the actual costs to have certain of the wines destroyed.”

3. In the pleadings, the plaintiff relies on contract and delict. In argument the plaintiff recorded that it “… does not persist with the delictual basis for its claim.”

4. The defendant pleaded in paragraph 7.3 of its Plea that “… it was an implied term of the contract of sale that the item sold would be free of latent defects and fit for its objectively intended purpose.”

5. With regard to the words “fit for its objectively intended purpose” as pleaded by the defendant in paragraph 7.3 of its plea, the plaintiff requested the defendant to provide the following further particulars:

“(I)s it the defendant’s case that the objectively intended purpose was simply to fill bottles with the wine regardless of whether the filler affected the characteristics of the wine or is it the defendant’s case that the objectively intended purpose was to fill bottles with wine without affecting the characteristics of the wine?”

6. The defendant responded as follows:

“The defendant’s case is that it was a tacit term that the item would be free of latent defects and fit for the intended purpose of filling bottles with wine so that, if used correctly it would not by itself spoil the wine. The defendant denies that it was a tacit or implied term that the intended purpose of the filler included the bottling of wine without affecting the characteristics of the wine regardless of the manner in which it was used.”

7. The defendant denies any breach on its part and also that any breach which may be proved would be the cause of the loss claimed by the plaintiff.

8. During oral argument, Mr Coetzee, who appeared with Mr Potgieter for the plaintiff, formally recorded the following:

8.1. The claim for damages is limited to the amount of R3 923 232.78 pleaded in paragraph 30.1 of the Particulars of Claim for the particular wines it claims were spoilt and for the costs in the amount of R121 200.00 pleaded in paragraph 30.3 of the Particulars of Claim, quoted above.

8.2. The plaintiff abandoned the claim for (future) loss of profit in the amount of R27 769 603.21 pleaded in paragraph 30.2 of the Particulars of Claim, quoted above.

8.3. The effect of this is that paragraph 30.2 of the Particulars of Claim was to be deleted and the amount in prayer a) thereof amended to R4 044 432.78.

8.4. The plaintiff only relied on a claim for damages flowing naturally from the breach of the contract, and not for any special damages as was pleaded in paragraph 31 of the Particulars of Claim.

8.5. The effect of this is that paragraph 31 of the Particulars of Claim was amended by the deletion of the content thereof from “, alternatively”.

8.6. Amended pages of the Particulars of Claim were delivered in the above respects, formally effecting the amendments.

9. The question of the quantification damages was separated out and the matter proceeded on the questions of breach and causation.

10. This spawned a trial of over twenty days, involving seven witnesses, of whom five were experts.

A: Application for certain evidence to be by virtual link

11. At the commencement of the trial the defendant applied for leave to lead the evidence of Dr John Danilewicz (“Danilewicz”) by an audio-visual virtual link. This was opposed and a formal application was brought by the defendant.

12. I granted the application. The reasons follow. They borrow extensively from the comprehensive and very useful written argument prepared by the defendant’s junior counsel, Mr Smuts, who also argued the application (he was led by Mr Duminy in the trial).

13. The basis of opposition may be summarised as follows:

13.1. The evidence with regard to Danilewicz’s personal circumstances, which forms the basis for the application, was hearsay and therefore inadmissible (“the hearsay argument”).

13.2. Danilewicz’s proposed evidence will be inadmissible, because the defendant pleaded that it had no knowledge of how the defective filler caused the plaintiff’s wine to become spoilt and any opinion which Danilewicz advances on this point will contradict this plea (“the pleadings argument”).

13.3. Danilewicz is not a material or an essential witness and his proposed evidence is not stated by the defendant to be material to its case. Moreover, the defendant did not show that Danilewicz will give evidence not already dealt with by its winemaker expert, Mr Van Rensburg (“Van Rensburg”), or that it will not be able to put up a defence without Danilewicz’s evidence (“the materiality argument”).

13.4. Although it may be inconvenient for Danilewicz to testify in person, it is not impossible. On the other hand, Danilewicz will not always be visible to the Court if the plaintiff’s counsel makes use of a separate screen to show documents, videos and photographs and furthermore makes use of a pointer to point to specific cells in spreadsheets, or points on photographs. The Court and the plaintiff’s legal representatives will therefore be inconvenienced by not being able to see Dr Danilewicz during the entire time of his testimony (“the convenience argument”).

13.5. Because the defendant filed the Rule 36(9) notices in respect of Dr Danilewicz with the knowledge that he would not be amenable to testify in person, the defendant is the author of its own misfortune (“the own misfortune argument”).

14. These arguments will be dealt with below, after a brief overview of the legal position, which is governed by section 37C of the Superior Courts Act 10 of 2013 and Uniform Rule 38(9).

15. The general requirements may be summarised as follows:

15.1. Section 37C provides that:

15.1.1. It must appear to the court that making such an order would prevent unreasonable delay, save costs, be convenient, or prevent the likelihood that any person might be prejudiced or harmed if he or she testifies or is present at such proceedings and otherwise be in the interests of justice.

15.1.2. Facilities for testifying in this way must be readily available or obtainable at the court.

15.1.3. The audio-visual facilities that are used by the witness or the court must enable the persons in the courtroom to see, hear and interact with the witness giving evidence and the witness to see, hear and interact with the persons in the courtroom.

15.2. The provisions of Rule 38 require notice of the application, that it “appears convenient or in the interests of justice” to the court to grant the order, and that a draft order be attached to the application setting out the terms of the order sought, including particulars of:

15.2.1. the witness who is to testify;

15.2.2. the address from where he or she will testify; and

15.2.3. the address to where the evidence will be transmitted.

16. Even prior to the Covid-19 era, in 2017, the court in Uramin (incorporated in British Columbia) t/a Areva Resource Southern Africa v Perie 2017 1 SA 236 GJ authorised the use of a video link to procure the evidence of witnesses who were in Dubai and Paris for business and would not attend court in Gauteng. This was done in terms of Rule 39(20),[1] and in deciding to do so, the court employed the same standard as set in Rule 38(9), namely that of convenience and necessity for the purposes of justice. The court remarked that “I have heard a number of such applications and heard evidence in this manner in a number of trials. My experience is that the approach of both South African courts and courts in other jurisdictions must continuously try to be relevant to and keep pace with rapidly changing demands placed upon judicial practice.”[2] The court found the use of the video link to have been appropriate and satisfactory in the circumstances.[3]

17. The advent of the Covid-19 pandemic in March 2020 introduced a paradigm shift in the use of audio-visual links at Court hearings, which became the standard for hearings for the duration of the pandemic. It subsequently precipitated a change in the norm for its use, and introduced the legal community to the convenience and cost-savings offered by virtual hearings. These changes must have played a role in the introduction of section 37C and Rule 38(9), in 2021.

18. Since the pandemic and these statutory changes, there have been significant technological advances in holding meetings and events virtually. Virtual trials and arbitrations have become commonplace and internationally recognised as being efficacious.[4] Current international jurisprudence supports a flexible approach which embraces the benefits of modern technology in having hearings being conducted virtually. Two cases which espouse this approach are Capic v Ford Motor Company Australia Ltd[5] and Joint Liquidators of One Blackfriars Ltd (In Liquidation).[6]

19. Each of the arguments of the defendant mentioned above will now be considered in turn.

(1) The hearsay argument

20. This argument was also advanced in Uramin in which the court held as follows at 239G-J:

“[16] Neither Dragone nor Barbaglia were, at the time of this trial, still in the employ of the defendant or the Areva Group and neither were based in South Africa. Neither are amenable to disruption of their working and personal lives and neither is subject to the control of or the wishes of the defendant. Defendant's attorney made an affidavit explaining that they were 'no longer under the supervision and control of the Areva Group'.

[17] Interestingly, neither Dragone nor Barbaglia made affidavits in support of this application or setting out why they were not available to be in the court in Johannesburg. This was the subject of some discussion at the hearing of the application. I took the view that the deponent to the affidavit is a senior attorney who had stated on affidavit that she had consulted with both Dragone and Barbaglia. There was and is no reason to doubt her bona fides in any manner. In any event, the contents of her affidavit were confirmed by both witnesses when we heard their evidence.”

21. This applies in the present matter. Danilewicz explained his circumstances in an email to Mr Potgieter (“Potgieter”), the attorney acting for the defendant, who is a senior attorney and officer of the Court. Potgieter reported under oath what was conveyed to him by Danilewicz. There is no reason to doubt his bona fides. The answering affidavit did not disclose any reason which the plaintiff has to dispute this.

22. The evidence as to Danilewicz’s circumstances as to not travelling to South Africa was therefore to be admitted.

23. Those circumstances are: Danilewicz is 86 years old. He finally retired about 3 years ago. He has had two hip replacements and a knee replacement, all of which impair his mobility. His wife suffers from retinal detachments which had caused blindness in her one eye and she is being treated for the condition in her other eye. She is totally dependent on Danilewicz for her activities of daily living. He is her full-time carer. They reside in the United Kingdom. In these circumstances, Danilewicz was willing to testify, but only by audio-visual link.

(2) The pleadings argument

24. This argument was probably based on the following obiter dictum in Standard Bank Factors Ltd v Furncor Agencies (Pty) Ltd and Others 1985 3 SA 410 (C) at 417H–418C:

“I may interpolate here that I can see good reasons for drawing this distinction. Although Bullen, Leake and Jacobs Precedents of Pleadings 12^th ed at 80 suggest that "there is no difference in effect between denying and not admitting an allegation" and that "the distinction is simply a matter of emphasis, a denial being more emphatic than a non-admission", I doubt that this is so. To my mind, there is a clear notional distinction between these two stances. A plaintiff faced with a positive denial must anticipate and prepare for the leading by defendant of rebutting evidence which contradicts the allegations he has made. A plaintiff faced with a non-admission need not anticipate and prepare to meet contradictory evidence to be adduced by the defendant. Indeed, there is authority for the proposition that he need not even anticipate a limited challenge by way of cross-examination of his witnesses. See Ntshokomo v Peddie Stores 1942 EDL 289 at 298. While that may conceivably be going too far (and again I refrain from deciding the point), I think, with respect, that the decision is undoubtedly correct in so far as it confirms that a denial (and, I would say, a fortiori a plea of non-admission) because of a lack of knowledge, will not entitle the pleader to contradict the plaintiff's averments by leading evidence to the contrary at the trial.” (my emphasis)

25. This view was criticised, in my view correctly, in N Goodwin Design (Pty) Ltd v Moscak 1992 (1) SA 154 (C) in which it was held as follows at 162B–163I:

“(T)he statement in Bullen, Leake and Jacobs Pleadings 12th ed at 80, is in my view valid that

'there is no difference in effect between denying and not admitting an allegation. The distinction is simply a matter of emphasis, a denial being more emphatic than a non-admission.'

A defendant is entitled to cross-examine plaintiff's witnesses as vigorously as he wishes and may lead rebutting evidence if he regards it necessary. Should that consist of expert testimony - which is likely - the Rules on that topic would ensure that plaintiff will not be caught by surprise.” (my emphasis)

26. This is also the view expressed in Erasmus Superior Court Practice:[7]

“A defendant is often in the position where he has no knowledge of certain facts pleaded by the plaintiff and is unable either to admit them or deny them. Subrule (2) allows a defendant in his plea to state which of the material facts alleged in the summons are not admitted and to what extent. In practice the defendant is not required to state why, in such circumstances, he has no knowledge, but it has been held that the non-admission must be accompanied by an explanation, such as no knowledge.

There is no difference in effect between denying and not admitting: the distinction is one of emphasis, a denial being more emphatic than a non-admission.”

27. I am therefore of the view that the proposed evidence of Danilewicz would be admissible at the trial and not precluded by the pleadings.

(3) The materiality argument

28. The defendant argued that this is not in line with the requirements for applications like the present, as it addresses none of the requirements set by section 37C of the Act, nor of Rule 38(9).

0cm; background: transparent; line-height: 150%">

29. I tend to agree.

30. The effect of the argument is to require the Court to assess, before having heard any evidence and based on argument, the importance of a particular witness to a party’s case. I do not think that such a criterion is, or should be, part of the correct approach as to whether a witness should be called.

31. If I am wrong in this, Danilewicz appeared to me to be an important witness who would be able to materially assist the Court. A perusal of this judgment confirms that this turned out to be the case in the clearest of terms. Indeed, other experts (especially Mr Loftie Ellis (“Ellis”)), called by the plaintiff as an expert) even referred to Danilewicz as a leading expert in the field and relied on his research and papers. Factors which were relied on in argument by the defendant included:

31.1. Danilewicz is the author and researcher to whom most reference is made by the wine experts in this matter.

31.2. His curriculum vitae reveals that he holds a PhD in chemistry, has lectured on wine oxidation at the Universities of Burgundy, Bordeaux and California and is either the author or the co-author of many scientific articles in this field. He was, at least on paper, much better qualified to testify about wine oxidation and its causes, including iron as a catalyst, than the other wine experts, Ellis and Van Rensburg. This turned out to be correct when oral evidence was heard.

31.3. Danilewicz is a highly qualified, experienced and distinguished chemist. He studied chemistry at Imperial College, University of London, and obtained his Ph.D in chemistry at the University of London. The first part of his career was as a scientist in chemistry. He later specialised in the chemistry of wines in which he has been involved for decades. He is an internationally renowned scholar and wine expert who published some of the most important and influential papers on the chemistry and mechanism of oxidation in wine. His work was referred to by Ellis, and in some of the academic papers placed before the court. His expertise is not in issue.

31.4. In answer to the arguments made by Ellis about iron in his report, Danilewicz pointed out in his reports that while iron as a catalyst may be responsible for increasing the rate of oxidation, the extent of oxidation in bottled wine is determined by the amount of oxygen available, and is independent of the concentration of the iron. Put another way, once the oxygen is depleted, oxidation stops, no matter the quantity of iron in the wine. This became a core feature in the trial and it features prominently in this judgment. This insight, which appears to have resulted from his pioneering work on the role of copper and iron in oxidation, is not contained in Van Rensburg’s report or Ellis’s report.

32. I considered that Danilewicz’s evidence might be of material importance in the matter and might significantly assist the Court in the trial. This turned out to be correct.

(4) The convenience argument

33. In Uramin, the court found at paragraph 30 that “it is sufficient reason that Dragone and Barbaglia are living and working elsewhere, do not desire to travel to South Africa, and have no obligation to either party by which they can be enticed so to do to find that this court should consider receiving evidence by video link”.

34. Potgieter stated in his affidavit that it would be quite possible for Danilewicz to see any and all material presented to him on his computer screen and still interact visually and audibly with the court and the legal representatives. During the trial, while at times Danilewicz did not have a physical document ready at hand while testifying, in each such case this was quickly remedied by the document being placed in the large screen in court (I estimate the screen to have been about 1 metre by 1.2 metres, albeit that this is a very rough estimate). Save for this aspect, which I regard as immaterial, what Potgieter stated in his affidavit proved to be correct on the whole.

(5) The own misfortune argument

35. Litigants are not obliged to limit themselves to experts who are able to travel to South Africa to testify. Leading experts in a particular field may be foreign (as appeared to eventuate in this matter). An inability to travel or a disinclination to travel, suitably reasoned, ought not per se to prevent a witness from being called when the technology which would allow him/her to testify is readily available

36. In the premise, I ordered that the evidence of Danilewicz would be heard by audio-visual link. As to costs, the plaintiff’s opposition was not entirely unreasonable. I am therefore of the view that the parties will pay their own costs in respect of the application in this respect.

B: The material issues

37. The plaintiff summarised its argument as follows in its heads of argument:

“We submit that, on the evidence presented by the plaintiff, and that of the defendant in respect of certain crucial aspects of the plaintiff’s case, the plaintiff has proven the following:

2.1 The wine filler sold by the defendant to the plaintiff was not properly pickled and passivated;

2.2 The lack of pickling and passivation constituted a manufacturing defect or flaw;

2.3 The defect impaired the utility of the filler;

2.4 As a result of the lack of pickling and passivation, iron ions (in the form identified as Fe3+) leached into the plaintiff’s wines;

2.5 The leached Fe3+ altered the chemical composition of the wine in that, as it entered the redox cycle, it oxidised polyphenols to create quinones which, in turn, bound free SO2;

2.6 The alteration of the chemical composition of the wines caused detrimental organoleptic changes in the wines;

2.7 The chemical reactions and resulting organoleptic changes (which occurred independently of oxygen) damaged or spoiled the plaintiff’s wines; and

2.8 Even if the wines became spoilt as a result of another or contributing cause, the lack of pickling and passivation would, in such a case, still have been a cause (not necessarily the cause) of the wine becoming spoilt.”

38. It is common cause that the plaintiff purchased from the defendant a 6-head Zambelli Tivoli filler and pump set (defined above as “the Filler”). It is also common cause that, at the time of the sale, the defendant was a merchant in the wine industry who had, for some 30 years, been offering wine industry-related products for sale, including bottling equipment.

39. The defendant pleaded that: “It is admitted that it was an implied term of the contract of sale that the item sold would be free of latent defects and fit for its objectively intended purpose.”

40. The plaintiff contends that this admission is in line with the case law confirming the principle that “a merchant who sells goods … in relation to which he publicly professes to have attributes of skill and expert knowledge is liable to the purchaser for consequential damages caused to the latter by reason of any latent defect in the goods.” (Holmdene Brickworks (Pty) Ltd v Roberts Construction Co Ltd 1977 (3) SA 670 (A) at 682H to 683A; Ciba-Geigy (Pty) Ltd v Lushof Farms (Pty) Ltd and Another 2002 (2) SA 447 (SCA) at par [48]).

41. This prompted much argument from the parties. This legal aspect is dealt with at the end of this judgment.

42. The plaintiff pleaded that the defendant breached the terms of the contract, firstly, by failing to deliver to it a Zambelli Tivoli filler and pump set and, secondly, by delivering a filler that had a number of defects. The defendant denied that it failed to deliver a Zambelli Tivoli filler and pump set, by implication pleading that what was delivered was the genuine product (“the Zambelli dispute”). With regard to the defects, the defendant pleaded that it “has no knowledge” and made no admission in that regard (“the defects dispute”).

43. In my view, the Zambelli dispute need not be decided in order to determine whether there was a breach of contract which caused damages. That determination, in my view, depends on the defects dispute. Put another way, if the Filler was not a Zambelli Tivoli filler and pump set, I do not think that this means that the damages were caused by that fact, a question which rather depends on whether there existed defects in the Filler which caused the wine to spoil. Had there been a claim for the return of the purchase price because a genuine Zambelli filler had not been supplied, then the Zambelli dispute would be relevant. There is, however, no such claim. I therefore do not consider it necessary to determine the Zambelli dispute.

44. Be that as it may, I find the following argument presented on behalf of the defendant in this respect compelling:

44.1. Walser's evidence was that he received an instruction manual for a Zambelli Tivoli filler with the machine that he bought, but that he threw it away.

44.2. The high-water mark of Walser's evidence that the Filler was not in fact a Zambelli Tivoli filler, was that it did not have a nameplate similar to the one shown in the instruction manual as "Attached to the Machine".

44.3. Dr Janet Cotton (“Cotton”), called as a metallurgical expert by the plaintiff, referred to the absence of a nameplate and a serial number on the Filler.

44.4. The absence of a nameplate does not prove that the Filler is not a Zambelli Tivoli filler.

44.5. During cross-examination Press provided a plausible explanation for why there may not have been a nameplate on the Filler, as follows:

“MR PRESS: They were identified to me as being Zambelli products. Some of them had labels on them. I understand that their fillers don't have the labels stuck onto them as a preference and the label is supplied to the, in the manual with, the user manual with the books so that the user can decide whether he wants to or they want to apply the label or not.”

44.6. In my view, the plaintiff did not prove that the Filler was not a genuine Zambelli Tivoli filler.

44.7. Be that as it may, whichever way this finding went, I am of the view that it has no material impact on the result of this matter.

45. With regard to the defects dispute, and specifically the words “fit for its objectively intended purpose” as pleaded by the defendant in paragraph 7.3 of its plea, as mentioned above, the plaintiff requested the defendant to provide the following further particulars:

46. As also mentioned above, the defendant responded as follows:

47. The plaintiff accepted in written and oral argument that it bears the onus in respect of the Zambelli dispute and the defects dispute.

48. I consider that the real and material question in this matter is whether the plaintiff succeeded in proving, as pleaded by it, that defects in the Filler caused “iron particles” to be introduced into the Wines which “… acted as an oxidation catalyst which, in turn, altered the qualities, characteristics and chemical composition of the wines.” Absent this being established, in my view the plaintiff’s case fails.

49. The plaintiff argued that given the principles laid down in Robinson v Randfontein Estates Gold Mining Co Ltd 1925 AD 173 at 177 and Shill v Milner 1937 AD 101 at 105, the Court is not called upon to decide other causes of damage to the wines since none have been pleaded by the defendant.

50. In my view, whether or not this is the case, it is not strictly necessary to decide on other causes of damage to the wines. In this regard:

50.1. To succeed, the plaintiff must prove the introduction of iron particles from the Filler which “… acted as an oxidation catalyst … ” and caused damage to the Wines.

50.2. The practical reality is that in considering whether the plaintiff succeeded in proving that the Filler caused damage to the wines, the factual matrix will inevitably involve (or stray into) a consideration of what caused the damage. I also refer to the authority in regard to pleadings dealt with above (in the section dealing with the application to lead evidence virtually) which undermines the plaintiff’s approach. Be that as it may, the consideration of the matter below results in the debate as to the effect of the pleadings not having to be resolved.

51. The plaintiff recorded in argument that it accepts that it has the onus to prove that the Filler was the cause of the alteration of the chemical composition of the wine and the detrimental organoleptic (meaning, roughly, sensory) changes and that it spoilt the Wines, but submitted that that onus does not include the onus to prove the absence of other possible causes. Relying on Thoroughbred Breeders’ Association v Price Waterhouse 2001 (4) SA 511 (SCA) at paragraph 66 (and its interpretation of the pleadings), the plaintiff argued that “(i) the plaintiff need only prove that the filler was a cause of the chemical composition of the wines being altered and they become spoilt (as opposed to the cause) and, (ii) should the plaintiff succeed in discharging that onus, this Court should find in its favour on the causation dispute.” In Thoroughbred Breeders’ Association it was held as follows at paragraph 66:

“The defence of a preponderance of fault on the part of the plaintiff, on which the Court a quo appears to rely, is incongruent within the field of contract. Where a plaintiff can prove that the breach of the defendant was a cause of the loss (as opposed to the cause thereof) he should succeed even if there was another contributing cause for the loss, be it an innocent one, the actions of a third party (compare Nedcor Bank Ltd t/a Nedbank v Lloyd-Gray Lithographers (Pty) Ltd 2000 (4) SA 915 (SCA) at paras [10] to [12]) or, logically, the carelessness of the plaintiff himself in failing to take reasonable precautions to avoid it. A defendant who commits a breach of contract does so independently of any of the extraneous factors mentioned above. All the requirements for his liability will have been fulfilled. In the absence of a contrary term in the agreement itself or of legislative intervention excluding or reducing his claim, he should therefore be held fully liable, regardless of whether the plaintiff's culpa was the dominant or pre-eminent cause of the loss.”

52. At its core essence, in my view, this matter turns on whether iron or rust or corrosion on the Filler, as a result of the manufacturing process, caused the Wines to spoil. And this, in turn, means what I have said above, namely that the real and material question in this matter is whether the plaintiff succeeded in proving that defects in the Filler caused iron particles to be introduced into the Wines which “… acted as an oxidation catalyst which, in turn, altered the qualities, characteristics and chemical composition of the wines.” Absent this being established, in my view, the plaintiff’s case fails.

53. There is also a debate between the parties as to whether any of the damages can, in law, (1) be claimed at all, (2) whether consequential damages can be claimed and (3) whether the damages claimed flow naturally from the breach. This will be dealt with at the end of this judgment.

C: Walser’s evidence: the same sulphur dioxide regime followed for 20 years

54. Much evidence and time was spent on this issue. I do not consider it necessary to consider it in depth.

55. Sulphur Dioxide (“SO₂”) is used to prevent wine from becoming oxidised. This is considered in some detail below. When it was put to Walser that he is not able to prove that SO₂ was added (for example, because he does not have the records of what was added), Walser answered as follows:

“MR WALSER: Yes, but we’ve been doing it the same before this happened, during this happened and after this happened. We have not changed our practices in sulphur regime.”

56. The court was requested by the plaintiff to find, “on a balance of probability, that, in respect of the twenty wines in question, the Plaintiff followed the same SO2 regime as before and added as much SO2 as was necessary to achieve a level of 40 parts per million before bottling the twenty wines in question.”

57. I do not agree that such a finding is to be made, but nor do I find that it cannot be made. At first glance this may seem peculiar. The reasons for this conclusion are set out below.

58. In my view, in any event, a finding on the above question is not dispositive or highly material to this matter which, as mentioned above, I believe, at its core essence, turns on whether the plaintiff succeeded in proving that defects in the Filler caused iron particles to be introduced into the Wines which “… acted as an oxidation catalyst which, in turn, altered the qualities, characteristics and chemical composition of the wines.” Absent this being established, in my view, the plaintiff’s case fails.

D: Factors argued in favour of the plaintiff’s case as to what caused the Wines to oxidise

59. Mr Coetzee submitted that various factors, including those which follow, show that the damage to the Wines was caused by defects in the Filler and excluded what he called “the Oxygen Theory” of Danilewicz which was effectively that either insufficient SO₂ was added or that there was too much oxygen for the SO₂ which was added.

60. Danilewicz’s evidence and explanations will be considered in the next section.

61. Various of the factors relied on by Mr Coetzee in argument follow.

62. First, the plaintiff had had no problems with its wines prior to those experienced in the instant matter and it had had no problems with its wines after those experienced in the instant matter. In this regard:

62.1. While of relevance, this nonetheless remains circumstantial and does not prove the cause.

62.2. Taken in isolation, it ignores the differences in bottling practices which were applied for the Wines (dealt with below).

63. Second, the plaintiff had followed its usual SO₂ regime of adding SO₂ prior to bottling described in the above section. While of relevance, this nonetheless remains circumstantial and does not prove the cause.

64. Third, the plaintiff had conducted a test by bottling 200 bottles of its Aasvoël wine with Amata, an independent bottler who bottles wines for winemakers, referred to further below, without certain protections (such as a vacuum corker). A Vinlab[8] test was done revealing that the oxygen level was not substantially increased in this exercise:

64.1. While of relevance, this nonetheless remains circumstantial and does not prove the cause.

64.2. Mr Duminy pointed out and argued inter alia as follows:

64.2.1. One does not know what the SO₂ was in the wine.

64.2.2. It is a 2020 wine. One does not know how it compares to the Wines.

64.2.3. One of the sets of Vinlab test results in this regard record, in respect of the wine from one of the tanks (tank 01) from which it was drawn, the pre-bottling SO₂ figures as 41mg/l Free SO₂ (“FSO₂”) and 97mg/l Total SO₂ (“TSO₂”) which is sufficient and 26mg/l FSO2 and 84mg/l TSO2 which would be on the low side.

64.2.4. Another set of Vinlab results records 36mg/l FSO2 and 92mg/l TSO2.

64.2.5. The set of Vinlab results for oxygen data figures on which the plaintiff relies, however, and in contrast, does not record any FSO₂ and TSO₂ data for the wine (Aasvoël).

64.2.6. Further, there is no indication whether the SO₂ was tested and evaluated or what the SO₂ levels were and when. What is recorded is therefore inadequate.

64.2.7. The results therefore do not assist insofar as the SO₂ regime is concerned.

64.3. It also goes without saying that in a test of this nature the best effort would have been made to minimise oxygen intake which may well skew the results from what was done with the actual bottling of the Wines. I do not mean this from a sinister perspective, but from the simple fact that what was been conducted was a test.

64.4. There were no problems with four of the Wines which are part of the plaintiff’s claim, despite having gone through the Filler. There is no explanation for this.

65. Fourth, Amata (now iBottle) bottles approximately 50% of wines without vacuum. Mr Duminy pointed out that this includes screw caps and in fact only 10% were corked without vacuum. He also noted that not all of the changed bottling practices employed by the plaintiff (dealt with below) with the Wines were used by iBottle.

66. Fifth, 8650 bottles of Moment of Silence were bottled by 15 November 2024 using the Filler with no problems:

66.1. Mr Duminy pointed out that this was a white wine and that there was adequate pre-bottling SO₂ (the FSO₂ was 37mg/l).

66.2. He contrasted this with the Im Winterhof wine which had a pre-bottling FSO2 of 4mg/l. This means that an enormous amount of oxygen was consumed. To get to the plaintiff’s required 40mg/l FSO₂ for this wine one would have needed to add 54mg/l TSO₂ in total (based on the ratio of FSO₂ to TSO₂ of 1 to 1.5). This means that there had to have been 10mg/l of oxygen to oxidise that wine which is huge (it is beyond the maximum oxygen level for wine of 9mg/l). Mr Coetzee countered that this was not so because this wine was an example of ‘stuck fermentation’ in which yeast can affect the SO₂levels. That, however, does not address the initial low pre-bottling SO₂ level.

67. Sixth, Ellis identified a sensorial characteristic of ‘burnt aldehyde’ in February 2020 more than a month before he observed rust in the Filler for the first time on 17 March 2020. Ellis recorded in an email at the time that there was a “bruin tint wat duidelik op oksidasie wys” (my translation: brown tint that clearly indicates oxygen). He testified:

“Can I just come back to 1.6 and tell you that, mister – M’Lord, if wine shows oxidation you pick up the aldehy(d)e …..In this case we had the burnt sensation with the aldehy(d)e and this is why it’s so important for me to indicate to you that this is not directly oxidation, at an early stage I’ve picked up a burnt oxidation character.” [the plaintiff’s underlining]

67.1. Ellis does not say that this wine was not oxidised.

67.2. On the contrary, his approach seems to be that the wine was oxidised, albeit that ‘burnt’ is not a direct indication of oxidation.

67.3. The ‘aldehyde’ descriptor is typical of oxidation.

67.4. Mr Duminy pointed out and argued that Ellis himself had said that there was no science behind what he testified. In other words, he cannot say that it is caused by the uptake of FE3+ which is what happens when iron oxide/rust is dissolved into wine (this aspect is explained and considered in lengthy detail below).

67.5. Probably most importantly, this description of burnt aldeyhyde character by Ellis only applied to one wine (Oorbegin). It does not apply to any other wine. It therefore does not apply across the board.

67.6. Mr Coetzee argued in reply, with reference to test results, that the chemical analysis is something different to the sensorial analysis. While this may be so, it cannot change whether iron/rust uptake oxidises wine or increases the amount or extent of the oxidation of wine. It does not: it only acts as a catalyst and increases the rate of oxidation, not the amount or extent thereof. That topic is dealt with in depth in the next section.

67.7. I consider that Mr Coetzee’s statement mentioned early in the next section, as well as the evidence quoted thereat, disposes of the ‘burnt aldehyde’ factor referred to above.

68. Seventh, the plaintiff had followed the same bottling practices as Amata:

68.1. This is not correct.

68.2. It is dealt with more conveniently under section G(3) below which applies equally here.

69. Most importantly, the science behind the oxidation of wines was explained by Danilewicz in detail which, in my view, excludes the conclusion drawn from the above circumstantial evidence. This is dealt with in detail in the next section.

70. Mr Coetzee argued that the above factors meant that Danilewicz’s explanations must be wrong.

71. The question is what the science is on the oxidation of wines to determine whether what Mr Coetzee argued has been shown to be possible.

72. I am therefore of the view that Mr Coetzee’s argument is far from conclusive, subject to much circumspection and fails to convince. As I will attempt to demonstrate below, in my view it does not pass muster in the light of the science presented by Danilewicz in evidence.

E: The oxidation of wine and the role of iron and oxygen: only oxygen oxidises wine, while iron and rust (taken up as Fe3+) only act as catalysts, with it, and the quantity thereof, having no effect on the extent of the oxidation but only on the rate thereof

73. Danilewicz, whose credentials are mentioned above, explained the role of iron and iron oxide (rust) in wine. Corroded iron or rust does not have some special property affecting wine beyond the role of a catalyst, and that, if it did, this would be a new concept in chemistry which would be studied intensely. The plaintiff’s counsel made it clear that he was not going to suggest this:

“MR COETSEE: When you say it has got a lot of iron in it, then I am supposing you are referring to the measured Fe level, but then you say the ordinary Fe3⁺, what do you mean by the ordinary Fe3⁺?

MR DANILEWICZ: I think I meant to the Fe3⁺, that is normally present in wine. It is not an imaginary Fe that has special catalytic properties. Because all the time you have got this proposal that rust produces this, what they call, corroded iron and somehow corroded iron implies to have a special property. So when I say ordinary Fe3⁺, I mean the one that is present in wine, the one that I know, not the one that I have never heard of.

MR COETSEE: You can accept that I will be asking you about Fe2⁺ and Fe 3⁺ the way you know it. I will not be referring you, or suggesting, or hinting at some mysterious other form of iron. It will be either Fe2⁺, or Fe3⁺. That you can take as a given Dr Danilewicz.”

“DR DANILEWICZ: … The idea that the point was made that during the bottling procedure, there was no rust visible Mr Ellis suggested that you don't have to see the rust. It could be there, not visible. And it could dissolve or go into the wine, and you wouldn't be able to measure it. That is totally not possible. So I don't know what exactly you’re drifting [sic - should be "driving"] at, but that's what I understand the situation is, as far as metals and rust are concerned. The thing is that if there was another species, some very, very potent catalyst, you've got to remember there'd have to be two because they enter the Redox cycle. And as far as I know, there is no other. And I don't know, if there were, it would be an extremely interesting phenomenon which I think people would be looking at intensely. Because if it happens in iron, it could happen in all the other metals, in cobalt, nickel, manganese. You know, that would be new concept in chemistry.

MR COETSEE: You can accept, Dr Danilewicz, I'm not going to suggest to you that there's a new concept in chemistry." (emphasis added)

74. As mentioned, I consider that Mr Coetzee’s statement in the above paragraph, as well as the evidence quoted above, disposes of the ‘burnt aldehyde’ factor referred to in the above section.

75. That an increased iron concentration will affect the rate of oxidation of the wine but not the amount of oxidation, as testified by Danilewicz, was accepted by the plaintiff’s counsel during his cross-examination:

"DR DANILEWICZ: So, I agree that if you add iron in either form, in this first phase of oxidation, you will increase the rate of oxidation, but you will not increase the amount of oxidation. Do you see what I mean? At the end of the phase, at the end of this rapid phase, the iron won't increase the amount of oxidation. It won't increase the amount of SO² that is used. (emphasis added)

[. . . ]

MR COETSEE: You can accept that there's no dispute about what you've just discussed."

76. Danilewicz explained that an increase in the iron content of wine, including if accompanied by copper, accelerates the rate of oxidation but does not increase the amount of oxidation. That depends on the availability of oxygen. In wine, oxidation cannot occur in the absence of oxygen.

77. He repeatedly pointed out, without contradiction, that the redox cycle responsible for the oxidation of wine (explained in detail below) was dependent on oxygen. Absent oxygen, oxidation of wine cannot occur and SO₂ would not be consumed in the presence of Fe³⁺ (explained in detail below).

78. Oxidation of wine requires the presence of oxygen. The more oxygen present, the greater prospect and the extent of the oxidation.

79. Conversely, Danilewicz explained that in the absence of (sufficient) oxygen (or once it is depleted), no amount of iron in the wine will cause oxidation:

“And when I said to you that there were other oxidants, well, there are, but they don't happen to be in wine. The oxidant in wine is oxygen. I don't think there's any presence of any significant sources of oxidation in wine, oxygen is the oxidant in wine. In other systems, it might be something else, potassium chromate, potassium manganate. There are lots of oxidants that could be, but not in wine, it's oxygen. And the iron that you're adding through this, in this process, will do nothing at all unless there's oxygen. Unless the cycle works, the iron can't cycle.”

80. This means that in the absence of oxygen, there is no functioning redox cycle.

81. Danilewicz was not challenged on any of the above aspects.

82. The question is whether iron or iron oxide (rust) from the Filler was a material cause of the oxidation.

83. This brings into focus the plaintiff’s summary of its case, quoted verbatim above, portion of which warrants repetition at this point:

2.1 As a result of the lack of pickling and passivation, iron ions (in the form identified as Fe3+) leached into the plaintiff’s wines;

2.2 The leached Fe3+ altered the chemical composition of the wine in that, as it entered the redox cycle, it oxidised polyphenols to create quinones which, in turn, bound free SO2;

2.3 The alteration of the chemical composition of the wines caused detrimental organoleptic changes in the wines;

2.4 The chemical reactions and resulting organoleptic changes (which occurred independently of oxygen) damaged or spoiled the plaintiff’s wines; and

2.5 Even if the wines became spoilt as a result of another or contributing cause, the lack of pickling and passivation would, in such a case, still have been a cause (not necessarily the cause) of the wine becoming spoilt.

84. In essence, it appears to me that the plaintiff argues that what is described in 2.2 above resulted in the damage to and spoiling of the wines in question.

85. I disagree, for the reasons set out below.

86. As mentioned, Danilewicz explained that various materials can cause oxidation in different substances (i.e. can be oxidants), but that wine only oxidizes if oxygen is present (i.e. the only oxidant in the case of wine is oxygen). Put another way, without oxygen being present in the wine or the air in the wine bottle headspace (the area between the cork and the wine), the wine cannot oxidise. He was not contradicted on this. It is noted that this is subject to there not being a problem resulting in the ‘Oxygen Transmission Rate’ (“OTR”) through or past the cork being excessive, but that is a much slower process (usually over a year) which does not apply to the instant matter in which the Wines were alleged to have been spoilt within a few months.

87. In chemistry terms, oxidation in wine occurs in what is known as a redox reaction or cycle. Further in this regard:

87.1. The following figure prepared by Danilewicz illustrates the redox reaction (it is from page 285 of the defendant’s expert bundle, hence the number on the top right which has nothing to do with the science).

87.2. Iron acts as a catalyst in the reaction (as does, for example, copper).

87.3. It is included in the above figure because of the part alleged by the plaintiff for it to have played in this matter.

87.4. Iron, however, does not result in oxidation on its own. It is a catalyst.

87.5. In the case of wine, oxygen is the only oxidant which causes oxidation.

87.6. Once the oxygen is depleted, there can be no oxidation.

87.7. The redox reaction works with iron (and rust) as follows:

87.7.1. Iron in the form of Fe₂+ releases an electron and becomes iron in the form of Fe₃+ (Fe is the symbol for iron).

87.7.2. On the right-hand side of the figure, the released electron reacts with oxygen, creating hydrogen peroxide, which would create an unpleasant malodorous oxidation product.

87.7.3. This is where the SO₂, which features so prominently in this matter, comes in. SO₂ prevents the creation of the unpleasant malodorous oxidation effect by binding with the hydrogen peroxide, thereby eliminating its effect. This is represented by the X at the bottom of the right-hand side of the figure.

87.7.4. The Fe₃+ receives an electron from the polyphenol on the left-hand side of the figure, resulting in the polyphenol being converted into a quinone. Quinones in wine would result in yellow/brown pigment in the colour of the wine.

87.7.5. SO₂ steps in here again. SO₂ prevents the creation of this oxidation effect by binding with the quinone, thereby eliminating its effect. This is represented by the X at the top of the left-hand side of the figure.

87.7.6. The left-hand side of the reaction has a consequence for the right-hand side which is the production of Fe₂+ which reduces the amount of oxygen (the consumption of oxygen), producing hydrogen peroxide which is bound by the SO₂, thus preventing oxidation of the wine. Once the oxygen is depleted, oxidation cannot happen.

87.7.7. In this cycle, for every molecule of oxygen consumed, four molecules of SO₂ are required.

87.7.8. The cycle spins and continues until (this takes approximately eight to twelve weeks):

87.7.8.1. the oxygen is depleted, in which event the wine will not become oxidised; or

87.7.8.2. the SO₂ is depleted (or, more correctly, reduced to below 10mg/l) and oxygen remains, in which event the wine will become oxidised.

87.8. SO₂ is added to wine to prevent oxidation by means of the above process.

87.9. If there is sufficient SO₂ to bind the quinones and hydrogen peroxide until the oxygen is depleted the wine will not oxidise. If not, the wine will oxidise.

87.10. Once the SO₂ reduces to a level of less than 10mg/l, the negative effects associated with the oxidation of wine will start to be released.

87.11. Accordingly, the wine will oxidise if the SO₂ is insufficient for the level of oxygen, or from the opposite perspective, if the oxygen is too great for the SO₂.

88. Key to understanding this matter is an appreciation of the aspect mentioned above that the redox cycle continues until:

88.1. the oxygen is depleted, in which event the wine will not become oxidised; or

88.2. the SO₂ is depleted (or reduced to below 10mg/l), in which event the wine will start to become oxidised.

89. Winemakers will add SO₂ to the level which they consider appropriately sufficient (normally including a buffer) to deal with the expected level of oxygen. There are legal limits as to what can be added. Walser aimed to have 40mg/l. Danilewicz and Ellis both considered this sufficient, provided that the oxygen in the wine was not too elevated.

90. Wine has an oxygen level of between 1 and 9mg/l of oxygen. This depends on (1) the dissolved oxygen in the wine, which usually happens while the wine is fermenting in tanks, and (2) the amount of oxygen in the headspace of the bottle after bottling (this is an effect of bottling). This oxygen is the subject of the redox reaction/cycle explained above which lasts approximately eight to twelve weeks, depending on the level of oxygen. This is a relatively short period. Oxygen is also introduced by a third mechanism through or past the cork. The rate at which this occurs is the Oxygen Transmission Rate (OTR) mentioned above which is much slower and usually takes more than a year. This time frame excludes this aspect as a playing a material role in this matter.

91. What the actual oxygen and SO₂levels of the wines at bottling were in fact, and the debate as to certain calculations by Danilewicz, are, in my view, irrelevant to the determination of this matter, as will emerge from what is set out below.

92. Danilewicz explained that iron and rust are dissolved in wine and taken up in the form of Fe3+. The question in this matter is whether this Fe3+ uptake (in the form of iron or rust) from the Filler caused the oxidation of the wines. I am of the view that this played no role, for the reasons set out below, and therefore the plaintiff’s claim must fail. This conclusion and result applies even if there was iron uptake (including in the form of rust) from the Filler. The evidence of Danilewicz and the explanation of the redox cycle are key to understanding this, hence them having been set out in much detail above.

93. Danilewicz and Ellis explained that:

93.1. Iron is always present in wine, in varying degrees. The acceptable worldwide average is 5.5mg/l, although concentrations are generally lower when modern stainless steel equipment is used. The legal limit is 10mg/l.

93.2. Iron dissolves in wine because wine is acidic. The same applies to iron oxide (rust).

93.3. Neither iron nor iron oxide (rust) oxidises wine.

93.4. They both act as a catalyst in the oxidation of wine in the redox reaction described above.

94. The post-bottling test results in respect of the Wines were between 1.21mg/l and 5.05mg/l of iron, with only 5 out of the 20 wines in question being above 3mg/l. The average was 2.9 mg/l, well below the worldwide average and acceptable limit (and very much below the legal limit). Bearing in mind that the minimum iron level of wine is 1mg/l, this means that on average at most 1.9mg/l of iron/rust could have been added to the Wines.

95. As mentioned above, the redox cycle continues until:

95.1. the oxygen is depleted, in which event the wine will not become oxidised; or

95.2. the SO₂ is depleted (or reduced to 10mg/l), in which event the wine will become oxidised;

and this process can take approximately eight to twelve weeks.

96. It is important to reiterate that Danilewicz explained that (he was not challenged on this):

96.1. The effect of adding iron or iron oxide (rust) to wine is to speed up the rate of oxidation.

96.2. It has no effect on the amount or extent of oxidation which occurs.

97. Mr Coetzee, I suspect having realised that the plaintiff’s case as to iron/rust uptake from the Filler would not increase the amount or extent of oxidation, attempted to add a rider to the analysis in cross-examination. He focused on the idea of the depletion of SO₂ in the wine by the initial uptake of extra Fe3+ from the dissolving or iron/rust on the Filler resulting in the oxidation. His idea, put to Danilewicz, was in essence that the FSO₂ used to bind the quinones produced by the Fe3+ introduced from the Filler would no longer be available, thereby depleting to insufficient levels the SO₂ available to combat oxidation. This was put in further detail as follows: if further iron or rust is added in the form of Fe3+, it will enter and become part of the redox cycle at the bottom of the middle part of the above figure, resulting in the movement of an electron from the polyphenol on the left-hand side to each molecule of Fe3+, converting the polyphenol to a quinone which causes oxidation and converting the Fe3+ to Fe2+ which continues the cycle. The quinone created binds with SO₂ (in the process described above, which prevents oxidation), thereby consuming it, as a result of which the SO₂level is reduced which compromises the quantity (and, therefore, ability) of the SO₂to prevent oxidation.

98. Danilewicz explained that this was a false premise for four main reasons:

98.1. The first obvious reason is that the creation of quinones and them being bound by SO₂ (i.e. SO₂being consumed in the process) is part and parcel of the cycle and catered for in it, with everything happening simultaneously (as explained and illustrated above).

98.2. The second just as obvious reason is that the addition of iron or rust does not increase the amount or extent of oxidation, it simply speeds it up.

98.3. The third reason is that, while the consumption of SO₂ is increased by the speeded-up reaction, so does the consumption of oxygen, thereby having no overall effect on the capacity for oxidation.

98.4. The fourth reason is as follows:

98.4.1. SO₂ has a greater molecular weight than Fe3+. The result of this is that for every 1mg/l of Fe3+ added, 0.57mg/l of TSO₂ (which is the total of FSO₂ and bound SO₂) will be consumed.

98.4.2. If 3mg/l of Fe3+ is added, which he described as a huge amount, this would result in the loss of 1.71 mg/l of TSO₂ at the time of initial introduction. This is in the context of 40mg/l being the level of FSO₂ aimed for (by Walser).

98.4.3. The reduction in SO₂ will therefore be in an immaterial amount which is neither here nor there, having no material effect on the reaction.

98.4.4. As dealt with above, on average 1.9mg/l of iron/rust could have been added to the Wines, which means that the addition of that amount would have resulted in, on average, an even more immaterial quantity of SO₂ being consumed.

98.4.5. Bearing in mind that 40mg/l of SO₂ is needed pre-bottling to protect wine, this is a negligible and irrelevant amount.

98.4.6. Even this is, of course, unrealistically skewed in favour of the plaintiff because, as dealt with above, 1mg/l is the minimum iron level of wine with the result that the actual levels were in all probability more, meaning that the amount of SO₂ consumed would have been even more negligible and irrelevant.

98.4.7. The redox reaction then continues as before, but at a quicker rate, consuming the oxygen.

99. In the above respect, Danilewicz’s evidence included the following (even a very large amount of iron would react with very little SO₂):

“So, it would mean three milligrams per litre of iron, three milligrams per litre would cause 1.7 milligrams [per litre] of SO₂ to react.”

and

“In fact, if you were to put in three milligrams per litre, that is a huge amount of iron, which I do not think you could sit [sic - should be "say"] any of the wines had more than that. You would have lost 1.7 milligrams per litre of SO₂. So, the effect of iron, you are saying that there are other oxidants that are present. True, Fe3, you have added, you have added an oxidant. The result is that the SO2 changes by a miniscule amount. So, this is why this questioning about another source of oxidation, the addition iron to the system. What would it do? Well, unless the cycle works, it would be very small.”

“So however much iron you throw at the system, if there is no oxygen, nothing is going to happen. So, the amount of oxidation, as I keep repeating, is dependent not on the amount of iron you put into the system, but the amount of oxygen that is present.”

100. He continued as follows:

"DR DANILEWICZ: If you put in three milligrams per litre, yes you need two irons, there is two of them and the SO₂ is 64, so you divide 64 by 111.6, I get that milligram per litre of iron you add, it will be 0.57 milligrams of total SO₂ will be consumed. It is not very much."

101. There was no expert evidence that rust causes the problems. On the contrary, the unchallenged expert evidence of Danilewicz is that wine is acidic and removes iron and rust taking it up in the form of Fe3+. There is nothing in the redox cycle which was not there before.

102. The result of all of the aforegoing, in my view, is that the uptake of iron/rust from the Filler did not cause the oxidation of the Wines.

103. In my view, this alone disposes of the case in favour of the defendant and the plaintiff’s claim must be dismissed.

F: When the Filler, if not properly pickled and passivated, would have displayed rust

104. Walser testified at length, including as follows:

104.1. The plaintiff purchased the Filler from the defendant.

104.2. The defendant’s offer to train the plaintiff’s employees to use the Filler was not taken up by the plaintiff because the procedures are simple.

104.3. Walser described the Filler before it was used for the first time on or about 11 or 12 November 2019 as being ‘spic and span’. This also appeared from photographs placed before the court. When it was delivered to the plaintiff, the Filler was in pristine condition.

104.4. No rust was observed on it despite it being used on 15 different days between 13 November 2019 and 11 February 2020.

104.5. The plaintiff started using the Filler on or about 13 November 2019 and used it for bottling 27 of its wines from that date to 11 February 2020, on 15 different dates, namely 13, 15, 22 and 27 November 2019; 3 December 2019; 9, 10, 13, 14, 16, 17, 25 and 27 January 2020; and 3 and 11 February 2020.

104.6. After being used for bottling the Filler was hosed down. Walser said that the machine was not physically dried and is left to dry by itself.

104.7. Once bottling had been completed for the season, the Filler was put in the office space next to the printer and the desk of an employee, Stefan Johannes (“Johannes”). The machine looked in perfect condition, in particular there were no signs of rust.

104.8. Walser said that he would say that on 9 January 2020 they (by which I consider he meant himself, Johannes and the workers involved) all looked into the filler bowl (there was no bottling from 3 December 2019 to 9 January 2020). They saw inside because they cleaned it. Nobody noticed any rust.

104.9. The Filler was thereafter used on 10, 13, 14, 16, 17, 25 and 27 January 2020 and on 3 and 11 February 2020. The same applies to these dates as well. Walser did not notice any rust on the Filler at that stage.

104.10. The plaintiff last used the Filler on 11 February 2020. No wines were bottled using it after that date. Immediately after that last use, it was rinsed with raw borehole water. This is because there would be contamination issues were wine to be left inside the Filler. Walser said he was most likely there when it took place. Since then, the Filler stood indoors next to the printer and Johannes’s work-space in the open plan warehouse. It was visible to Johannes and him from 11 February 2020.

104.11. No rust was seen until Walser visited Ellis on 17 March 2020. Walser saw rust on the Filler that day for the first time. He took the Filler in his bakkie to Ellis on the same day. Ellis took photographs on his cell-phone on 17 March 2020 in the driveway of his office. Rusted areas were clearly visible.

104.12. Subsequent to the problems arising, Walser bottled one of the wines which had spoilt (called Aasvoël). He used Amata to bottle these. One batch was bottled using inert gas and vacuum (i.e. to reduce the possible effect of oxidation), and one without. Videos were played depicting this. Approximately 200 bottles were involved. Walser observed the bottling and provided the wine. He did nothing else in the process. There were no problems with both batches.

104.13. Walser and Ellis both knew that if they used Amata to bottle there would be oxygen.

104.14. Winemaking is complex with many steps. Bottling is the final step. It is a critical step. One can change the product at many steps in the process by mixing and adding. One can add and remove items as permitted by law until bottling, but once bottled, that is it and the wine cannot be changed.

105. The aforegoing evidence establishes the following during which no rust/oxidation was observed:

105.1. Although there was no actual evidence on this aspect, it flows from the aforegoing evidence that the Filler was, once manufactured, transported to the defendant which must have taken some time. This time is not taken into account in the figures below, but would only serve to exacerbate the conclusion.

105.2. The plaintiff purchased the Filler and took possession thereof on 11 or 12 November 2019.

105.3. It was first used by the plaintiff from 11 or 12 November 2019 to 3 December 2019, a period of just over 3 weeks.

105.4. It stood at the plaintiff’s premises from 3 December 2019 to 9 January 2020, a period of 38 days, including both start and end days (5 weeks and 3 days).

105.5. It was used on 9, 10, 13, 14, 16, 17, 25 and 27 January 2020 and on 3 and 11 February 2020.

105.6. It stood at the plaintiff’s premises from 11 February 2020 to 17 March 2020, a period of 36 days (2020 was a leap year, hence the extra day of 29 February), including both start and end days (5 weeks and 1 day).

105.7. No rust/oxidation was seen until Walser visited Ellis later on 17 March 2020 when rusted areas were clearly visible. What infer from this is that the rust must have become visible some time before that.

106. This means that the Filler was in the plaintiff’s possession and use for at least four months in which no rust/oxidation was observed.

107. Cotton testified that if the Filler was not properly pickled and passivated on manufacture and iron leached out, the effect would be immediate. She answered the following question in the affirmative:

“Yes, if it is exposed to air, you would expect it to be – for red rust to be evidence immediately, yes?”

108. She testified that, with normal exposure to oxygen, there will immediately be oxidising if iron is able to leach out. The reaction happens instantaneously. The layer must be fully intact and covering the whole item without any gaps (i.e. homogenously and continuously, which is imperative) to work. That is the purpose of final treatment. If not, iron oxide is immediately created in the form of red rust. One would expect red rust to be evident immediately if exposed to air.

109. Press testified in similar vein: that iron contamination on a surface results in corrosion within hours. He said that it should have appeared in the first days of operation and certainly during the break in operations between November and December and the five-week break after January.

110. On questions from the Court, as to how rapid (it having been described as instantaneous) the reaction would be in the light of her earlier testimony, Cotton testified that if the Filler was not properly pickled and passivated on manufacture, it would take a matter of weeks for the rust/oxidation to show. This includes time standing (for example at the manufacturer) or in a show room when not being used. She testified that it would be quicker if used. The invisible reaction will happen immediately in the filler at a nano level but will not manifest in anything visible (at a macroscopic level) for a matter of weeks.

111. The Filler was in existence for much more than a matter of weeks without showing any rust/oxidation:

111.1. As mentioned, while there was no evidence as to the actual time involved, the Filler must have spent some time in transport and at the defendant’s premises prior to possession having been taken by the plaintiff (although this is not taken into account in the analysis below).

111.2. It was at the plaintiff’s premises in use and standing for over four months. It was used in this time as summarised above. Cotton stated that the use of the Filler makes the rust/oxidation process faster.

111.3. Cotton opined that the wine running over the surface of the tank of the Filler and the act of cleaning could wash the rust away (“you may get rid of that evidence by way of use”, “evidence being washed away”), but in my view this does not have a material impact because (1) it means that the rust would have had to have been there nonetheless in order for it to be washed away, but it was not seen (2) the Filler was being used which sped up the appearance of rust (3) the Filler was hosed, not washed and (4) yet the Filler stood for a period of 5 weeks and 3 days between 3 December 2019 and 9 January 2020 with no appearance of rust, as well as for other periods. This is 38 days including the start and end days.

112. I think that what resolves the matter on the probabilities is the following:

112.1. In response to a series of questions for the purposes of clarification from the Court, Cotton gave the following answer which indicates the re-appearance of the rust would be after a matter of days:

“MS COTTON: Yes, so what I mean by the evidence being washed away is that any haematite that could have possible been visually, microscopically visual, would have been washed away through the flow of the product through the wine filler and then also the cleaning process. If the wine filler was then used the very next day or several days later, I would not expect there to be any visual rust, but certainly, that haematite product could be washed away with use and find itself in the product.”

112.2. The last bottling date was 11 February 2020 (2020 was a leap year). The Filler was taken to Ellis on 17 March 2020, when the rust was first seen. This was 36 days including the start and end days (5 weeks and 1 days). The underlining is to emphasise that the rust could (or, more accurately, probably would) have appeared earlier as the Filler was not being used and the rusted areas were described as being clearly visible on 17 March 2020, which would reduce this 36 day period.

112.3. The Filler had earlier stood at the plaintiff’s premises from 3 December 2019 to 9 January 2020, a period of 38 days, including both start and end days (5 weeks and 3 days).

112.4. This means that it had earlier stood for a longer period with no rust appearing (a difference which probably would have been more because the rust must have appeared earlier, it having been described as being clearly visible on 17 March 2020).

112.5. This in turn means, on the probabilities, that:

112.5.1. had the relevant part of the Filler bowl not been properly pickled and passivated, the rust would have appeared in that earlier longer period;

112.5.2. rust on that the relevant part of the Filler bowl was therefore not caused by that part not being properly pickled and passivated.

113. As an aside, the condition of the Filler as identified in the evidence (dust, insects, dirt etc) indicates that it may not have been stored in the office at all times and rather exposed to a more ‘aggressive’ environment which may have played a role in the corrosion, as testified by Press, but this is not needed for the analysis of, and conclusion on, this aspect.

114. On plaintiff’s evidence, therefore, I am of the view that, had the inside of the Filler bowl not being properly pickled and passivated, on the probabilities, it would have shown rust/oxidation well before it actually did on 17 March 2020, and should have done so in the earlier longer period of no use leading up to 9 January.

115. The consequence of the aforegoing is that, I am not in a position to find on a balance of probabilities that the rust/oxidation on the inside of the Filler bowl (as opposed to certain welds on the upper lip and other areas)[9] was caused by the Filler not being properly pickled and passivated (and therefore that any oxidation of any of the Wines would have been caused thereby, which has, in any event, been determined in the previous section).

116. In my view, this is a further and independent basis why the plaintiff’s claim fails.

G: Further problematic questions impacting on the plaintiff’s case

117. There are further problems with the plaintiff’s case in the context of section F above.

(1) The quantity and materiality of any iron/rust pickup from the Filler

118. The plaintiff’s case is that the iron/rust pickup was from the wine being in the Filler.

119. In the bottling process the wine is pumped in a continuous process from the tanks into the Filler through which it flows into the (six) wine bottles in place.

120. While there may possibly be interruptions at times, for example if the filler gets too full, for which there are sensors to stop the pumping, it is a continuous process. This has the result that any portion of the wine is in the Filler for a matter of seconds, at most a minute or two, and even less is in contact with any corrosion on the Filler.

121. Even though, in my view, its pick-up makes no difference to the extent of oxidation, as considered above, there was in any event no evidence as to the quantity of rust/iron which could be picked up from a few seconds or a minute or two’s contact with rust on the Filler. On this basis, too, the plaintiff has in my view failed to make out a case.

122. This was exacerbated by the fact of variation in the iron values in different samples of the same wine. I do not believe that this was shown to be the result of different time spent by wine in the Filler (which could only be a matter of seconds, at most a minute or two), as argued by the plaintiff. A similar problem for the plaintiff presented with the variations in TSO₂in the wines.

123. Of more import than my belief, is that there was no scientific basis proffered for this.

124. Put simply, in my view, the plaintiff failed to make out a case in this regard.

(2) Related to, but standing independently of, the above aspect is that there was no evidence as to the quantity of rust/iron which would be required to cause the wine to spoil (again, even though its pick-up makes no difference as explained above)

125. It would have been a simple task for the plaintiff to conduct a test with iron/rust and wine, but this was not done. On this basis, too, the plaintiff has failed to make out a case.

126. In regard to conducting a test, one of the reports of Van Rensburg (filed at the instance of the defendant) summarised such a test. Although he did not testify, the test (and most of his report) was dealt with Ellis in cross-examination and he agreed with most of it.

127. The purpose of the experiment was to obtain an idea of the extent to which the exposure to iron in a filler bowl during bottling could lead to increased levels of iron in the wine and what effect that would have on the oxidation of the wine, with oxygen excluded.

128. A teabag with 4.65 grams of iron filings was put in 4.65 litres of wine in a filler for more than 10 minutes and in another the teabag was left in the bottle overnight. In the control bottle the FSO₂ was 53mg/l and TSO₂ was 138mg/l. Copper was 0.21mg/l and iron was 0.51mg/l. The analysis of wine which had the iron for 10 minutes and overnight was recorded. For the 10 minutes item, the FSO₂ was 53mg/l and TSO₂ 138mg/l. Iron was (in mg/l) 0.53, 0.51, 0.54 and 0.57, which was a moderate uptake in the wine.

129. This is statistically irrelevant.

130. For the overnight item, iron was 3.67mg/l. The longer the exposure, the higher the uptake of iron.

131. What is important is not whether the results of the test are accurate or not, but rather that this further illustrates that no case has been made out by the plaintiff that the minimal period for which the wine would have been exposed to any iron or rust in the Filler (considered above) would have had the effect of the pick-up of sufficient Fe3+ to have the effect on the Wines claimed by the plaintiff (although all indications are that it would not have had that effect).

(3) Changed bottling practices

132. Ellis testified that “There were differences in the plaintiff’s practices beyond the use of the Zambelli, such as the use of inert gas, vacuum and the third party bottlers, Amata.”

133. Prior to using the Filler, the plaintiff employed the services of Amata to bottle its wines. When it bottled the Wines itself using the Filler, the plaintiff deviated from bottling practices used by Amata when it bottled the plaintiff’s wines prior to the use of the Filler, as follows:

133.1. Not protecting the wine in the filling bowl by means of an inert gas layer to keep oxygen away from the wine as much as possible.

133.2. Not sparging the bottles with inert gas to remove oxygen from them.

133.3. Not applying a vacuum at the time of corking.

133.4. Not ensuring that its staff was properly trained and experienced at bottling in such a way as to ensure minimum oxygen uptake during bottling.

134. Further of the plaintiff’s practices which may be relevant identified by the defendant in argument included:

134.1. Not determining that the SO₂ content of the wines was stabilised before bottling, in the sense of actually measuring them before bottling. This was a recommended practice, according to Ellis.

134.2. Not ensuring that the FSO₂ content was sufficient to protect the wines from oxidation, in the sense of actually measuring them before bottling.

135. Walser testified as follows:

135.1. Prior to using the Filler, the plaintiff used the services of Amata for bottling.

135.2. Amata is a specialist in bottling wine. It bottled for many winemakers.

135.3. The Amata mobile bottling unit was established by Martin Fourie, an ex-student of Ellis. Ellis helped with the planning because it has to be done correctly.

135.4. Amata used its own employees to bottle the wine. Amata had a team of employees for this specific purpose. They go from place to place to fill bottles. They use their own equipment. They apply management practices and controls to show how they have performed their tasks and have written records. Walser observed the bottling by Amata and supplied the wine, but did nothing else.

135.5. The plaintiff bought the Filler to do the bottling itself.

135.6. Amata trained employees over the years to do the work and to manage the machinery.

135.7. On the other hand, Johannes and Walser trained the plaintiff’s workers to operate the Filler. None had previous experience apart from possibly three days with a rented four head filler rental.

135.8. Amata used inert gases to reduce the risk of oxidation in the bottling process. The plaintiff did not do so when it used the Filler.

135.9. Amata’s employees cleaned their filler before, during (between different wines) and after bottling. Amata also cleaned the filler off-site. Plaintiff did this itself with the Filler.

135.10. Amata uses a vacuum corker to prevent or minimise oxygen ingress. On the other hand, Plaintiff does not have a facility to use a vacuum with which to remove air from the head space of the bottle. This means that air remains in the bottle.

135.11. Amata uses inert gases to reduce the possibility of oxidation. Plaintiff does not do so.

136. Archibald Smartryk Coetzee, who currently owns the Amata business (it is now called “iBottle” – I do not know if it is held in an entity, but that is of no moment) testified that the above practices were inherited from the previous owner of the business and are still applied.

137. Johannes was described as the plaintiff’s cellarmaster. Johannes was not called to corroborate the plaintiff’s case and there was no indication that he was not available to testify. The failure to call Johannes is of some significance because he is the person who Walser said:

137.1. Filled out the pink cards from which the bottling dates appear.

137.2. Helped with the Plaintiff’s bottling procedure and sulphur dosage, and should have been able to corroborate Walser’s evidence in this regard.

137.3. Helped to train the plaintiff’s staff in the operation of the Filler.

137.4. From time to time cleaned the Filler after use.

138. None of the employees who assisted with the SO₂ dosage testified.

139. Ellis himself testified that his first comment was that the changes in the wines which had spoilt were from bad bottling practices. He said that there are other possible reasons for the oxidation of wines, but it is normally the fault of the winemaker. Ellis testified that most of the deterioration of wines after bottling comes from the practices during and before bottling.

140. It is therefore not a simple matter of saying, as the plaintiff attempted to argue, that nothing changed in the bottling of the Wines. The fact is that much did.

141. In addition, not all of the wines (5 out of the 20 in question, being Smaug the Magnificent, Orbito, ESB, TBC and Jimmy) which passed through the Filler were not oxidised which supports the conclusion that it was not the Filler which resulted in oxidation.

142. While it is not necessary to determine what in fact caused the rust/oxidation on the Filler which was first observed more than four months after the plaintiff took possession thereof, both Cotton and Press testified that the disruption of the oxide layer can take place in a variety of ways, for example that imperfections to the oxide layer can be caused by physical contact with the surface or chloride in water used on the Filler (a factor favoured by Press), which in this case was slightly above the recommended limits. Cotton explained that breakdown of the passive layer depends on various factors such as environment temperature, time of exposure, Ph level and oxygen concentration in the electrolyte. It could also be that during the time it was in the plaintiff’s possession it was knocked in a manner which could have compromised the integrity of the stainless steel.

143. The consequence of all of the aforegoing is that, on the plaintiff’s own evidence, and that of its experts, dealt with above, were it to be necessary to decide the point, I am not in a position to find on a balance of probabilities that the rust in the Filler which would have come into contact with the Wines was caused by the inside bowl of the Filler not being properly pickled and passivated (and that the oxidation of the plaintiff’s wine complained of in this matter was caused by the Filler not being properly pickled and passivated).

144. This is an additional basis to the main bases considered in sections E and F above for the plaintiff not having made out a case.

H: The plaintiff’s late argument on shelf life

145. In oral argument in reply (during the last hour of a trial lasting over 20 days), Mr Coetzee raised the point, as I understand it, that an increase in iron in wine which would oxidise in any event would result that that wine would have a lesser shelf life and that this was a breach. In other words, that the uptake of iron/rust caused wine which would have oxidised in any event to have oxidised quicker, which is itself a breach.

146. No such case is pleaded by the plaintiff.

147. The pleaded case is that wine was oxidised, caused by the uptake of iron particles, not that oxidised wine is oxidised quicker resulting in a lesser shelf life which causes a loss.

148. A further problem is that no case is made out as to the different shelf life of the wine which would result and the materiality thereof: unsurprisingly, in the context of this after-thought, one is left in the dark as to whether it would be a matter of days, a week, a few weeks, a month or a few months or otherwise.

149. A consideration of the evidence which there is indicates that no case would be able to be made out for the following reasons: the evidence dealt with above established that oxidation would take between 8 weeks and 12 weeks depending on the level of the iron (from 1mg/l to 5 mg/l), which is an immaterial difference (and excessively in the plaintiff’s favour because the iron levels averaged 2.9mg/l, as dealt with above).

150. Danilewicz testified that a wine with sufficient oxygen (or insufficient SO_2,whichever way one looks at it) will oxidise, whether with an iron level of 1mg/l or 5 mg/l or otherwise. The amount of iron does not impact on the amount of oxidation: it will remain the same. The amount of oxidation depends on the amount of oxygen. The amount of iron only impacts the rate at which oxidation occurs. The wine with 5 mg/l would oxidise quicker, in about 8 weeks. With 1mg/l it would take about 12 weeks. These are actual examples, being the plaintiff’s wines 10 and 14 as numbered by it which were analysed by Danilewicz. As mentioned, this is an immaterial difference (and excessively in the plaintiff’s favour because the iron levels averaged 2.9mg/l, as dealt with above).

151. That is in the short rapid first phase and in these (extreme, from the perspective of iron levels, which is what is in issue) examples it would be over within weeks of each other. When the wines are sent to the retailer, both would appear oxidised.

152. Sight has not been lost of the two phases of oxygen exposure. Iron is involved in the initial rapid phase, not the later slow phase of oxygen passing through the cork at the Oxygen Transmission Rate (OTR).

153. I am therefore of the view that this point is of no merit.

I: Conclusion on the merits (leaving aside the legal question as to damages dealt with below)

154. I am of the view that the plaintiff has not established on a balance of probabilities that the Filler not being properly pickled and passivated and as a result becoming corroded was the cause of the wine becoming oxidised.

155. I am of the view that the plaintiff has not established on a balance of probabilities that iron or rust uptake from the Filler spoilt the Wines.

156. The plaintiff’s claim will therefore be dismissed.

J: The damages claimable in law

157. The plaintiff’s claim is for contractual damages.

158. The debate between the parties concerned two questions of law:

158.1. Whether the damages claimed flow naturally from the breach (also known conveniently as general damages or intrinsic damages: Shatz Investments v Kalovyrnas 1976 (2) SA 545 (A) at 550BC), special damages no longer being claimed, as referred to above.

158.2. Whether consequential damages could, in law, be claimed at all in the circumstances.

159. Both of these aspects are hurdles which the plaintiff must overcome to be in a position to claim the damages which it does in this matter.

(1) General damages

160. These damages are for loss which flows naturally and generally from the kind of breach which the law presumes the parties contemplated as a probable result of that breach. As held in Holmdene Brickworks (Pty) Ltd v Roberts Construction Co Ltd 1977 (3) SA 670 (A) at 687, they are:

“… damages that flow naturally and generally from the kind of breach of contract in question and which the law presumes the parties contemplated as a probable result of that breach …”

161. Mr Duminy submitted as follows: the standard measure of general damages is to make good the bargain which, in this instance, is the difference between the market value of the item sold and the market value of the item delivered. Where an agreement is rescinded for breach of contract, the measure of damages is the difference between the contract price and the market price at the time of performance. Examples of special damages include loss of profit on resale, loss of reputation and loss of trade.

162. The above-mentioned principles do not appear to me to be contentious.

163. The plaintiff’s main remaining claim is for R3 923 232.78 for the difference between the value of the wine it withdrew from the market, less saved costs. Mr Duminy submitted that this was a claim for loss of profit which is ordinarily regarded as a claim for special damages and not damages which flow naturally from the breach (general damages) (Shatz Investments v Kalovyrnas 1976 (2) SA 545 (A) at 550GH).

164. I have underlined the word ordinarily because it indicates in unambiguous terms that this is not an absolute rule. What I understand from this is that the legal conclusion as to whether damages flow naturally from the breach depends on the facts and circumstances of the matter.

165. Assuming that Mr Duminy is correct that the above-referred claim is for loss of profit, what must be considered is whether on the facts and circumstances of this case it is excluded from general damages.

166. The Filler was purchased by Walser Wine Projects (Pty) Ltd from Wine Machinery Group (Pty) Ltd. Its purpose was to bottle wine. It is common cause that it should not spoil wine if properly used. It would be a breach of a term of the contract of sale were the wine to be spoiled even if the Filler was properly used. The loss of that wine would flows from that breach. I have underlined the word wine because its involvement is core to the contract and the agreed intended use of the Filler.

167. I think that these facts and circumstances provide the answer in the affirmative in the plaintiff’s favour as to the damages flowing naturally from the breach of the contract.

(2) Consequential damages

168. There was no debate between the parties that the plaintiff’s claim is for consequential loss.

169. It is admitted on the pleadings that the defendant is a merchant who sells products related to wine-making, including bottling equipment.

170. In Kroonstad Westelike Boere Ko-Op v Botha 1964 (3) SA 561 (A) it was held as follows:

“… there is insufficient judicial support for the wide view that a merchant, who sells goods in which it is his business to deal, is merely on that account liable for consequential damages caused to the purchaser by a latent defect, of which the seller was unaware, in the thing sold.” [10]

“In my opinion the preponderant judicial view, and which this Court should now approve, is that liability for consequential damage caused by latent defect attaches to a merchant seller, who was unaware of the defect, where he publicly professes to have attributes of skill and expert knowledge in relation to the kind of goods sold. … Whether a seller falls within the category mentioned will be a question of fact and degree, to be decided from all the circumstances of the case.” [11]

171. The defendant argued that the plaintiff’s contractual claim cannot succeed unless it proves that the defendant was a merchant seller who publicly professed to have attributes of skill and expert knowledge in relation to fillers such as the Filler.

172. The plaintiff tendered no evidence in support of the necessary allegation that the defendant publicly professed to have attributes of skill and expert knowledge in relation to the kind of goods sold.

173. The plaintiff relies on the admission that the contract of sale contained an implied term of fitness for purpose and submits that this satisfies the requirements articulated in Kroonstad.

174. The defendant submits that this is not the correct reflection of the law and argues that the breach of the implied warranty against latent defects that would render an item sold useless or partially useless for the purpose for which it was sold is a ‘redhibitory defect which underlies the Aedilitian remedies (the actio quanti minoris and the actio redhibitoria).

175. Liability for consequential loss, however, for breach of the implied warranty against latent defects, is confined to cases that satisfy Kroonstad. In Holmdene Brickworks it was held at 682-683 as follows:

“The legal foundation of respondent's claim is the principle that a merchant who sells goods of his own manufacture or goods in relation to which he publicly professes to have attributes of skill and expert knowledge is liable to the purchaser for consequential damages caused to the latter by reason of any latent defect in the goods. Ignorance of the defect does not excuse the seller. Once it is established that he falls into one of the above mentioned categories, the law irrebuttably attaches this liability to him, unless he has expressly or impliedly contracted out of it. (See Voet, 21.1.10; Pothier, Contrat de Vente, para. 214; Kroonstad Westelike Boere Ko-op. Vereniging v Botha, 1964 (3) SA 561 (AD); also Bower v Sparks, Young and Farmers Meat Industries Ltd., 1936 NPD 1; Odendaal v Bethlehem Romery Bpk., 1954 (3) SA 370 (O).) The liability is additional to, and different from, the liability to redhibitorian relief which is incurred by any seller of goods found to contain a latent defect (see Botha's case, supra at p. 572).”

176. A claim for special damages which is not based on ‘what is articulated in Kroonstad carries with it the necessary elements of contemplation and agreement required to be proved for such a claim. This basis for the claim was abandoned, as mentioned above, which appears to me to have been a correct decision on the evidence. Walser confirmed that the little that passed between the parties concerned only the speed of filling, in respect of which there is no complaint:

“MR DUMINY: According to your pleadings, you had discussions with him over a period from February 2018 to November 2019 ?

MR WALSER: Yes. So the discussion was all about the speed of filling.

MR DUMINY: Yes.

MR WALSER: There was only a six head filler available, so it is got six spouts and I was not sure if I needed a four - spout one or a six- spout one. So discussions was all about the flow and the rate of flow. That was the discussions regarding the filling.

MR DUMINY: Alright and you had, sometime later in November 2019, you experimented with a four - head filler and could see its speed. So you had a basis for comparison?

MR WALSER: That is correct.

MR DUMINY: But that is really as far as it went?

MR WALSER: Yes.”

177. The defendant contends that for this independent reason, the plaintiff’s contractual claim should be dismissed, with costs.

178. In the light of the decisions above on the merits of the matter, this question need not be decided. Mr Duminy submitted, in my view correctly, that unless Kroonstad is satisfied, consequential damages cannot be claimed. Mr Coetzee relied on Amler’s Pleadings 10ed at 337 which set out three circumstances in which damages can be claimed. He relied on Minister van Landboutegniese Dienste v Scholtz 1971 (3) SA 188 (A) and Kroomer v Hess 1919 AD 207. In my view, his argument is not supported by these cases.

179. Kroonstad was concerned with the consequences of a contract of sale in certain factual circumstances. They arise where there is a contract of sale, the merx is latently defective, the seller is not aware of the defect and has publicly professed skill and expert knowledge in relation to the kind of goods sold. The implication or imputation of the “term” underlying liability depends on those facts being proven. In that sense, it could be classified as a tacit term. It is imposed as a matter of law, and in that sense it would be equally valid to regard it as an implied term. Neither Kroonstad nor any of the later decisions on the point, attribute different consequences or apply the Kroonstad requirements differently depending on whether the term is regarded as implied (by law) or tacit (i.e. by agreement).

180. My view, were this issue need to be decided, is that the requirements in Kroonstad were not satisfied and therefore that the argument of the defendant is to be preferred. While this may appear harsh to a purchaser, it appears to me to be the state of the law and would be another basis for the claim to fail.

Costs

181. The defendant requested the qualifying and preparation fees of all of the expert witnesses for whom summaries had been delivered at its instance, which included three who did not testify. I indicated to Mr Duminy that I do not consider that such costs can be awarded without more. He argued that while Van Rensburg did not testify, his report was mostly accepted by Ellis which disposed of many disputes. Accordingly, even though he did not testify, his costs should be allowed. This however, only applied to Van Rensburg’s first report and not his second report. Further, there were some aspects of the first report which were not admitted. Van Rensburg had also refused to attend a meeting with Ellis, stating (as recorded in the ‘joint minute’ prepared by Ellis (he was the sole participant) that Van Rensburg said “I have my opinion and you have yours.” I do not regard that as acceptable. In my view, only the costs of the witnesses who actually testified for the defendant should be awarded.

182. There was no dispute that the employment of two counsel was warranted and that scale C in terms of Rule 67A should apply, the matter being complex.

Order

183. The following order is granted: The plaintiff’s claim is dismissed with costs, including the costs of two counsel where so employed, with scale C in terms of Rule 67A applying, and including the qualifying and preparation fees of the expert witnesses Danilewicz and Press.

A Kantor

Acting Judge of the High Court

[1] See par [25], footnote 4. This was because section 37C and Rule 38(9) had not yet been enacted.

[2] At par [3].

[3] At par [38] – [45]. See also the Judge’s remarks in Folley v Pick n Pay Retailers (Pty) Ltd & Others [2017] ZACHC 86 (23 August 2017) at par [20]. In addition to the cases already quoted, there are others dealing with remote hearings, mostly in the context of the Covid pandemic, such as Krivokapic v Transnet Ltd t/a Portnet [2018] 4 All SA 251 (KZD); Union-Swiss (Pty) Ltd v Govender and Others 2021 (1) SA 578 KZD; RVD v IJDV 81157/2017, 24 March 2021 Gauteng Division; and Puma Sports Distributors (Pty) Ltd v Hughes and Others 1820/18, 10 November 2020, Western Cape Division.

[4] A good overview of recent developments in this field is to be found in Hafez, M ‘Remote hearings and the use of technology in arbitration.’ Global Arbitration Review: The Middle Eastern and African Arbitration Review 2023. The article can be accessed at:

https://globalarbitrationreview.com/review/the-middle-eastern-and-african-arbitration-review/2023/article/remote-hearings-and-the-use-of-technology-in-arbitration

[5] 2020 FCA, a judgement of the Federal Court of Australia-New South Wales. This case can be accessed at: https://www.austlii.edu.au/cgibin/viewdoc/au/cases/cth/FCA/2020/486.html?context=1;query=capic%20ford%20motor%20company;mask_path=

[6] [2020] EWHC 854 (CH), a judgement of the Chancery Division of the High Court of England and Wales. This case can be accessed at https://www.bailii.org/ew/cases/EWHC/Ch/2020/845.pdf. For a later evaluation by the same Judge of his experience of the use of the technology, see the same case name but with citation [2021] EWHC 684 (Ch) at par [25] (which can be found at https://www.bailii.org/ew/cases/EWHC/Ch/2021/684.pdf): “In summary, the remote hearing proved to be more than a second-best work around in the face of the Covid 19 pandemic. The fully remote mode of trial certainly created real challenges for the parties and their representatives, not least because of the short period to make adjustments in the preparations for trial. However, my overall assessment is that not only were those challenges overcome by appropriate and mutually agreed adjustments on the part of counsel, the parties and court but that the trial was conducted more efficiently and far more conveniently as a fully remote trial. It was also more accessible to the public than it would have been had it taken place in a traditional court room in the Rolls Building.”

[7] Superior Court Practice Vol 2, D1 Rule 22-8.

[8] Vinlab is a facility which conducts laboratory tests on wine which were accepted by the parties as being reliable.

[9] Identified by Press. Both he and Cotton testified as to welding and the heat involved therewith being a factor in the compromise of stainless steel.

[10] At 571DF. See also Holmdene Brickworks v Roberts Construction 1977 (3) SA 670 (A) at 682-683; Ciba-Geigy (Pty) Ltd v Lushof Farms (Pty) Ltd en 'n Ander 2002 (2) SA 447 (SCA), par [48] – [50].

[11] At 571GH.