THE SUPREME COURT OF APPEAL

OF SOUTH AFRICA

 

Case number : 427/01

Reportable in part*

 

In the matter between :

 

STELLENBOSCH FARMERS’ WINERY

GROUP LIMITED FIRST APPELLANT

STELLENBOSCH FARMERS’ WINERY

LIMITED SECOND APPELLANT

 

and

 

MARTELL & CIE S A FIRST RESPONDENT

MARTELL ET CIE (SA) (PTY) LTD SECOND RESPONDENT

SEAGRAM AFRICA (PTY) LTD THIRD RESPONDENT

JOSEPH E SEAGRAM & SONS INC FOURTH RESPONDENT

 

 

 

CORAM : NIENABER, FARLAM, BRAND JJA, HEHER and LEWIS AJJA

 

HEARD : 16 AUGUST 2002

 

DELIVERED : 6 SEPTEMBER 2002

 

Summary: Oral agreement - dispute as to its terms - approach to be adopted in resolving disputes of fact - analysis and evaluation of probabilities

 

JUDGMENT

 

* Para [5] is reportable within the context of a broad summary of the facts.

NIENABER JA/

NIENABER JA :

[1] Recollection can be fallible. And in business the failure to confirm an event promptly and on paper can be fatal. If ever a case proved these propositions, this was it. The core dispute between the parties could not have been narrower. It was whether the respective representatives of the two parties, on 17 April 1997 and in the appellants’ boardroom at Stellenbosch, orally agreed on a sales volume target for the ensuing fiscal year of 5317000 or 5366000 litres of Martell cognac-style brandy. The appellants were producing, promoting and selling Martell brandy in the southern African market under franchise from the respondents. Because of a formula in their agreement huge sums of money would in future, as it happened, turn on this marginal discrepancy between the two figures. One side confirmed its understanding of what was orally agreed between them and the other side did not. And that, in the end, after much was said and done, makes the difference.

[2] There are four respondents, all belonging to the Seagrams group of companies, who figured as plaintiffs in the Court below and two appellants, both belonging to the Stellenbosch Farmers’ Winery group, who were the defendants, but nothing in the present case depends on their respective corporate identities and for the sake of convenience I shall henceforth refer to the two groupings simply as ‘Seagrams’ and ‘SFW’. During 1988 and at a time when foreign companies found trading in South Africa uncomfortable, Seagrams concluded a written agreement with SFW in terms of which the latter would manufacture and market three qualities of Martell brandy in southern Africa, namely Martell Five Star, Martell VO and Martell XVO Classique. SFW would be furnished with certain secret information, technical assistance and flavouring ingredients and would pay Seagrams royalties calculated on volumes of sales. This agreement was due to terminate in June 1992.

[3] During July 1990 and at New York a further agreement, described in argument as an ‘evergreen and rolling agreement’, was concluded, the express terms of which were incorporated into a document termed ‘summation’. It introduced the concept of ‘annual sales objectives’ on which, year by year, the parties would have to agree in advance. Clause 1 reads as follows:

‘1. Duration

1.1. 5 Year rolling contract with automatic annual extension if annual sales objectives are achieved.

1.1.1. 90% achievement of annual sales objective still implies the above.

1.1.2. Anything less than 90% achievement of annual sales objective implies that one year of the 5 year contract is lopped off.

1.1.3. Should annual sales objective be achieved in the subsequent year it would mean immediate re-instatement of 5 year period as per 1.1.

1.1.4. If two parties cannot agree on the annual plan, the prior year’s share of market percentage will be used as the base.

1.1.5. Plan will be constructed by category (i.e. V.O.5 Star) but contract will work on a total basis only.

1.1.6. Achievement of anything below 80% of annual sales objective would give Seagram the right to renegotiate the agreement unless it was due to unforeseen prevailing circumstances.’

 

[4] A meeting duly took place on 17 April 1997 and it was common cause that agreement was reached on the ‘annual sales objective’ for the next successive fiscal year commencing on 1 July 1997. The critical question, as stated earlier, was: what was the actual figure on which agreement was reached? Was it 5317000 litres (as SFW contends) or 5366000 litres (as Seagrams contends)? What was also common cause between the parties was that the actual sales for the fiscal year July 1997 to June 1998 were 4284748 litres. If the target figure agreed upon more than a year earlier was 5317000 it represented 80,58% of the agreed annual sales objective and the agreement would carry on uninterruptedly; but if it was 5366000 it represented an under-achievement by SFW equivalent to 79,859% and would provide Seagrams with the escape from the agreement it had long sought. Seagrams found the agreement to be constricting because it inhibited it from exploiting its three brand names itself in southern Africa. If SFW under-achieved it would enable Seagrams to insist on a renegotiation of the entire agreement. In that event it was in the end not disputed by SFW that a further tacit term of the agreement would permit Seagrams, if SFW should refuse to renegotiate the terms of the agreement, to terminate it by giving SFW reasonable notice to that effect. And that is precisely what happened: Seagrams demanded a renegotiation; SFW disputed its entitlement to do so; Seagrams gave notice of the termination of the agreement with effect from 30 June 1999 and thereafter sued SFW for a declaratory order in the Cape High Court that the agreement had come to an end on that date. The matter eventually came before Traverso DJP who found in favour of Seagrams. This is an appeal, leave to pursue it having been refused by the Court a quo but granted by this Court, against that finding.

[5] On the central issue, as to what the parties actually decided, there are two irreconcilable versions. So too on a number of peripheral areas of dispute which may have a bearing on the probabilities. The technique generally employed by courts in resolving factual disputes of this nature may conveniently be summarised as follows. To come to a conclusion on the disputed issues a court must make findings on (a) the credibility of the various factual witnesses; (b) their reliability; and (c) the probabilities. As to (a), the court’s finding on the credibility of a particular witness will depend on its impression about the veracity of the witness. That in turn will depend on a variety of subsidiary factors, not necessarily in order of importance, such as (i) the witness’s candour and demeanour in the witness-box, (ii) his bias, latent and blatant, (iii) internal contradictions in his evidence, (iv) external contradictions with what was pleaded or put on his behalf, or with established fact or with his own extracurial statements or actions, (v) the probability or improbability of particular aspects of his version, (vi) the calibre and cogency of his performance compared to that of other witnesses testifying about the same incident or events. As to (b), a witness’s reliability will depend, apart from the factors mentioned under (a)(ii), (iv) and (v) above, on (i) the opportunities he had to experience or observe the event in question and (ii) the quality, integrity and independence of his recall thereof. As to (c), this necessitates an analysis and evaluation of the probability or improbability of each party’s version on each of the disputed issues. In the light of its assessment of (a), (b) and (c) the court will then, as a final step, determine whether the party burdened with the onus of proof has succeeded in discharging it. The hard case, which will doubtless be the rare one, occurs when a court’s credibility findings compel it in one direction and its evaluation of the general probabilities in another. The more convincing the former, the less convincing will be the latter. But when all factors are equipoised probabilities prevail.

[6] In the instant case the Court a quo commended Fleck and Kikillus on Seagrams’s side, and criticised Msiza, in particular, on SFW’s side, dismissing parts of his evidence as ‘absurd’. But this appraisal does not seem to have depended on an analysis of the various factors enumerated in the previous paragraph but largely on the Court a quo’s estimation of the overall probabilities. If that estimation is shown to be suspect, so too must be the Court a quo’s conclusions on credibility. It is therefore on that exercise, an evaluation of the general probabilities, that the outcome of this case ultimately hinges.

[7] I propose to discuss the probabilities with reference to various successive phases as events unfolded between the parties during the period under discussion.

The meeting of 17 April 1997

[8] Before dealing with the meeting itself it is opportune to say something about the events immediately preceding it. The previous meeting, at which the target volume for the period 1996/1997 was agreed, was held in October 1996, during the then current fiscal year, when the annual sales objective was settled at 5604000 litres for the three brands. (SFW’s official budget figure for the year was 5574000 litres.) Both sides acknowledged the agreement in writing. Bullen, one of SFW’s marketing people, confirmed the figure in a letter to Seagrams and Kikillus, Seagrams’s marketing director, prepared a minute, also confirming the agreed figure, which he forwarded to SFW. During January 1997 Kikillus wrote a letter to Bullen asking for an estimate of sales for the coming year and suggesting that a meeting be convened to agree on a target figure for the 1997/98 year. Bullen responded by giving him an estimate, based on SFW data up to December 1996, of 5884000 litres. The meeting was then convened for 17 April 1997. By that time, however, the estimate of 5884000 litres was no longer realistic. The sales figures for January, February and March reflected a marked decline in sales for all three brands over that period as was apparent from the monthly reports which SFW had routinely forwarded to Seagrams. It has always been the practice for SFW to furnish Seagrams with monthly reports of actual sales, broken down into period, brand and region. This was of direct interest to Seagrams since royalties were calculated on sales. These reports, so it was contended on behalf of SFW, should have forewarned Seagrams’s representatives that the figure of 5884000 supplied by Bullen was outdated and hence no longer reliable.

[9] As to the meeting itself there was a good deal of common ground. Henning, the senior man, Visser, Msiza and Holzkampf represented SFW and Seagrams was represented by Fleck, its managing director, and Kikillus. Henning opened the proceedings by introducing Msiza. Msiza put before the meeting a compilation of statistical data referred to as ‘the Blue Book’ about the past performance of the Martell brands. For ease of reference the first page thereof is annexed to this judgment as annexure A. The supporting statistics furnished showed a significant decline in sales since December 1996. That was also the point of certain graphs prepared by Holzkampf and projected on to a screen. The Blue Book contained no proposal on the next annual sales objective but it did suggest an increase in advertising spending of some 16%. (The cost of advertising was by agreement shared between SFW and Seagrams.)

[10] Msiza testified that in view of the poor performance of the brand during recent months he proposed what he called a ‘flat figure’ of 5217000 litres as the new annual sales objective. This is the figure appearing in annexure A as the estimate for 1996/97, representing an expected increase of 1,3% for the year ending June 1997, compared to the figure for the fiscal year ending June 1996 which was 5148000. Msiza’s proposal implied that there would be no growth during the coming year. This proposal, all the witnesses agreed, was instantly rejected by Seagrams’s representatives who pointed out that they would be unable to justify, to their principals overseas, a target reflecting zero growth in sales if there was to be a simultaneous increase of 16% in spending on advertising. So much, then, for what was common ground. I turn to what was in dispute.

[11] Seagrams sought support for its version (that there was agreement on 5366000), in a document headed ‘Latest Forecast’. Fleck and Kikillus testified that each of them was given a copy thereof. I annex, as annexure B, a copy of the one given to Fleck. On it there are two sketches, one of Henning and one of Visser, which, so Fleck testified, he drew while the meeting was in progress. On Kikillus’s copy there are annotations representing his contemporaneous conversion of thousands of litres into cases.

[12] SFW, on the other hand, sought support for its version (that 5317000 was the agreed figure), on a document, a copy of which I annex as annexure C, which is a copy of the first page of the Blue Book, with some calculations on it in Msiza’s handwriting. Each side claimed that the document on which it relied corroborated its version and that the document on which the other side relied did not figure at all in the discussions during the meeting.

[13] According to Fleck and Kikillus they went into the meeting with Bullen’s forecast of 5884000 litres in mind. But after considering the later statistics and Holzkampf’s presentation on a screen of his graphs showing the dramatic tailing off on sales after December, they realised that the figure Bullen mentioned to them was no longer achievable. They were accordingly prepared to compromise on a lower figure. That lower figure was SFW’s own budget forecast for the 1997/98 fiscal year. It was 5366 [thousand]. It is a figure that can be derived from annexure B as the sum of 1450 + 3900 + 16, the ciphers appearing on annexure B under the column ‘budget forecast’ for the period 1997/98. (At that point the budget had not yet been formally approved for the 1997/98 year by SFW’s board of directors. Such approval would only follow in June 1997.) If that, so Fleck and Kikillus explained, is what SFW itself regarded as the forecast of sales for the period concerned why should Seagrams be prepared to settle on a lower figure? Because of the concession on their part, to reduce the target figure from 5884000 to 5366000, it was further agreed between all those present, so they said, that part of the advertising expense be reserved for reconsideration later in the fiscal year should the progress on sales not measure up to expectations. (For that evidence there was confirmation in a note to that effect by Msiza in the Blue Book.)

[14] Neither Fleck nor Kikillus made a note of the figure 5366 or, for that matter, of the agreement regarding the increase in the ‘ad-spend’ on their respective copies of the Latest Forecast document.

[15] It is necessary to say something about this document. It was formally discovered by Seagrams and not by SFW. Its very existence was initially denied by SFW’s witnesses. In particular it was denied that annexure B was ever produced at the meeting or discussed as such. All SFW’s witnesses who testified were sceptical about its provenance, first, because these were at that stage still confidential figures and, secondly, because of the heading ‘budget forecast’ which was a term that, according to them, was inherently contradictory and was never used within SFW circles. Of course, if that evidence stood, it could well have been destructive of Seagrams’s case since the figure 5366, being a confidential SFW figure, appeared nowhere else in the documentation presented at the meeting. But during the course of the trial Seagrams produced expert evidence that the Latest Forecast document with the detailed information it contained could only have emanated from SFW. And this was eventually conceded by SFW. That the two documents were received by Fleck and Kikillus on that day is moreover proved by Fleck’s drawings and by Kikillus’s handwriting on them. SFW’s witnesses, after having at first denied that the document was ever produced at the meeting, could ultimately do no better than to say that they were not responsible for its production and that none of them had ever seen or heard it being discussed during the one and a half hours or so the meeting lasted. It was not a routine SFW document and it must have been produced especially for the meeting that afternoon. The evidence was that the SFW’s representatives had not met before the meeting in order to discuss strategy. Nor was the meeting conducted according to a formal agenda. Discussions must have taken place amongst those present without the knowledge of others. The only rational explanation is that Henning had the document prepared in advance without the knowledge of the others and that he distributed the only two copies to Fleck and Kikillus, at a time when none of the other SFW witnesses was paying close attention to what was passing between them. Henning presumably did so in order to demonstrate the downward trend of sales during the few months immediately preceding the meeting. He did not testify. He had died before the trial commenced.

[16] Much was made of this retreat on the part of SFW by both the Court a quo and counsel for Seagrams. And indeed, the very fact that its authenticity was initially disavowed does create the impression that the document was deliberately suppressed by SFW; and consequently, that it contained information that was embarrassing to its cause. In truth the document contained nothing that was per se awkward for SFW. Its immediate significance went the other way: its absence would have been an embarrassment for Seagrams’s case since it was the only document reflecting, even if only arithmetically, the figure of 5366.

[17] The Court a quo found, correctly so, that annexure B was indeed produced at the meeting. And if that is so it does create a probability in favour of Seagrams. For if SFW had itself budgeted for that figure, albeit provisionally at that stage, it would conceivably have been more difficult for it to persuade Seagrams’s representatives to agree to an even lower figure. That is a point in its favour. The agreed figure, after all, would almost certainly be a compromise based on SFW’s anticipated sales during the coming year. Nor was it an upper limit. It still left SFW with a margin of slack of 20% in terms of clause 1.1.6 of the 1990 agreement in order to protect its contractual stranglehold on the brand. On the other hand, it is also true that neither party was compelled to capitulate and agree to an unrealistic target since clause 1.1.4 of the 1990 agreement allowed for a default figure if no compromise could be reached. (That default figure, calculated in later correspondence by SFW’s representatives to be 4790465 litres, was lower than both the two figures now in contention.)

[18] What about the document on which SFW relies, annexure C? Msiza’s version is as follows. Annexure C was part of the Blue Book upon which he made certain annotations, so he said, as the discussions progressed. When his suggestion of a ‘flat figure’ of 5217 was rejected, Fleck made it plain that Seagrams would at the very least insist that the improvement of 1,3% (which was the improvement of the 1996/97 estimate over the 1995/96 year) be maintained for 1997/98. The 1.3% appears on annexures A and C. Msiza then calculated the figure that would be yielded if a 2% growth is factored into the equation on a base of 5217 [thousand]. That calculation produced 5321 [thousand], which is noted on annexure C. Holzkampf was then asked to project onto the screen how his previous three graphs for the three individual types would have to be adjusted if the target figure was shifted to a total sales figure of 5321. This required a breaking down into the three types of the total of 5321. Msiza attempted his own breakdown. On annexure C appears the annotation:

1350 3850 15

5215

 

 

This breakdown came short of the total of 5321. A discussion followed amongst all those present and out of this discussion a new breakdown was suggested which Msiza noted down on annexure C. It reads:

1430 3875

5305 12

5317

 

 

 

 

This breakdown was then reflected on Holzkampf’s graphs per type. And that was the figure eventually agreed upon as the annual sales objective for the fiscal year 1997/98. Msiza testified that he then blocked in the figure of 5317, as indeed appears on annexure C. Elsewhere in the Blue Book he also made a note ‘5% reserve into 2nd six months’ which refers to the agreement, not disputed, about the ‘ad-spend’ figure. Visser and Holzkampf, although neither could remember the actual figures as such, confirmed the course of events as related by Msiza. Visser also testified that he too may have made notes but that all his documentation was destroyed when he was later transferred to a post abroad.

[19] One is thus left with two versions, each tailored to a particular document which the other side does not recognise as having been discussed at the meeting. In my view one must accept that both documents featured at the meeting and that each side’s denial of the other side’s document shows that the respective witnesses’ evidence was largely reconstructed rather than recollected. Since neither reconstructed version is inherently contradictory one is driven, once again, to revert to the probabilities. In my view the probabilities relating to the meeting itself favour SFW’s version. I say so for the reasons that follow.

[20] As far as Seagrams’s version is concerned it is unlikely, if the figure of 5366 had been agreed to with specific reference to the Latest Forecast document, that Fleck or Kikillus would not have made a note of it on the document of which each had a copy, given that each of them scribbled something on his own copy. Admittedly they also did not note down the figure 5317. That presumably shows that they left it to SFW to do the necessary follow-up paperwork. Moreover, as counsel for SFW was at pains to stress, when Seagrams in August 1998 did commit themselves for the first time to the figure they alleged was agreed, it was not to 5366 but to the full SFW budget figure of 5365715 that reference was made. This latter figure could only have been taken from a telefax sent to Kikillus a year earlier on 11 August 1997 when Msiza, at his request, furnished Kikillus with the 1997/98 ‘budget figures’. Even on their own showing the more exact figure was not available to them at the meeting on 17 April 1997.

[21] Whereas the Seagrams’s figure of 5366 does not as such appear on annexure A, SFW’s figure of 5317 does appear on annexure C. I have some difficulty with the submission addressed to us on behalf of Seagrams that these were mere scribblings jotted down by Msiza but which had nothing to do with what was discussed. In my view these scribblings do serve as an indication of what counsel for SFW referred to as ‘the route’ the discussions took in boosting the figure from 5217, initially proposed by SFW, to 5317. As such it does lend support to SFW’s version. That version is further supported by the reason SFW’s witnesses gave for their willingness to agree to an upward adjustment of the figure. It was a concession made to Seagrams for the latter’s willingness to agree to the proposed increase in the spending for advertising, for a portion of which Seagrams would be liable. Fleck and Kikillus agreed to a 16% increase in the advertising spending, but only if the volume target were lifted above the 101,3% level and on condition that a percentage of the spending on advertising be reviewable after six months. This was the gist of the notes Msiza made on his Blue Book. In sum, therefore, I believe that the balance of probabilities as to what occurred at the meeting of 17 April 1997 favours SFW’s cause rather than Seagrams’s.

The letter of confirmation of 29 April 1997.

[22] Contrary to what happened the previous year no minute of the meeting was prepared but a letter of confirmation was written by Msiza and faxed to Seagrams on 29 April 1997, some twelve days after the meeting. That letter is annexed hereto as annexure D. It confirms both the agreement as to advertising spending and SFW’s figure, in thousand of litres, of 5317. In itself this letter constitutes a powerful probability in SFW’s favour. It was written shortly after the meeting when the figure would have been fresh in Msiza’s mind. It confirms a figure which Seagrams’s witnesses say was never even mentioned at the meeting. This was at a time when the differential between 5317 and 5366 was marginal and not remotely controversial. Why would Msiza conjure up an imaginary figure? And if it was simply a mistake, how could he have made it? Unlike other documents sent by SFW to Seagrams by Msiza, this particular letter was not copied to Henning and Visser. Much was made of this fact in argument on behalf of Seagrams. The omission was explained by Msiza and Visser and the point of the criticism loses much of its force since it was common cause that the letter was indeed received by Seagrams. More importantly, the content of the letter of confirmation gains greatly in plausibility when Seagrams’s response, or rather lack of a proper response, is taken into account.

Seagrams’s response to the letter of confirmation

[23] According to Kikillus he immediately realised, on receiving the fax, that the agreed figure was incorrectly stated therein. He mentioned it to Fleck in passing. Fleck instructed him to have it corrected. But instead of telephoning or faxing Msiza, as one would have expected him to do, he decided to wait until he would see Msiza on 5 May 1997 at a marketing meeting in Stellenbosch. Kikillus went to the meeting without taking along either the letter of confirmation, annexure D, or the Latest Forecast document, annexure B. Late in the day he mentioned to Msiza that the figure in the latter’s letter of confirmation was wrong, without, however, stating in what respect it was wrong or what the correct figure should be. Msiza, according to him, then promised to rectify it ‘in the next fiscal’ by which he presumably meant during July 1997. Kikillus did not mention the problem of the correction of the figure to Visser, who was also in attendance on that occasion, since he did not want to cause embarrassment to Msiza who was still, so he explained, relatively new to the job. And that, according to him, is where he left matters, in the hands of Msiza to correct it during the forthcoming fiscal year.

[24] Msiza, while admitting that he attended the meeting and that Kikillus was present, denied that such a conversation had ever taken place. Kikillus’s evidence on this whole episode, I am sorry to say, is far from convincing. It is, in my view, incomprehensible that he did not mention what he regarded as the correct figure to Msiza when he wanted the latter to rectify it and that Msiza promised to correct something without having been told what was wrong with it and how it should be rectified; so too, that Kikillus did not insist that Msiza should forthwith substitute a corrected letter of confirmation for the supposedly incorrect one of 29 April. Nor is it plausible that he should be willing simply to leave matters in Msiza’s hands to attend to it only when he would be furnishing statistical data in the coming fiscal year - which would in any event have created for Msiza the problem that the later so-called corrected figure would be in conflict with the uncorrected figure earlier stated in his letter of 29 April. One would have expected Kikillus, having been instructed by Fleck to sort the problem out without delay, to have asked Msiza to confirm in writing that his earlier letter was wrong, and, when the latter failed to do so, that Kikillus would himself have confirmed Msiza’s undertaking to rectify the matter. None of this happened. He waited and yet, when Msiza eventually did send the new sales figures for July 1997 on 5 August 1997, there was no mention of either the promise to correct his letter of confirmation nor of the alleged agreed target figure. It was only then, on 6 August 1997, that Kikillus, thanking Msiza for the information he forwarded, faxed him a letter in return, without, however, referring to Msiza’s earlier promise to correct matters, stating: ‘Unfortunately, plan is not reflected in your volume breakdown. Please could we include this with immediate effect.’ By ‘plan’, it is common cause, Kikillus meant the annual sales objective.

[25] Msiza’s evidence was that he telephoned Kikillus on receipt of the letter because he was puzzled by it, since this was the first time in the history of the relationship between the parties that such a request had ever been made. Kikillus explained to him that he wanted a breakdown into brands and regions of the ‘Plan’ figure. Msiza thereupon explained to Kikillus, so he said, that the only statistics broken down in these categories he had at his disposal were the SFW’s budget figures (which by then had been approved by the board). He promised to furnish this information to Kikillus and he did so on 11 August. It reflected a breakdown into brands and regions of the budget figure of 5365715. (It was on this information, as stated earlier, that Seagrams in later correspondence relied as being proof of the figure agreed to at the meeting of 17 April 1997.) Kikillus denied that such a telephone conversation had ever taken place. There is no note or any other documentary substantiation of Msiza’s evidence in this regard. Yet it is not, in my opinion, inherently unlikely that such a conversation could have taken place, considering the novel nature of the request made to him. Nor is it immediately apparent why Seagrams should insist on a breakdown of the agreed target volume since it is the globular figure at the end of the fiscal year that has contractual significance and not the monthly breakdown. And if Seagrams wanted figures merely to track the performance of the individual brands against expectations and projections SFW’s budget figures would have served that purpose equally well. Seagrams could have done the conversion itself, knowing what it believed the agreed figure to be.

[26] I am accordingly less impressed than was the Court a quo with the submission, so strongly advanced on behalf of Seagrams, that the budget figures furnished by Msiza from month to month, as opposed to ‘plan’ figures, were ‘meaningless’ to Seagrams; and that the furnishing thereof therefore served as corroboration of what it said was agreed to on 17 April 1997.

[27] The centre of gravity of the probabilities clearly lies in Msiza’s letter of confirmation of 29 April 1997, read with Seagrams’s lack of an appropriate response thereto, if it truly contradicted the figure stated in that letter. These considerations are not counterbalanced by the mere fact that Msiza thereafter furnished Seagrams, month by month, with the breakdown of actual sales of the three brands juxtaposed with a breakdown of SFW’s budget figures.

[28] Having regard to the probabilities analysed thus far Msiza’s version must accordingly be accepted as being more probable than that of Fleck and Kikillus. The events thereafter, in my view, fall into the same pattern.

The correspondence during the period July 1997 to November 1997

[29] Seagrams’s main contention was that it repeatedly asked for a monthly breakdown of ‘Plan’; that Msiza properly understood this to be a request for a breakdown of the agreed target volume; and that, by furnishing, without qualification, statistical information based on 5365715 litres, he in effect reaffirmed Seagrams’s version of what was agreed to at the meeting of 17 April 1997. Msiza’s answer was that he had explained to Kikillus telephonically that he lacked the expertise and equipment to convert the breakdown of the budget figures supplied to him by SFW’s statistical section (which of course was based on 5365715) to a breakdown of the agreed target volume of 5317000. The differentiation between the two base figures, a matter of a mere one percent, was so minimal that the breakdown of the budget figures (as opposed to the ‘plan’ figures) would in any event have served the same purpose. The Court a quo’s finding that the monthly budget figure was ‘totally useless and irrelevant’ to Seagrams and that Msiza’s evidence in this regard was ‘absurd’ completely misses this point. The subsequent breakdowns Msiza supplied for August 1997, September 1997 and October 1997 remained throughout the breakdown of the SFW budget figures for those periods and were consistently so described by him.

The correspondence in November 1997

[30] On 11 November 1997 Kikillus requested a summary of the Martell figures for its use during the visit of the president of Seagrams, and enclosed a ‘blank format’ which included a heading ‘plan’. Msiza responded by using Kikillus’s format, but he continued to furnish the budget figure breakdown. Viewed in context this does not amount to an acknowledgement by Msiza that this was the figure agreed to on 17 April 1997. Thereafter Msiza continued to adopt the format requested by Kikillus. At first blush it may seem to favour Seagrams’s cause but on analysis it adds very little, if anything, to it.

Msiza’s meeting with Brandon Morris in March 1998

[31] During February 1998 Kikillus telefaxed Msiza and suggested that a meeting should be held, during his temporary absence overseas, with Brandon Morris, Seagrams’s strategic planning manager, to discuss certain matters relating to, inter alia, ‘expenditure’ and ‘plan’. Msiza agreed and a meeting was arranged. Msiza testified:

‘The first thing that came to my mind, was I would be having a meeting with someone who had never worked on Martell before. So I had to provide a lot of background information on the brand. Not only was this person new to working with Martell, but that person had not been to our 17th April meeting. So that person did not know - probably did not know that here were two different figures on Martell, an internal SFW budget and a contractual target that we have with Seagrams. So I then decided well, for this meeting I have to make this person aware that we have our internal budget, and there is the Seagrams contractual target, and also bringing him up to speed with all the things listed in Mr Kikillus letter, which requested this meeting.’

 

Msiza accordingly prepared a document for Morris’s attention. In this document under the heading ‘budget 1977/8’ he inserted the figure 5321000. This is a figure he took off annexure C, being a calculation of 102% of 5217000. It was, he said, a mistake on his part and a careless one at that, since the agreed sales target, which is what he had in mind, was 5317 and the SFW budget figure was 5366. At the subsequent meeting he had with Morris, when he presented the document to him, he explained this error. His testimony was:

‘… in trying to use the official contractual volume we have with Seagrams, I just quickly glanced at the blue book and saw the figure of 5321 and I mistakenly used that one instead of the 5317 here, which you see here. So I used the 5321, but what I did do, was at that meeting, I highlighted to Brandon Morris that there are two different figures. There is SFW’s internal budget volume and there is the contractual target we have with Seagrams and he wrote here, that is his handwriting, he wrote it, while I was presenting to him, under the column budget, he wrote there, internal SFW plan, is 44 000 litres more.

Which would bring it to 5365 -- Yes.’

And again:

‘I explained to him that there are two different figures and he wrote that the SFW internal plan is actually 44 000 litres higher than the contractual one.’

 

That Morris fully understood the distinction Msiza sought to draw between the SFW budget figure and the agreed volume figure also appears from a document, dated 17 March 1998, in which, under the heading ‘volume versus plan’, Morris recorded:

‘According to your latest estimates, total Martell sales will equal 4,6 litres (518K cases). This will represent 90% of the plan target agreed with Marais Kikillus and Peter Fleck at last year’s planning meeting’.

(On that arithmetic the agreed volume would have been 5,1).

 

[32] What is significant about this meeting between Msiza and Morris is not so much the exact figures that were bandied about, but (a) the clear distinction drawn by Msiza and understood by Morris between the agreed target volume and the SFW budget figure; (b) that he referred back to exhibit B, albeit incorrectly, as his aide memoire as to what was in fact agreed at the meeting of 17 March 1997; (c) that he did not purport to furnish SFW’s budget figures but the agreed volume figure. That means that in his mind there was no identity or coincidence between the two figures, which controverts the suggestion that in the earlier documentation he had by implication accepted Seagrams’s version that the agreed volume figure was exactly the same as SFW’s budget figure; and, lastly, (d) that all of this happened in March 1998 long before the dispute flared up between the parties in August 1998 after the final actual sales figure had been determined.

The correspondence after July 1998

[33] On 1 July 1998 Msiza furnished his monthly report for June 1998. That completed the statistics for the fiscal year which is now under scrutiny. It showed a cumulative sales figure for the entire year of 4284748 litres. On 2 July 1998 Kikillus sent Morris an internal note reading:

‘Herewith latest Martell numbers - what % of plan did one achieve?’

 

Morris was not called to explain how he responded and what base figure he used to work out the percentage. Up to that point Seagrams had not yet committed itself on paper as to the figure it claimed to be the agreed figure, but after consultation with its legal advisers and on 11 August 1998, Fleck wrote to SFW stating, inter alia:

‘The annual sales objective for the 1997/8 year, as agreed upon between ourselves and yourselves, was 5 365 715 litres. As appears from the figures recently provided to ourselves, the actual sales achieved for this period were 4 284 748 litres. In the circumstances, you have achieved sales of less than 80% of the sales objective for this year.

We are, in the circumstances, entitled to renegotiate the agreement between ourselves and yourselves, and will be contacting you shortly in order to commence such negotiations.’

 

This letter caused concern in SFW’s ranks. Stroebel, its managing director, asked Bullen to report to him. Bullen telephoned Msiza who was at that point busy with promotional activities in Cape Town. Msiza gave Bullen certain information, whereupon Bullen searched Msiza’s office for the relevant file. It is not necessary to unravel, in this judgment, the events and misunderstandings that led to several mistakes in Stroebel’s letter of reply to Seagrams of 18 August 1998. It is a letter which occasioned SFW some embarrassment and which called for an explanation on the part of Bullen and Msiza. It was strongly argued on behalf of Seagrams that it also called for an explanation from Stroebel himself but Stroebel was never called by SFW. His letter read, inter alia:

‘The figure you have used (5365715 liters) was an SFW budget incentive figure and was provided to you on 11 August 1997 for your use in a Seagram’s presentation during a presidential visit (Letter enclosed).

SFW originally proposed a volume figure of 5217000 in April 1998 (sic). (See enclosed) You were not satisfied and requested 100 000 liters more. This was eventually agreed upon, hence the 29 April letter.

When the in-house budget for SFW is prepared I personally push up volumes on which an incentive will be paid to the sales force. This was the figure Pawn Msiza provided you with on 11 August i.e. 5 months after the volumes were agreed and do not represent the base for the agreement.’

 

The letter contains a number of mistakes. The most important one was of course that the figure of 5317000 litres was obtained by boosting the figure of 5217000 at Seagrams’s insistence, by an additional 100000 litres. That was not SFW’s evidence. This mistake was due to an assumption Bullen made which he conveyed to Stroebel. The statement, linking the letter of 11 August 1997 to the presidential visit which took place in November 1997, was due to a misinterpretation by Stroebel of Bullen’s internal memo to him. Stroebel’s own statement that he personally pushed up the volumes when the in-house budget was prepared, was an exaggeration and inaccurate as far as it went. These are all matters that had to be explained by SFW witnesses and on which they were, not without justification, severely criticised. The criticism did not, however, in the end unnerve the explanations that were given by Msiza and Bullen. Stroebel’s evidence, if he had been called, could not have added much to it. Nor did the criticism have any real bearing on the essential probabilities. I think counsel for SFW was right in describing it all as something of a red herring.

[34] In assessing the probabilities, phase by phase as events unfolded, as well as comprehensively and in retrospect, the conclusion seems to me to be inescapable that of the two versions before court as to what the parties agreed to, SFW’s is the more probable. That being so, Seagrams has not succeeded in discharging the onus which it assumed for itself in suing for a declaratory order. It further follows that SFW’s appeal must succeed.

[35] The following order is made:

(1) The appeal succeeds with costs, including the costs of two counsel relating to the application for leave to appeal before the Court a quo as well as before this Court.

(2) The following order is substituted for the order granted by the Court a quo:

‘The Plaintiffs’ claim is dismissed with costs, including the costs of two counsel’.

 

………………….

P M NIENABER

JUDGE OF APPEAL

Concur:

FARLAM JA

BRAND JA

HEHER AJA

LEWIS AJ A