1 This appeal arises from the decision of the Gen---eral Division of the High Court in Sizer Metals Pte Ltd v Chubb Insurance Singapore Ltd [2022] SGHC 51 (the “Judgment”). In HC/S 1248/2019, Sizer Metals Pte Ltd (“Sizer”) successfully sued Chubb Insurance Singapore Ltd (“Chubb”) for the loss of four shipments of tin concentrate in drums (the “Four Shipments”) that were insured with Chubb under a Marine Cargo Open Policy which incorporated standard Institute Cargo Clauses (A) (1/1/1982) (“ICC(A)”) terms. The Judge held that Chubb was liable to indemnify Sizer since the loss of the Four Shipments was due to theft that had occurred during the insured voyage. Dissatisfied with the Judge’s decision, Chubb appealed.

2 This appeal hinges on the Transit Clause in cl 8.1 of the ICC(A), which stipulates the operative duration of the insurance, ie, when the insurance attaches and when it terminates. Ultimately, the key point at issue is whether the thefts of the Four Shipments occurred during the period of insurance. Chubb contends that the Judge came to the wrong conclusion that Sizer had discharged its burden of proof that the thefts occurred during the period of insurance (ie, the insured voyage). In the court below, the primary question before the Judge was whether the thefts of the Four Shipments occurred prior to or after the cargo insurance in question had come on risk for the duration of the insurance cover. It was undisputed that the thefts were first discovered by the end receiver in Penang, Malaysia (“Penang”), whereupon it was revealed that the drums contained iron oxide (also known as “iron ore”) instead of tin concentrate (also known as “tin cassiterite”). It is common ground that the tin concentrate was replaced with the iron oxide after the drums were sealed in Kigali, Rwanda (“Kigali”). Put differently, the parties do not dispute in this appeal that the subject matter insured under the insurance policy, ie, tin concentrate, did exist at the outset and was filled into the drums.

3 In our judgment, Sizer prima facie discharged its legal burden of proving that the thefts had occurred during the period of insurance on the basis of undisputed facts and contemporaneous documentation. The evidential burden then shifted to Chubb who had the opportunity to produce evidence either rebutting Sizer’s evidence or supporting Chubb’s counter theory that the theft had occurred in Excellent Mining’s premises (ie, prior to the commencement of the insured voyage). However, in our view, Chubb failed to discharge its evidential burden. Consequently, we conclude that Sizer had discharged its legal burden on the balance of probabilities. We elaborate on our reasons for arriving at this conclusion below. Where relevant, we also make reference to the dissenting judgment of Woo Bih Li JAD (“the Minority Judgment”) and address some of the points raised therein.

4 Sizer is a company incorporated in Singapore that carries on the business of trading base metals. On 16 September 2013, Sizer and Chubb entered into a Marine Cargo Open Insurance Policy No 92359646 (the “Policy”). Pursuant to the Policy, Chubb agreed to insure Sizer’s purchases comprising base metals including tin concentrate against any loss, damage, or expense arising out of their conveyance from Kigali to various inland destinations such as the port at Dar es Salaam, Tanzania (“Dar es Salaam”), and thereafter by sea carriage to overseas destinations such as Penang. The schedule to the Policy incorporated the ICC(A) terms which covers all risks save for express exclusions (see cl 1 of the ICC(A)). For our present purposes, these exclusions are not relevant as Chubb did not raise any express exclusions in defending Sizer’s claims.

5 Sizer entered into two sale and purchase contracts with Excellent Mining Co Ltd (“Excellent Mining”), a company incorporated in Rwanda, for the purchase of tin concentrate on 15 September 2017 (the “First Contract”) and 30 May 2018 (the “Second Contract”). The First Contract concerned six shipments of tin concentrate. The first five shipments under the First Contact were shipped and received by the end receiver in Penang, the final destination, without incident. However, the last shipment of tin concentrate under the First Contract bearing International Tin Supply Chain Initiative (“ITSCI”) Shipment Number EXC/RW/0000011 for 27 drums of tin concentrate (“the Sixth Shipment”) was discovered after its arrival in Penang on 10 July 2018 to have been replaced with iron oxide.

6 The Second Contract concerned three shipments of tin concentrate. For ease of identification, the three shipments under the Second Contract are referred hereto as follows: (a) ITSCI Shipment Number EXC/RW/0000013 for 30 drums of tin concentrate (“the Seventh Shipment”); (b) ITSCI Shipment Number EXC/RW/0000015 for 36 drums of tin concentrate (“the Eighth Shipment”); and (c) ITSCI Shipment Number EXC/RW/0000017 for 40 drums of tin concentrate (“the Ninth Shipment”). The Seventh Shipment arrived in Penang on 25 July 2018. The Eighth and Ninth Shipments arrived in Penang on 6 September 2018. It was similarly discovered after delivery to the end receiver in Penang that the contents in the drums of the Seventh to Ninth Shipments had also been substituted and replaced with iron oxide. In short, from the Four Shipments a total of 86.075mt of tin concentrate in 133 drums had been replaced with iron oxide. Pursuant to the Policy, Marine Cargo Insurance Certificates were issued on 25 May 2018, 12 June 2018, 23 June 2018, and 3 July 2018 to signify insurance cover in respect of the Four Shipments.

7 After the discovery of the thefts of the tin concentrate, Sizer sent notices of claim for the shipments to Chubb on 16 July 2018 for the Sixth Shipment, on 31 July 2018 for the Seventh Shipment, and on 24 September 2018 for the Eighth and Ninth Shipments.

8 Operational activities, be it in connection with the appropriation of the consignment purchased under the First and Second Contracts or the preparation of the consignments for export, would be accompanied with the relevant documentation tracking each of these activities (see [9(a)]–[9(e)] below). In relation to the Four Shipments, these operations were conducted in Excellent Mining’s premises in Kigali.

9 In the Judgment (at [8]), the Judge set out the various stages involved in the transporting of the tin concentrate from Excellent Mining to the end receiver, Malaysian Smelting Corp Bhd (“MSC”), in Penang. We adopt the Judge’s list with the addition of other activities that are important to the reasoning behind the majority’s judgment (the “Majority Judgment”):

(a) Leading up to each shipment, approximately 20 to 25mt of tin concentrate was procured by Excellent Mining and brought to Excellent Mining’s premises in plastic bags (Judgment at [8(a)]).

(b) A representative from Alex Stewart International Rwanda Ltd (“ASIR”) would collect samples of the tin concentrate for analysis. The tin concentrate was then weighed and filled into empty second-hand steel drums in the presence of an ASIR representative. It is undisputed that the drums were filled with tin concentrate through a bunghole at the top of each drum. After the drums were filled, the bungholes were welded to close the openings. Tamper-proof clips (also known as “Precintia clips”) were spot welded on the sealed bungholes. A layer of white alkyd paint was applied to the top of the drums and the ITSCI Shipment Number and/or lot number and the addresses of both Excellent Mining and Sizer were written on the paint coating. A representative from the ITSCI and a Mineral Field Officer (“MFO”) from the Rwanda Mines, Petroleum and Gas Board (“RMB”) were in attendance during this process (Judgment at [8(a)]).

(c) The drums were stored at Excellent Mining’s premises pending the results of sampling from ASIR (Judgment at [8(a)]).

(d) Three certificates were issued before the drums were transported from Excellent Mining’s premises:

(i) The first was the Certificate of Sampling, Weighing and Packing (the “Packing Certificate”) issued by ASIR. The Packing Certificate documented the sampling process, packing of the tin concentrate into the drums, weighing of the drums containing the tin concentrate, and sealing of the drums.

(ii) The second was the Certificate of Assay (the “Assay Certificate”) issued by ASIR certifying the mineral content of the tin concentrate sampled.

(iii) The third was the International Conference on the Great Lakes Region (“ICGLR”) certificate which was issued jointly by the RMB and Rwanda Standards Board after the drums were inspected by RMB’s certification officers at Excellent Mining’s premises. Inspection of the drums by RMB’s certification officers would be at Excellent Mining’s premises after issuance of the Packing Certificate and the Assay Certificate.

(e) After the Packing Certificate, the Assay Certificate, and the ICGLR certificate were issued, Excellent Mining would have to obtain the certificates of origin and export declaration sheets from the Rwanda Revenue Authority (“RRA”). Before these documents were made available to Excellent Mining, the RRA would have to ensure that: (i) all taxes and dues for the export of the shipments of tin concentrate have been paid; and (ii) the relevant documentation (eg, the Packing Certificate, the Assay Certificate and the ICGLR certificate) were in order.

(f) Thereafter, the drums would be ready for transport. The drums would be stuffed into a 40ft container outside of Excellent Mining’s premises in the presence of the ASIR and RMB representatives. At this stage, no sign of tampering or thefts was observed. After stuffing, an ASIR representative would affix temporary seals on the doors of the 40ft container (Judgment at [8(b)]).

(g) The 40ft container would be driven on a 45 to 60minute-long journey to Bolloré Logistics Rwanda Ltd’s (“Bolloré”) bonded warehouse (the “Bonded Warehouse”) in Kigali (Judgment at [8(c)]).

(h) The 40ft container containing the drums would clear customs at the Bonded Warehouse, where the temporary seals affixed by ASIR were broken and the drums were inspected by representatives from ASIR, Bolloré, and the RRA (Judgment at [8(d)]).

(i) After inspection, the 40ft container would be re-sealed with three security seals:

(i) a seal by ASIR;

(ii) a seal by Bolloré; and

(iii) a seal by the RRA.

(j) The 40ft container would then be trucked on an inland transit between 1400 to 1500km, with overnight stops over several days to the port at Dar es Salaam (Judgment at [8(e)]).

(k) At Dar es Salaam, the drums were devanned from the 40ft container and stuffed into a 20ft container for the seaward carriage of the shipment to Penang (Judgment at [8(g)]). The oversight of the devanning and stuffing of the containers was undertaken by Bureau Veritas (Bolloré’s appointed surveyors) who would issue a stuffing survey (“the Stuffing Report”), which set out the following details:

(i) That the security seals on the 40ft container were intact.

(ii) The gross weight of each drum, as well as the net weight of each drum for the Eighth and Ninth Shipments. The gross weights but not net weights of each of the drums for the Sixth and Seventh Shipments.

(iii) The seal numbers on the drums and whether they tallied with the seal numbers recorded by ASIR in the respective Packing Certificates.

(l) After the drums were stuffed into the 20ft container, three seals were affixed to the door of the 20ft container:

(i) a seal by the Tanzania Revenue Authority;

(ii) a seal by Bureau Veritas; and

(iii) a seal by the liner shipping company.

(m) The doors of the 20ft container would be sealed in the presence of representatives from the Tanzania Revenue Authority, Bureau Veritas, and the liner shipping company (Judgment at [8(g)]).

(n) The 20ft containers were stored at the port in Dar es Salaam pending loading onboard the designated carriers.

(o) The 20ft containers were shipped to Penang on the following vessels:

(i) For the Sixth Shipment, on board the vessel “MIAMI TRADER” to a transhipment port, and then on board the vessel “TEERA BHUM” for carriage to Penang.

(ii) For the Seventh Shipment on board the vessel “VICTORIA SCHULTE” to a transhipment port, and then on board the vessel “TEERA BHUM” for carriage to Penang.

(iii) For the Eighth Shipment on board the vessel “NORDMED” to a transhipment port, and then on board the vessel “XIN CHANG SHU” for carriage to Penang.

(iv) For the Ninth Shipment on board the vessel “BUXFAVOURITE” to a transhipment port, and then on board the vessel “XIN CHANG SHU” for carriage to Penang.

(p) Upon arrival in Penang, the 20ft containers were placed on a truck for delivery to the end receiver, MSC.

(q) At MSC’s warehouse, the seals of the 20ft containers were broken and the 20ft containers were opened in the presence of Sizer’s appointed surveyor, Alex Stewart International Malaysia (“ASIM”).

10 After the thefts were discovered, on or around 3 October 2018, one representative from Sizer and one representative from Chubb, namely Dr Luigi Petrone (“Dr Petrone”), visited Excellent Mining’s premises and the Bonded Warehouse to investigate. On 7 May 2019, Chubb rejected Sizer’s claims on the basis that its investigations had led it to conclude that the tin concentrate had been replaced with iron oxide at Excellent Mining’s premises. On 30 August 2019, Rwanda’s National Public Prosecution Authority (“NPPA”) released a report (the “NPPA Report”) after investigations conducted by the Rwanda Investigation Bureau (“RIB”). In the NPPA Report, the NPPA concluded that the thefts had not been committed in Rwanda. Chubb does not dispute that the NPPA Report was issued but disagrees with the findings in the NPPA Report.

11 On 16 November 2021, the second last day of the trial, the parties submitted to the Judge the Agreed Statement of Facts and Issues No 4 (“ASOF 4”). For now, we need only refer to three sets of agreed facts that are material and common to the four shipments in question:

12 It is pertinent that the parties appear to have accepted that as a matter of construction of cl 8.1 of the ICC(A), the duration of the insurance cover contemplated is delineated by the delivery terms of the sale and purchase contracts between Sizer and its seller, Excellent Mining. Clause 8.1 reads as follows:

The delivery term in the First Contract this was indicated as, “Warehouse in Kigali, Rwanda”. ¹ While the delivery term in the Second Contract was indicated as, “FCA Kigali, as per incoterms 2010, customs cleared”. ²

13 With that understanding in mind, the trial below proceeded on the clear footing that: (a) the point of commencement of the inland transit (ie, the point of delivery to Sizer) for the Sixth Shipment was after the drums had left Excellent Mining’s premises in Kigali; and (b) the point of commencement of the inland transit (ie, the point of delivery to Sizer) for the Seventh to Ninth Shipments was after customs clearance at the Bonded Warehouse in Kigali and before the start of the inland transit to the port at Dar es Salaam. Collectively, these periods shall be referred to as the “Transit Period”.

14 Before the Judge, the parties thus agreed that the primary issue was where the thefts of tin concentrate had occurred. In this regard, parties also agreed that Sizer bore the burden of proof to show that the losses had occurred after the commencement of transit; more specifically, during the Transit Period.

15 In support of its case, Sizer put forward, inter alia, the following evidence:

16 Further, Sizer argued that it would have been impossible for the 86.075mt of tin concentrate to have been swapped (ie, substituted and replaced) at Excellent Mining’s premises given, inter alia: (a) the logistics required; (b) the time and space constraints; (c) the location of Excellent Mining’s premises in the city centre; (d) the compound walls of the premises; and (e) the security guards on duty and Mr Theodore Sindikubwadbo (“Mr Sindikubwadbo”) (director of Excellent Mining) whose residence was in the compound of Excellent Mining’s premises. Sizer argued that based on the evidence of its expert witness, Mr Howard Nathan Wheeler (“Mr Wheeler”), the theft could have happened in transit, such as at the port at Dar es Salaam, or during the inland transit between Kigali and Dar es Salaam.

17 Conversely, Sizer argued that Chubb’s expert witness, Dr Petrone, was inexperienced and had failed to conduct a thorough investigation of the thefts, which in turn led him to erroneously conclude that the thefts of the tin concentrate had occurred at Excellent Mining’s premises.

18 Chubb argued that the burden was on Sizer to provide evidence of where the thefts of the tin concentrate had occurred, not merely a hypothesis of how it could have happened. In this regard, Chubb argued that the burden was on Sizer to show how and when the thefts occurred and that Chubb was not required to show on a balance of probabilities that the thefts had happened at Excellent Mining’s premises.

19 Chubb argued that apart from Mr Wheeler, none of Sizer’s witnesses had provided any evidence to show that the thefts had occurred after the commencement of transit. In any event, Mr Wheeler’s evidence merely hypothesised where the thefts could have had occurred based on nothing more than a “desktop” evaluation of documents before him. Further, the drums were stowed in 40ft and 20ft containers with their seals intact. Sizer did not produce any evidence to show that the seals of the containers had been tampered with. Neither did Sizer produce any evidence that the 40ft or 20ft containers had been tampered with. In addition, Chubb pointed to the lack of evidence to demonstrate that the equipment required to swap the tin concentrate was available at the port at Dar es Salaam. Chubb argued that it was incumbent on Sizer to have adduced evidence from Bureau Veritas, the surveyors appointed by Bolloré, who were in turn appointed by Sizer to transport the 40ft and 20ft containers from Rwanda to Penang. Accordingly, as Sizer had failed to call witnesses from these entities to give evidence in the proceedings, an adverse inference should be drawn against Sizer.

20 Chubb also pointed to two pieces of evidence in support of its case. First, was the presence of three minerals, namely, cassiterite (tin oxide, also known as “tin concentrate”), coltan (columbite-tantalite) and wolframite (iron manganese tungstate) (collectively, the “3Ts minerals”), which were found mixed with the iron oxide found in the drums upon their receipt at MSC’s premises in Penang. Chubb argued that the presence of the 3Ts minerals in the iron oxide showed that the swap had taken place in Rwanda, as Tanzania did not have any tantalum or tungsten mines. Second, was the use of white alkyd paint on the lids of the drums which was visually similar to the cans of white paint seen in Excellent Mining’s premises.

21 Finally, Chubb submitted that as no witness was called to testify as regards to the NPPA Report, it could at best only “be taken at face value”. Moreover, as the NPPA Report lacked proper investigation to lead to findings of fact and had instead relied on the report of Bureau Veritas stating that the containers were in good condition (although both parties had agreed that there was some tampering with the drums), it could not be relied upon.

22 The Judge’s approach to the resolution of the dispute was shaped by the agreed issues contained in the ASOF 4, which we set out below:

23 With the agreed issues in mind, the primary question before the Judge was whether the thefts of the Four Shipments occurred prior to or after the cargo insurance in question had come on risk for the duration of the insurance cover. The Judge noted that he was presented with an “almost binary choice” for his consideration, which was whether the thefts had taken place during the period of insurance cover or not (Judgment at [39]). In practical terms, this translated to: (a) whether the theft of the Sixth Shipment occurred before the consignment left Excellent Mining’s premises or during the remainder of the Transit Period; and (b) whether the thefts of the Seventh to Ninth Shipments occurred before the consignments cleared customs at the Bonded Warehouse or during the remainder of the Transit Period. The Judge also held that as “all relevant facts [were] known such that all possible explanations, except an extremely improbable one, [could] properly be eliminated”, he could apply the process of elimination to determine whether the thefts had happened at Excellent Mining’s premises relying on the authority of Rhesa Shipping Company SA v Edmunds (The Popi M) [1985] 1 WLR 948 (“The Popi M”) (Judgment at [43]).

24 First, the Judge agreed with the evidence of Mr Sindikubwadbo who was Excellent Mining’s director at the material time, that the drums of tin concentrate were under tight security while in Excellent Mining’s premises (Judgment at [63]). In reaching this conclusion, the Judge accepted, inter alia, that:

25 The Judge also found that the evidence of Mr Felicien Nkomeje (“Mr Nkomeje”), a Certification Specialist at the RMB, corroborated Mr Sindikubwadbo’s evidence. Although the MFO, Ms Campire Laurence (“Ms Laurence”), was not called to testify during the trial, the Judge accepted Mr Nkomeje’s evidence that no issues were raised by Ms Laurence at the material time (Judgment at [60]).

26 The Judge also found that the evidence of Mr Norman Mwashi (“Mr Mwashi”), the Managing Director and Chief Chemist of ASIR, had corroborated the evidence of Mr Sindikubwadbo. Mr Mwashi had not noticed any signs of tampering when the drums of tin concentrate were transported to the Bonded Warehouse (Judgment at [62]).

27 The Judge also accepted the evidence of Mr Wheeler that there were significant logistical difficulties in swapping the tin concentrate with iron oxide at Excellent Mining’s premises because (Judgment at [66]):

(i) removing the lids of the drums;

(ii) emptying out the contents of the drums;

(iii) filling the drums with iron oxide;

(iv) resealing the lids of the drums;

(v) filling and painting the lids of the drums; and

(vi) repositioning the drums.

28 In the Judge’s opinion, it was “clearly logistically impossible for the thefts to have taken place at Excellent Mining’s premises, especially with the several layers of checks and surveillance at Excellent Mining” (Judgment at [68]). Accordingly, with all the logistical difficulties mentioned above, the swap of tin concentrate with iron oxide at Excellent Mining’s premises would only have been feasible with the involvement of Excellent Mining (Judgment at [69]).

29 The Judge also accepted Mr Wheeler’s opinion that the seals on the containers’ doors being intact did not mean that no one had gained access into the sealed containers (Judgment at [73]–[75]). In the Judge’s view, the swap of the contents in the drums had been performed by a “professional and well-organised gang of thieves”, who would have had no problem gaining access to the sealed drums in the sealed containers without tampering with the seals (Judgment at [76]). In the Judge’s opinion, the security measures were weakest when the drums were in transit from the Bonded Warehouse to the port at Dar es Salaam which would have only required the container’s driver and assistant to be compromised in order for the thefts to have taken place (Judgment at [79] and [84]). However, the Judge disagreed with Mr Wheeler’s opinion that the thefts could have occurred at the port in Dar es Salaam, as there was the presence of human traffic, security guards, and CCTV cameras at the port area (Judgment at [80]).

30 As the parties’ experts did not opine on the likelihood of the thefts occurring during the remaining stages of the Transit Period, the Judge did not think it necessary to consider the possibility of the theft occurring during those stages. In any case, the Judge was of the view that it would have been very difficult for the thefts to have occurred on the seaward voyage to Penang or at the port in Penang, as there were too many people around for thefts of such magnitude and sophistication to have gone unnoticed (Judgment at [81]). In addition, the Judge remarked that Chubb’s belated suggestion that it was possible for the thefts to have occurred at MSC’s premises in Penang was contrary to Dr Petrone’s evidence and was thus of no significance (Judgment at [82]–[83]).

31 Mr Desmond Sim Kok Whye (“Mr Sim”), the General Manager of Zama Marine Services and Consultancy Sdn Bhd (“Zama”), who had been appointed by Chubb to investigate the loss of the tin concentrate, and Mr Yeoh Oon Huat (“Mr Yeoh”) an ad-hoc surveyor for Zama testified as factual witnesses for Chubb at the trial. However, between the two men, only Mr Yeoh personally inspected the drums at MSC’s premises in Penang. Mr Sim testified that it was unlikely that the thefts occurred on the journey to Dar es Salaam. However, the Judge rejected his evidence as partial (Judgment at [86]) and illogical as vis-à-vis the agreed facts (Judgment at [88]). We pause here to put in context the Judge’s remarks on the illogicality of Mr Sim’s testimony. Mr Sim claimed that he did not know the lids of the drums had been cut and that the premise of his investigation was that “there was no cutting of the drum, there was no theft, … something must have gone wrong from the point of origin”. ³ Mr Sim’s case theory was thus that the drums had never been filled with tin concentrate which contradicts the agreed facts on this point. Indeed, it was also an agreed fact that the tin concentrate in the drums were swapped with iron oxide (see [11(b)] above). In any event, we observe that since Mr Sim testified not as an expert witness, but as a factual witness, his opinion on where the thefts were likely or unlikely to have had occurred should be rejected. Further, Mr Sim was not on-site during the inspection of the drums and thus had no personal knowledge of their condition.

32 Mr Yeoh testified that upon his inspection of the drums, he did not notice any sign of tampering. However, we note that Mr Yeoh’s testimony is also plainly inconsistent with Chubb’s case premised on Dr Petrone’s suggested modus operandi of the thefts which included the cutting of the drums’ lids to facilitate the swap of the tin concentrate and iron oxide, and subsequent re-welding of the lids and the application of paint to conceal the welding marks.

33 Second, in considering the NPPA Report, the Judge held that the NPPA was Rwanda’s prosecution authority, and that if it had found that no thefts were committed in Rwanda, “it would not have been necessary for it to further investigate the method of thefts, which would have happened elsewhere” (Judgment at [95]). The Judge also held that Dr Petrone (and Chubb) did not have any basis to allege that the NPPA had conducted an improper investigation as he did not know the extent of investigation performed by the NPPA and the RIB (Judgment at [96]). In this regard, the Judge stated that while Sizer had not called the official from the NPPA in charge of the investigation to testify in court and that caution was warranted when relying on the NPPA Report, Chubb had failed to produce reasonable or credible evidence to disregard the NPPA Report, and that the findings made in the NPPA Report could not be ignored (Judgment at [97]).

34 Third, the Judge rejected Chubb’s submission that the concomitant presence of the 3Ts minerals detected in the iron oxide samples taken in Penang from the drums, weighed in favour of the finding that the thefts occurred at Excellent Mining’s premises.

35 The Judge rejected Dr Petrone’s opinion that the 3Ts minerals which were “present on the floor of the open yard or on other equipment/tools used at Excellent Mining” were admixed with the iron oxide (also located in Excellent Mining’s premises) prior to the drums being filled. By this contention, iron oxide and the 3Ts minerals would have to be present at the same location and at the same time. Dr Petrone’s opinion was drawn from three observations: (a) the 3Ts minerals are rarely found together in the same mine; (b) Excellent Mining trades in the 3Ts minerals; and (c) the samples of iron oxide that he took in Penang did contain 3Ts minerals (Judgment at [106]). In turn, Dr Petrone’s conclusion was largely based on another of Chubb’s expert witnesses, Dr Hans-Eike Gäbler (“Dr Gäbler”), a chemist who tested four samples of iron oxide from the drums in Penang to ascertain whether the cassiterite found in the samples: (a) came solely from two mines sites in Nyaruvumu and Gituntu; and (b) if all four samples came from the same source. Dr Gäbler reported, inter alia, that it was “very plausible” that the cassiterite grains in all four samples originated from the same source. In this regard, he clarified that a source did not necessarily refer to a specific geographical region or a specific mine. Instead, a source could also refer to a mixture of material from one mine with another, which would present its own signature (Judgment at [101] and [103]). The Judge did not find that Dr Gäbler's report assisted Chubb’s case. It was undisputed that the origin of the cassiterite, ie, the tin concentrate, was from Excellent Mining. Each consignment of tin concentrate was mixed on-site from plastic bags of tin concentrate obtained from various mines. Moreover, Dr Gäbler did not analyse the origin of the other two 3Ts minerals, coltan and wolframite, which would have been pertinent to this issue (Judgment at [105]).

36 In addition, the Judge rejected as speculative Dr Petrone’s basis for his contention that the iron oxide found in the drums was from Excellent Mining’s premises. Dr Petrone claimed that he saw a “red substance” believed to be iron oxide in Excellent Mining’s premises. However, the Judge noted that Dr Petrone did not take a sample of the red substance for testing. Dr Petrone also conceded on the stand that there could be other red substances that are not iron oxide (Judgment at [107]).

37 The Judge also did not accept Dr Petrone’s theory that the presence of the 3Ts minerals in the drums meant that the swap took place at Excellent Mining’s premises, and that the iron oxide had been admixed with the 3Ts minerals prior to the swap. Instead, he noted that this was not the only possible explanation for the presence of the 3Ts minerals. The Judge observed that if the 3Ts minerals had indeed been found on the ground of Excellent Mining’s premises, it was possible that they could have been admixed with the tin concentrate when the plastic bags containing them were bled and emptied onto the ground before being filled into the drums. Thus, when the swap with the iron oxide occurred, remnants of the tin concentrate and the 3Ts minerals remained present in the drums. This alternative explanation was supported by the evidence of Sizer’s expert witness, Dr Mirjana Küzma (“Dr Küzma”). Dr Küzma testified that the presence of the 3Ts minerals could be explained by the remnants of the tin concentrate in the drums after the swap (Judgment at [109]).

38 Again, the Judge rejected as unconvincing Dr Petrone’s claim that the four samples he took from the drums consisted purely of the iron oxide that was swapped, without the remaining tin concentrate that was originally in the drums. Dr Petrone argued that since the iron oxide filled the whole of the drums, there was little room for thorough mixing of the iron oxide in the drums. The remaining tin concentrate left behind after the swap would be found mostly at the bottom of the drums. Therefore, as the four samples were not extracted from the drums’ bottoms, they would have only been samples of the iron oxide that was swapped. However, Dr Petrone’s testimony was inconsistent with the stated sampling methodology set out in his expert report, whereby he recorded that the four samples were composite samples made up of samples retrieved from three different levels of the drums, ie, the top, middle and bottom levels. The Judge accepted Dr Petrone’s oral testimony over the sampling methodology recorded in his expert report, but he ultimately dismissed the former, opining that the iron oxide in the drums must have been subjected to considerable movement during the long journey. Moreover, Mr Wheeler testified that cassiterite is a “dusty and free-flowing” material (Judgment at [110]).

39 All said, the Judge held that Dr Petrone’s case theory that the swap took place at Excellent Mining’s premises due to the presence of 3Ts minerals in Excellent Mining’s premises was not conclusive but instead speculative (Judgment at [109]).

40 Fourth, on Dr Petrone’s claim that the white alkyd paint used on the drums was visually similar to the cans of white paint seen in Excellent Mining’s premises, the Judge noted that Dr Petrone did not take a sample of the paint from the cans of white paint seen at Excellent Mining’s premises. He only collected paint flakes from around the rim of a drum’s lid and a sample of the paint layer from a cut out of a drum’s lid. Both samples were sent to SOCOTEC UK Ltd’s (“Socotec”) laboratory for analysis. Socotec subsequently issued its report (the “Socotec Report”), which stated that there was a strong correlation between the samples of paint, although there was no complete match due to the samples being contaminated. To the Judge, the Socotec Report was thus inconclusive (Judgment at [116]). Be that as it may, Sizer’s expert, Dr Küzma, and Chubb’s expert, Dr Petrone, arrived at diametrically opposed interpretations of the findings in the Socotec Report. Dr Küzma opined that the two white paints were not the same while Dr Petrone opined otherwise (Judgment at [117]). However, ultimately the Judge favoured Dr Küzma’s opinion that the paint samples were markedly different and could not have been from the same source (Judgment at [119]–[125]). Nonetheless, the Judge concluded that even if the two paint samples were an exact match, this would not assist Chubb’s case. This was because Dr Petrone’s case theory assumed that the same group of persons applied the paint after the thefts were committed. However, no evidence was adduced to show that Excellent Mining was the only one using the particular paint which happened to be a paint-type that was commercially available in Rwanda and Tanzania (Judgment at [126]).

41 Further, Dr Petrone claimed that the thick layer of white alkyd paint which was used to cover up the welding marks was present since the drums left Excellent Mining’s premises to the time they arrived at Penang, and therefore the thefts had occurred there (Judgment at [127]). This being supported by photographs demonstrating that the thickness of the paint applied to the drums was the same. However, the Judge too rejected this as inconclusive (Judgment at [128]–[129]).

42 Fifth, the Judge held that Chubb’s difficulty in establishing the bases of Dr Petrone’s case theory came from his lack of credibility as an expert witness. He noted that Dr Petrone’s investigation was “unsatisfactory, superficial and substandard”, listing a number of important tasks Dr Petrone had omitted to do in the course of his investigation (see Judgment at [132] and [134(a)]–[134(g)]). In addition, the Judge found that Dr Petrone “harboured a prejudiced belief and had a preconceived judgment that the thefts took place at Excellent Mining from the very start of his investigation” (Judgment at [137]). This was evidenced by Dr Petrone’s case at trial that Excellent Mining was complicit in the thefts, although this was not mentioned in his expert report or in his affidavit of evidence-in-chief (“AEIC”) (Judgment at [136]). Therefore, the Judge found that he was obliged to treat Dr Petrone’s evidence with utmost caution (Judgment at [137]).

43 In summary, the Judge was of the view that Sizer had been able to prove its case on a balance of probabilities that the thefts had taken place during the Transit Period. We examine the Judge’s reasons for concluding as such at [130]–[139] below.

44 Accordingly, Sizer’s claim for the undisputed sum of US$1,154,508.94 was allowed (Judgment at [139]).

45 In this appeal against the decision of the Judge, Chubb makes three arguments. First, that the Judge erred in reversing the burden of proof. In particular, Chubb argues that while there was a lack of evidence showing where and when the thefts had occurred, the Judge should not have adopted a process of elimination in concluding that the thefts had occurred during the Transit Period. In this regard, Chubb submits that the Judge erred in holding that he had an “almost binary choice” between finding that the thefts had occurred at Excellent Mining’s premises or during the Transit Period, as there was a third possibility – that the theft had occurred at MSC’s premises in Penang. This third possibility was wrongly rejected by the Judge as the parties had not in fact agreed that the theft occurred before the arrival of the shipments in Penang. Accordingly, it was Sizer’s burden to show that the losses did not occur in Penang, which Sizer had been unable to do. Further, Chubb also points to the Judge’s treatment of the NPPA Report to make the argument that he unfairly placed the burden on Chubb to show that the NPPA Report was not credible, where the burden should instead have been on Sizer to prove that the NPPA Report was credible.

46 Second, Chubb argues that there was insufficient evidence to support the Judge’s finding that the thefts had occurred during the inland transit between the Bonded Warehouse in Kigali to the port in Dar es Salaam. In this connection, Chubb submits that the Judge should not have relied on the evidence of Sizer’s expert, Mr Wheeler because: (a) he did not have the requisite expertise; (b) his opinion was speculative; and (c) he was unreliable and changed his opinions while on the stand.

47 Third, Chubb submits that the Judge’s finding that the thefts had occurred during the inland transit of the Transit Period was contrary to the other relevant facts brought up in the course of the trial and accepted by the Judge. These facts include:

48 Sizer agrees that the burden was on it to prove on a balance of probabilities that the thefts had occurred during the period of insurance coverage (ie, during the Transit Period). However, Sizer argues that there was no requirement for it to prove where exactly the thefts occurred, but rather it only had to show that the thefts did not occur before or after the Transit Period.

49 In this regard, Sizer argues that the Judge properly considered the following evidence in concluding that the thefts did not occur at Excellent Mining’s premises:

50 As for the transport of the drums from Excellent Mining’s premises to the Bonded Warehouse, Sizer submits that the Judge had considered the fact that the short 45 to 60 minute-long journey did not provide sufficient time for the theft to have occurred. Based on the evidence of Mr Wheeler, Sizer submits that the Judge was correct to find that the weakest link was along the land route from the Bonded Warehouse in Kigali to the port at Dar es Salaam.

51 As for the transport of the drums from Dar es Salaam to Penang and at the port in Penang, Sizer argues that the Judge correctly concluded that the theft would not have likely occurred during this leg of the journey.

52 Finally, Sizer argues that the Judge properly considered the evidence that the theft would not have likely occurred at MSC’s warehouse as there were several CCTVs installed, and no signs of tampering of the seals on the containers and the containers itself were found. In this regard, Sizer also argues that Chubb is belatedly raising the point about the thefts having possibly occurred in Penang, which was a point that it had not raised in its pleadings or affidavits. In any event, Sizer argues that Chubb’s arguments about the thefts occurring in Penang flies in the face of logic and is against the weight of the evidence led from Chubb’s own witnesses.

53 Chubb’s contention is that the Judge erred in holding that the loss of the Four Shipments occurred during the Transit Period. Specifically, the Appellant’s Case at para 21 identifies three issues in the appeal for this court’s determination, which are elaborated on at [45]–[47] above. We have further distilled them into key two issues:

54 Counsel for Chubb, Mr Goh Phai Cheng SC (“Mr Goh”) confirmed that Chubb is not appealing against the Judge’s findings and conclusion in relation to: (a) the 3Ts minerals; and (b) the identity of the white alkyd paint on the drums’ lids. Chubb had argued below that: (a) the presence of the 3Ts minerals in the iron oxide showed that the swap had taken place in Rwanda, as Tanzania did not have any tantalum or tungsten mines; and (b) the use of the white alkyd paint which was visually similar to the cans of white paint seen in Excellent Mining’s premises suggested that the swaps had occurred there (see [20] above). Mr Goh’s confirmation means that there is no appeal against the Judge’s rejection of Dr Petrone’s claim that: (a) both the 3Ts minerals and iron oxide were from Excellent Mining’s premises and that the 3Ts minerals were admixed with iron oxide prior to the swap of tin concentrate with iron oxide in the same premises; and (b) the paint samples and photographs demonstrate that the thefts had occurred in Excellent Mining’s premises. In particular, Chubb does not appeal against the Judge’s rejection of Dr Petrone’s claim that the cans of paint in Excellent Mining’s premises contained the same paint used on the drums’ lids to conceal the welding marks.

55 In light of how the parties presented their cases below and in the appeal, we set out the following road map of our analysis of the key issues outlined above. First, we make some general observations on the nature and scope of an “all risks” cargo policy and what the primary requirements are to prove an assured’s claim under the ICC(A), especially under cll 1 and 8.1. Second, we turn to consider whether Sizer had established its case against Chubb. In this connection, we discuss the law relating to the burden of proof in respect of a claim made by an assured against the insurer under an ICC(A) cover including the shifting of the evidential burden between Sizer and Chubb throughout the course of the trial. Third, we consider whether the Judge erred in holding that Sizer had proved its case on the balance of probabilities that the thefts occurred during the Transit Period.

56 Cargo is commonly insured against “all risks” under the ICC(A). Clause 1 of the ICC(A) states that the policy covers “all risks of loss of or damage to the subject-matter insured except as provided in Clauses 4, 5, 6 and 7 below” [emphasis added]. ⁴ This apparently comprehensive cover is restricted by the express exclusions enumerated in cl 1. In an “all risks” policy, therefore, the coverage is not against risks of a specified class or classes; “all risks” under the ICC(A) covers any fortuitous accident or casualty resulting in loss of or damage to the subject matter insured unless expressly excluded. On proof of a fortuitous loss, Donald O’May and Julian Hill, Marine Insurance Law and Policy (Sweet & Maxwell, 1993) at p 166 is instructive:

57 In the same vein, Lord Birkenhead LC, in the leading case involving “all risks”, British & Foreign Marine Insurance Co Ltd v Gaunt [1921] 2 AC 41 (“British & Foreign Marine Insurance Co Ltd”) said at 47:

58 It is clear that the burden is in some respects lighter than a policy of enumerated risks, and this important question of proof is one of the main advantages of an “all risks” cover. This, however, does not mean that the assured will not be required to disprove any counter theory that may be put forward by the insurer designed to show that the loss was not fortuitous. In this case, the parties had agreed on theft as the casualty that caused the loss of the Four Shipments. Theft is a recognised casualty covered in an “all risks” policy. Adopting Lord Birkenhead’s dictum, once Sizer demonstrated that the losses were caused by some accident or casualty (ie, theft), it need not go further to show how the theft had occurred.

59 As discussed above at [12]–[0], cl 8.1 of the ICC(A) sets out the duration of the insurance cover, which in the present case is delineated by the delivery terms of the sale and purchase contracts between Sizer and Excellent Mining. As a matter of general principle, an insurer is liable if the fortuitous loss caused by an insured peril occurred during the period of cover. In this case, there is no dispute that theft of the Four Shipments and the resulting loss occurred at the same time. The question for determination is whether the theft/loss occurred: (a) during the period of cover (ie, the Transit Period) as argued by Sizer; or (b) before the commencement of the Transit Period as argued by Chubb.

60 Before considering the primary question of whether the theft/loss occurred before or during the Transit Period, we first discuss the issue of burden and standard of proof on the time of occurrence of the fortuitous loss.

61 The parties agree that the burden of proof lies with Sizer to show that the fortuitous loss, ie, the thefts of the tin concentrate, occurred during the Transit Period. The legal burden remains on Sizer throughout the trial and does not shift. To discharge its legal and evidential burden at the first instance, Sizer must produce evidence in the form of witness testimony or documents to prove that the risk under the Policy had attached when the thefts/losses occurred. If Sizer succeeds in demonstrating on a prima facie basis that the losses occurred during the Transit Period, the evidential burden then shifts to Chubb who has the opportunity to produce evidence either rebutting Sizer’s evidence or supporting Chubb’s counter theory that the theft had occurred in Excellent Mining’s premises (ie, prior to the commencement of the Transit Period). In other words, Chubb has to produce sufficient evidence to raise the counter theory for the Judge’s consideration as the trier of fact. This evidential burden to adduce evidence can shift to Sizer once it has been discharged by Chubb. That the evidential burden can shift between the parties is the usual sequence of play in a trial. As explained by the Court of Appeal in Britestone Pte Ltd v Smith & Associates Far East, Ltd [2007] 4 SLR(R) 855 at [58] and [60]:

62 We will consider later in this Judgment the overall evidence presented to persuade the Judge to arrive at the decision in Sizer’s favour in the court below. In so doing, we will examine whether the overall evidence that Sizer and Chubb provided to show that a particular fact or event (in this case whether the theft had occurred prior to or during the Transit Period) was more likely than not to have occurred. At this juncture, we segue to consider Chubb’s complaint that the Judge reversed the burden of proof.

63 Chubb’s complaint is that the Judge had impermissibly reversed the burden of proof. Chubb raises two reasons in support of its position:

64 This complaint is curious and, in our view, misplaced. Putting aside the case of The Popi M for a moment, what transpired before the Judge, was indeed a classic sequence of play involving the shifting of the evidential burden from one party to the other (see [61] above). The Judge was always alive to the fact that the legal burden was on Sizer to prove its case (eg, Judgment at [28]). The Judge was also aware that he was “not always bound to make a finding one way or the other with regard to the facts averred by the parties” (Judgment at [41]). Sizer called four factual witnesses and two expert witnesses. Chubb called two factual witnesses and two expert witnesses, essentially to support its counter theory that the theft occurred in Excellent Mining’s premises prior to the commencement of the Transit Period.

65 Even if, the Judge erred in distinguishing The Popi M, we find it difficult to understand Mr Goh’s criticism that the Judge had reversed the burden of proof. The Judge expressed that the resolution of this case involved an almost binary choice for his consideration: either the thefts took place during the Transit Period or they did not (Judgment at [39]). Here, the Judge’s reference to a “binary choice” has to be understood in the context of cl 8.1 of the ICC(A) and the primary dispute between the parties. As stated above, the parties’ main dispute concerns where the thefts had occurred. This is relevant because cl 8.1 stipulates that insurance cover under the Policy would only span the duration of the Transit Period. Therefore, by virtue of cl 8.1, the Judge had to decide whether the thefts occurred prior to or after the risk had attached under the Policy. As it was not Chubb’s case in the proceedings below that the thefts had occurred after the insurance cover had ended (see [77]–[83] below), the Judge only had to decide whether the thefts had occurred before or during the Transit Period. Naturally, if the Judge found that it was not possible for the thefts to have occurred at one of these periods, then it stood to reason that he must be satisfied that the theft was more likely than not to have occurred during the alternative period. The Judge accepted that “the test is not whether the claimant’s case is more probable than the defendant’s, but whether the claimant’s evidence (and not hypothesis) has been proved on a balance of probabilities” [emphasis in original] (Judgment at [32]). It is clear to us that the Judge analysed the evidence relied on by the parties to ascertain whether Sizer had discharged its legal burden of proof (Judgment at [33] and [48]). Having rejected Chubb’s counter theory that the thefts had occurred at Excellent Mining’s premises, it was not wrong of the Judge to conclude that Sizer had proved its case on the balance of probabilities that the thefts had occurred during the Transit Period based on the evidence it presented, and all things considered. This analysis is simply an application of the principle discussed earlier, namely on the shifting of the evidential burden from one party to the other. We elaborate on this view later in the Majority’s Judgment at [120]–[128] below.

66 In the decision below, the Judge distinguished The Popi M from the facts of the present case to apply a process of elimination in deciding if the thefts had occurred during the Transit Period given that all the relevant facts were known (Judgment at [43]). Accordingly, as the Judge was of the view that the evidence adduced by Sizer had sufficiently eliminated the probability that the thefts had occurred at Excellent Mining’s premises and the immediate inland transit thereafter up until the shipments arrived at Dar es Salaam, what remained was the probability that the “thefts must have occurred at some point during the Transit Period” [emphasis in original] (Judgment at [48]).

67 We begin with Chubb’s argument on appeal that the Judge erred in distinguishing The Popi M from the present case and thereby adopted and applied the process of elimination erroneously (see [63] above).

68 In The Popi M, an insured vessel sank, and the shipowners sued the hull underwriters for the total loss of the vessel arising from the sinking. According to the shipowners, the vessel sank after a collision with a submarine. On the other hand, the hull underwriters alleged that the ship had sunk because of the effects of prolonged wear and tear. At first instance, the trial judge found the shipowners’ case of a collision with a submarine “improbable”. However, as the trial judge found the hull underwriters’ case premised on prolonged wear and tear “impossible”, he allowed the shipowners’ case. The English Court of Appeal dismissed the subsequent appeal. Dissatisfied, the hull underwriters appealed to the House of Lords, who allowed the appeal. Lord Brandon delivering the judgment of the House of Lords was dismissive and suggested that the trial judge’s view was based on an exchange by fictional characters, namely, Sherlock Holmes’ dictum to Dr Watson. In particular, Lord Brandon stated as follows (at 955G to 956E):

69 There are three principles that can be distilled from the passage quoted above from Lord Brandon in The Popi M:

(a) First, the court is not bound to always make a finding with regard to the facts put forward by the parties. It is open to the court to find that the plaintiff has been unable to discharge his burden of proof. Such is the case where the evidence placed before the court is unsatisfactory.

(b) Second, the legal burden of proof is on the plaintiff to show on a balance of probabilities that a particular event is more likely to have occurred than not. Where the court finds that the occurrence of an event is improbable, the plaintiff has failed to discharge his burden of proof.

(c)  Third, where all the relevant facts are known such that all possible explanations can properly be eliminated, the court would possibly be justified in finding that an improbable event was the truth.

70 With respect, it appears that the Judge wrongly concluded that Lord Brandon had disapproved of the method of elimination in The Popi M. To recapitulate, the Judge held (at [42]):

However, a closer reading of what Lord Brandon said shows that what was expressly disapproved of in The Popi M was the “method of elimination” proposed by the fictional character Sherlock Holmes that “when you have eliminated the impossible, whatever remains, however improbable, must be the truth” [emphasis added]. In other words, the court cannot simply accept a state of affairs as having been proven on a balance of probabilities, where the plaintiff’s case was the least improbable of the various possibilities, ie, allowing an improbable possibility to form the basis of a finding of fact only because it was less improbable than other possibilities put forward by the parties, and where the evidence “leaves [the judge] in doubt whether the event occurred or not”. In our view, the prohibition in The Popi M was thus not against a “method of eliminating possibilities to determine the true state of affairs” per se, but rather with employing this method of factual analysis without due regard paid to whether: (a) the full spectrum of facts that could explain the occurrence is before the court; and (b) whether the occurrence of an event, on the evidence and on the application of common sense, can be said to be proved on a balance of probabilities.

71 Our view is reinforced by the observation that none of the local cases which have cited and approved The Popi M, discussed or disapproved the method of elimination put forward by the Judge. We now turn briefly to the following cases: Clarke Beryl Claire (personal representative of the estate of Eugene Francis Clarke, deceased) and others v SilkAir (Singapore) Pte Ltd [2002] 1 SLR(R) 1136 (“Clarke Beryl Claire”); Surender Singh s/o Jagdish Singh and another (administrators of the estate of Narindar Kaur d/o Sarwan Singh, deceased) v Li Man Kay and others [2010] 1 SLR 428 (“Surender”) and Wartsila Singapore Pte Ltd v Lau Yew Choong and another suit [2017] 5 SLR 268 (“Wartsila”).

72 In Clarke Beryl Claire, it was stated at [63]:

73 The court in Wartsila at [87]–[89] stated:

74 Finally, as succinctly stated by the High Court in Surender at [121]:

75 The primary principle that can be gleaned from the cases cited above is that for the court to determine the possible cause(s) of the loss, it is the plaintiff’s burden to show that it was more likely than not that the alleged event actually happened, and not just that it was the least improbable possibility amongst a number of other possibilities. This is consistent with Lord Brandon’s holding in The Popi M that a plaintiff cannot discharge his burden of proof by showing that his version of events is less improbable than the other versions put forward.

76 We alluded earlier that even if the Judge had misunderstood Lord Brandon’s holding in The Popi M, this is ultimately of no consequence for the reasons stated above at [65]. To be clear, the question in The Popi M was whether the loss of the vessel was proximately caused by an enumerated insured peril. In that case, there were two competing causes for the loss proposed by the shipowners and the hull underwriters, both of which the trial judge had considered improbable. The trial judge felt unable to accept the insurer’s defence based on expert evidence of wear and tear. He accepted the theory of collision with a submarine notwithstanding the inherent improbability of the theory of collision. In contrast, the present case is not about determining the proximate cause of the loss by an enumerated insured peril. The primary dispute between Sizer and Chubb is whether or not the thefts had occurred during the Transit Period. The fact that Excellent Mining’s premises was “eliminated” as the locus of the theft was not to say that the Judge had been compelled to choose between two competing improbable locations or more specifically, stages of the inland transit (ie, Excellent Mining’s premises or some place during the Transit Period). It bears reiterating that we have to be mindful that in the present case, the issue in the context of cl 8.1 of the ICC(A) is whether the theft had occurred prior to or after the risk under the Policy had attached. Where the losses could only have occurred at either one or the other period, any subsequent finding on the probability of one of the periods would logically lead to the opposite conclusion on the other. From this perspective, in our view, the Judge did not err in applying a method of elimination which entailed eliminating the possible locations of the thefts to determine the true state of affairs.

77 In this appeal, Chubb also argues that the Judge failed to make a finding of fact in relation to the possibility of the thefts occurring at MSC’s premises in Penang. To recapitulate, the Judge held (at [81]) that neither parties’ experts had opined on the likelihood of the thefts occurring during the remaining stages of the Transit Period after the shipments left the port in Dar es Salaam. He also noted that, on balance, it would have been very difficult for the thefts to have occurred on the voyage to Penang and at the port in Penang, due to the magnitude and sophistication involved in executing the agreed method of theft. In addition, although Chubb orally submitted that it was possible for the thefts to have occurred at MSC’s premises in Penang, the Judge held (at [82]–[83]) that this belated suggestion was contrary to Dr Petrone’s testimony and Chubb’s case at trial and was thus of little significance.

78 In our view, the Judge did not err in not considering this issue as parties should, in general, be bound by their pleadings (V Nithia (co-administratrix of the estate of Ponnusamy Sivapakiam, deceased) v Buthmanaban s/o Vaithilingam and another [2015] 5 SLR 1422 (“V Nithia”) at [38]). As stated by the Court of Appeal in V Nithia at [40], a departure from this rule is permitted only in limited circumstances, where no prejudice is caused to the other party in the trial or where it would be clearly unjust for the court not to do so.

79 The allegation that the loss could have happened at MSC’s premises in Penang, after the end of the Transit Period, was never Chubb’s pleaded case. To be clear, Chubb had in its Defence (Amendment No 2) (the “Defence”) filed on 24 February 2021 set out the following defence in paras 13 and 14: ⁵

80 What was pleaded in the Defence was thus that the loss had occurred before the Transit Period, and not after.

81 In our judgment, Chubb’s assertion that the Judge failed to consider the possibility of the thefts occurring in Penang is mischievous. Chubb’s criticism that the Judge should have considered that possibility, or that Sizer had failed to prove that the theft did not happen in Penang is ill-founded, not only because it was belatedly canvassed in full for the first time by Chubb in the present appeal, but also because this was not pleaded.

82 In addition, as observed by the Judge, neither parties’ expert had opined on the possibility of the thefts occurring in Penang. In this regard, we reiterate that Sizer’s case is that the thefts occurred during the Transit Period and Chubb’s case was that the thefts had occurred at Excellent Mining’s premises. Whilst Sizer points out in its Respondent’s Case that there was some evidence before the Judge which showed that the thefts could not have occurred in Penang, it was clear from the Judgment that the Judge was of the view that it was not strictly necessary to consider this issue because it was not Chubb’s case that the thefts had occurred after the shipments had been delivered to MSC’s warehouse in Penang (Judgment at [82]–[83]).

83 In our view, to allow Chubb to now depart from its pleaded position on appeal, would be extremely prejudicial to Sizer having not had the opportunity to lead evidence on a new un-pleaded position. We see no reason to depart from the general rule (and more so given that this is raised on appeal) that parties are to be bound by their pleadings.

84 The Judge found that Sizer had discharged its legal burden of proof on the balance of probabilities by showing that the thefts were unlikely to have occurred at Excellent Mining’s premises and along the short trip thereafter to the Bonded Warehouse for customs clearance. Therefore, it was probable that the thefts had occurred during the Transit Period (Judgment at [48]).

85 As outlined above at [55], we divide the analysis of the present case into two stages: (a) first, whether Sizer had prima facie discharged its legal burden of proof on the basis of the testimony of its witnesses, both factual and expert, contemporaneous documentation and the agreed facts; and (b) second, whether Chubb had discharged its evidential burden to prove that the thefts had occurred at Excellent Mining’s premises as per its case at trial. In our view, Sizer has prima facie proved its case and Chubb has failed to rebut Sizer’s case on the merits.

86 We begin with the first inquiry as to whether Sizer had prima facie discharged its legal burden of proof. In order to discharge this legal burden, Sizer’s obligation is threefold. It has to establish that: (a) tin concentrate (as the insured subject matter) was packed into the drums at Excellent Mining’s premises; (b) the losses suffered were as a result of a fortuitous casualty; and (c) the fortuitous casualty occurred after the risks insured under the Policy had attached. We discuss each of these points in turn.

87 It is an agreed fact that the tin concentrate, the subject matter of the insurance, was loaded into the drums at Excellent Mining’s premises and subsequently replaced with iron oxide (see [90] and [94] below). However, there was some distraction in the court below introduced by Chubb suggesting that tin concentrate had never been filled into the drums at Excellent Mining’s premises in the first place. Upon questioning by the Judge, Chubb’s counsel, Mr Yee Mun Howe Gerald (“Mr Yee”), took a surprising position: ⁶

88 Mr Yee referred the Judge to para 5(a) of the Defence, ⁷ which states: ⁸

89 The import of Chubb’s averment in para 5(a) of the Defence is far from precise. In any event, such a defence based on the non-existence of insurable subject matter, if any, was unsubstantiated by evidence adduced at the trial. In particular, this position – that tin concentrate was not filled into the drums – was roundly contradicted by the expert report of Chubb’s key expert witness, Dr Petrone. In summary, Dr Petrone concluded in his expert report that the thefts had occurred at Excellent Mining’s premises. However, the fundamental premise underlying his conclusion was that tin concentrate had been filled into the drums at the outset and were then swapped out for iron oxide while stored at Excellent Mining’s premises. In his expert report, Dr Petrone consistently makes reference to the replacement of the tin concentrate cargo in the drums and even in his conclusion states that “[t]he contents of drums of four shipments of Tin Concentrate from Excellent Mining in Kigali had been replaced with worthless iron oxide at some stage prior to the drums’ arrival at MSC in Penang”. ⁹ If Chubb’s own pleadings are contradicted by its own expert witness’s testimony, little weight should be accorded to the former.

90 In our view, Chubb never maintained the position that the drums were not filled with tin concentrate in the first place. This is evident from: (a) Chubb’s acceptance of the ASOF 4; (b) its written closing statement filed on 1 December 2021 (“Chubb’s Closing Statement”); (c) its reply submissions filed on 8 December 2021; and (d) Mr Yee’s oral closing submissions made on 8 February 2022.

91 As alluded to above, an allegation premised on the fact that tin concentrate had never been filled into the drums would be more fitting as a defence based on the non-existence of insurable subject matter. If Chubb intended to pursue this defence it would have had to plead reliance on cl 1 of the ICC(A). One such instance of the defence of non-existence of subject matter succeeding is in the case of Fuerst Day Lawson Ltd v Orion Insurance Co Ltd [1980] 1 Lloyd’s Rep 656 (“Fuerst”). In Fuerst, the plaintiff was a company trading in essential oils. The defendants were the representative underwriters. In 1976, the plaintiff was insured under an Institute of London Underwriters Companies Combined policy on essential oils in steel or iron drums on steamer and/or steamers and/or conveyances or any other method of transportation from “anywhere to anywhere”, lost or not lost, against all risks. The policy contained the usual Institute Cargo Clauses warehouse to warehouse clause. By 18 contracts concluded on various dates, the plaintiff bought on cost and freight terms from an Indonesian company, Farmaport, 495 drums of essential oils. On discharge the drums were found instead to contain water with slight traces of an essential oil on the surface. The defendant contended that the drums never contained anything other than water with a slight trace of oil. While cl 1 of the ICC(A) was not raised in Fuerst, it is evident that the defendant was relying on a defence of non-existence of insurable subject matter to defend against the plaintiff’s claim. On the facts, Mocatta J concluded that the plaintiff had failed to discharge its burden of proof of establishing on the balance of probabilities that the drums they had agreed to purchase from Farmaport ever started on their transit from Farmaport’s warehouse or any other warehouse. Indeed, the contemporaneous documents concerning the sampling and analysis of the content of the drums was greatly suspect and there were real questions as to reliability. Thus, there was a real doubt as to whether the insured subject matter, ie, essential oils had ever been filled into the drums. As no such defence was pleaded in the present case, Chubb could not seek to argue that the tin concentrate had not been filled into the drums.

92 The only issue for the Judge’s determination, as was patently made clear in Chubb’s defence, is that the loss had occurred before the Transit Period in reliance on cl 8.1 of the ICC(A). This was emphasised repeatedly in Chubb’s written opening statement filed on 26 October 2021 (“Chubb’s Opening Statement”), Chubb’s Closing Statement as well as the agreed issues in the ASOF 4 between the parties.

(a) In the ASOF 4, under the List of Agreed Issues at paras (a) and (b) (see [22] above), the inquiry was clearly focused on whether the loss occurred during the Transit Period or not.

(b) In Chubb’s Opening Statement, it identified the core issue for determination as whether it was liable to indemnify Sizer for the losses arising from the four shipments. In this regard, Chubb highlighted that a “sub issue” was “whether the drums were filled with the Replaced Cargo before or after the ‘commencement of the transit’ under clause 8.1 of the ICC A 1982” [emphasis in original omitted]. ¹⁰

(c) In Chubb’s Closing Statement, it identified the “main issue” as “whether the losses occurred during the period of coverage under the Policy”. ¹¹

93 Further, as detailed in [9(d)] above, there were three documents issued prior to each shipment of drums leaving Excellent Mining’s premises:

Pertinently, each of these certificates affirmed that tin concentrate had been packed into the drums for the Sixth to Ninth Shipments.

94 Whilst the Packing Certificates and the Assay Certificates were not agreed as to their authenticity and content and were thus excluded from the agreed bundle of documents (“Agreed Bundle”), Chubb gains no mileage there. In so far as these certificates go towards showing that tin concentrate had been packed into the drums at outset, this same fact is also supported by the parties’ express agreement in the ASOF 4 that tin concentrate had been packed into the drums for the Sixth to Ninth Shipments. To illustrate, we set out the critical points of agreement from the ASOF 4:

Crucially, it bears reiterating that the ASOF 4 was dated 16 November 2021, ie, the second last day of the trial below. As such, any of Chubb’s earlier objections raised at trial or in its Defence concerning the contents of the drums were clearly abandoned and no longer live at the time.

95 In sum, Chubb did not run a case that the drums were never filled with tin concentrate in the first place. Put differently, Chubb chose not to present a defence based on the non-existence of insurable subject matter as that would have required alleging that the contemporaneous documents were fraudulent; a choice that was not appealing to Chubb. In the end, Chubb accepted that the drums were filled with tin concentrate at the outset and defended Sizer’s claims based on Dr Petrone’s case theory.

96 As we noted at [56]–[58] above, where an “all risks” policy is at play, such as in the present case, Sizer need only demonstrate that the losses were caused by some accident or casualty. With the ASOF 4, the requirement of proof of a fortuitous casualty is met. It was agreed that the losses suffered by Sizer were attributable to thefts of the tin concentrate in the drums, which involved its substitution with iron oxide.

97 We pause here to emphasise the fortuitous nature of the losses suffered by Sizer. Under the “all risks” Policy in the present case, once the parties agreed that theft had occurred causing the loss of the tin concentrate, the requirement of a fortuitous casualty was satisfied. Therefore, Sizer need not go further to prove how the thefts occurred to satisfy its legal burden. Notably, it was not Chubb’s pleaded defence that Sizer and/or Excellent Mining had been complicit in the thefts of the tin concentrate. An allegation that the thefts took place in Excellent Mining’s premises fell short of an accusation that Sizer and/or Excellent Mining had fraudulently shipped iron oxide in the drums instead of tin concentrate or that Excellent Mining’s management and/or staff were in some way involved in the thefts. The Minority Judgment takes the position that complicity need not have to be pleaded by Chubb so long as it did dispute that the thefts had occurred during the Transit Period given that the legal burden of proof was on Sizer (see [191] below). We respectfully disagree. As we will elaborate below, it fell on Chubb to raise the issue of complicity if this was necessary to support its case theory on how and where the thefts were committed.

98 First, Chubb did not run its case on the basis that the loss was brought about by Sizer’s wilful misconduct, which is an express exclusion in cl 4.1 of the ICC(A). If Chubb did intend to allege that Sizer was complicit in the thefts, this necessarily had to be pleaded given that this amounted to an entirely different defence against a claim under the Policy. Pertinently, the complicity of any other party in the thefts, including Excellent Mining, would not have been material to a claim under the Policy in any case, given that the express exclusion in cl 4.1 of the ICC(A) covers only the wilful misconduct of the insured.

99 Second, although Dr Petrone had belatedly sought to suggest at trial that Excellent Mining was complicit in the thefts, the Judge observed that this was but an afterthought which heavily undermined his credibility (Judgment at [136]). Dr Petrone had not alluded to Excellent Mining’s complicity in either his expert report or AEIC. Despite this, he accepted in the course of questioning by the Judge that his case theory that the swap occurred in Excellent Mining’s premises, depended on the complicity of the staff of Excellent Mining. Notably, Chubb mounted a defence based on Dr Petrone’s case theory about the thefts of the drums of tin concentrate when such complicity was never a part of Chubb’s pleaded defence. The Judge strongly criticised Dr Petrone’s evidence, particularly of the manner in which Dr Petrone advanced his case theory. We agree that Dr Petrone’s opinion was put forward with an audacious inadequacy of factual support or basis. In any event, it is a fact that no prosecution in Rwanda followed from the thefts of the four shipments, thus pointing to the lack of culpability of Excellent Mining and Bolloré.

100 Third, complicity is deliberate and intentional conduct that involves persons participating together in wrongdoing. Who the participants are, and each person’s involvement are material facts that ought to have been pleaded by Chubb in order to establish its defence.

101 We next turn to consider whether the operation of the fortuitous casualty (ie, the thefts of the tin concentrate) occurred before or after the risks insured under the Policy had attached.

102 To recapitulate, as the plaintiff suing under the Policy, the legal burden is on Sizer to demonstrate that the losses occurred during the Transit Period (see [61] above). To this end, Sizer produced evidence in the form of witness testimony (by way of four factual witnesses and two expert witnesses) and documents to prove that the risk under the Policy had attached when the thefts/losses occurred. Once Sizer satisfies the court on a prima facie basis of this, the evidential burden shifts to Chubb to either rebut Sizer’s evidence or produce evidence supporting its own counter theory that the theft had occurred in Excellent Mining’s premises prior to the commencement of the Transit Period.

103 In this section, we begin by examining the evidence that Sizer produced to prove that the risk under the Policy had attached when the thefts/losses occurred. In proving its case, it is in our view, perfectly permissible for Sizer to rely on the use of circumstantial evidence to draw a particular inference of fact. An inference of fact, if drawn, is one factor amongst others which the court must consider alongside all the other circumstances of the case. Generally, the use of inferences of fact is an aid available to both parties. To this end, Sizer may justify with the aid of inference that the thefts occurred after customs clearance and during the Transit Period. Thereafter, if the evidential burden shifts to Chubb, Chubb may also rely on inferences to justify its position that the swap occurred at Excellent Mining’s premises.

104 We begin with our observations of several undisputed facts which cumulatively give rise to prima facie proof, whether arising directly or inferentially, that the tin concentrate in drums was stolen during the period of insurance cover (ie, the Transit Period):

(a) The first undisputed fact is that iron oxide instead of tin concentrate had been delivered to the end receiver, MSC, in Penang. It is not disputed that the tin concentrate in the drums were substituted and replaced with iron oxide, after the drums had been sealed. From this, we can be certain that there was loss during the transportation of the drums from Excellent Mining to MSC’s premises in Penang. As agreed, the loss is due to theft which is a fortuitous event.

(b) The second undisputed fact is that all the tin concentrate in the drums was substituted and replaced. This suggests that there was opportunity for tampering with the contents after the sealing of the drums and containers. The Sixth to Ninth Shipments cleared customs at the Bonded Warehouse in Rwanda without incident. Indeed, Chubb accepted that the checks by customs at the Rwanda-Tanzania border are stringent. For instance, any discrepancy between the container seal numbers and the seal numbers stated in the cargo documents would result in the containers and their cargo being seized by the relevant authorities. This too suggests the absence of any tampering of the drums or the 40ft containers when they left Excellent Mining’s premises. The Minority Judgment points out that the lack of tampering of the seals of both the drums and the containers is immaterial given that the parties proceeded on the premise that the thefts had occurred without any such tampering (see [148]–[149] below). This, however, misses the broader point made here that the customs checks were stringent and any anomaly in the shipments would have been flagged out. This included, but was not limited to, any noticeable tampering of the seals on the drums and the containers. Thus, it is a matter of reasoning from a reasonable inference of the facts that the thefts occurred after customs clearance. In relation to our observations in this paragraph and [104(a)] above, we emphasise that the customs clearance and the receipt of iron oxide in Penang are matters of fact that are undisputed.

(c) The third undisputed fact is that there was a casualty that occurred which caused the loss. This can be discerned from the extraordinary or exceptional nature and character of the modus operandi of the thefts given the magnitude of the task involved as per Dr Petrone’s case theory. The Four Shipments followed the normal inland transit which has been described by the parties to be from the Rwanda-Tanzania border to Dar es Salaam and then storage at the port. Therefore, if we take the modus operandi as described, its very exceptional character speaks to a casualty (ie, that something unexpected or fortuitous happened to the tin concentrate in drums to result in the losses) during the Transit Period. There is no need to talk about the seaward carriage as the swap could not have occurred onboard the sea carriers. This was also not contended by the parties. As we have explained at [56] and [58] above, under an “all risks” policy all that has to be shown by Sizer is that a fortuitous casualty had occurred resulting in the loss. Given that this has been agreed by the parties (ie, that the loss was due to theft), Sizer does not have to prove the persons involved and/or the precise manner or mode of how the swap of the drums’ contents had taken place in order to establish its claim under the Policy.

(d) The fourth undisputed fact is that the first five shipments under the First Contract were received without incident. These five shipments had followed the same journey as the four later shipments where losses were sustained. Comparing these two similar journeys, it stands to reason that since the Four Shipments had cleared customs without incident and the iron oxide was received at the other end in Penang, the thefts/losses could only have occurred during the period of the insurance (ie, the Transit Period).

105 The Minority Judgment expresses the view that the four undisputed facts raised above are individually immaterial to the question of whether the thefts occurred during the Transit Period (see [145]–[151] below). With respect, this fails to appreciate the interaction between these four undisputed facts and the inferences to be drawn from them as a whole. We know that despite tin concentrate having been packed into the drums at Excellent Mining’s premises, iron oxide was subsequently discovered in the drums at MSC’s premises in Penang (ie, the first undisputed fact) and that a casualty, in the form of theft, caused this loss (ie, the third undisputed fact). We also know that these drums cleared customs at the Bonded Warehouse in Rwanda without incident, which demonstrates that the thefts could not have occurred before then (ie, the second undisputed fact). Further, it is a fact that the only difference between the successfully delivered first five shipments under the First Contract and the four subsequent stolen shipments under the First and Second Contracts is that for the latter shipments, iron oxide (instead of tin concentrate) was received in Penang (ie, the fourth undisputed fact). This points toward the thefts occurring during the Transit Period, especially given that both the successful and stolen shipments travelled overland on the same route from Kigali to Dar Es Salaam and cleared customs without incident, with only the overland transit of the last Four Shipments being unsuccessful because of theft. Therefore, pulling these strands together, an inference can be drawn that it is more likely than not that the thefts occurred during the Transit Period.

106 We now turn to the documentation for the Sixth to Ninth Shipments. We will discuss some critical pieces of documentation and their significance.

107 First, there is documentary evidence showing that the drums containing tin concentrate had cleared customs in Kigali intact. It is helpful at this juncture to recall the parties’ agreed position that under the First and Second Contracts, the insurer’s risk attached after the drums had left Excellent Mining’s premises in respect of the Sixth Shipment and after customs had been cleared at the Bonded Warehouse in Kigali in respect of the Seventh to Ninth Shipments (see [0] above).

108 As we noted above at [94], the ASOF 4 recorded the parties’ agreement that tin concentrate had been packed into the drums at Excellent Mining’s premises. This comports with the details contained in the Packing Certificates, Assay Certificates and ICGLR certificates for each shipment which were issued before the drums were transported from Excellent Mining’s premises. Subsequently, after the tin concentrate was weighed and packed into the drums, the small opening of each drum was welded shut and sealed with Precintia clips before a layer of white alkyd paint was applied on the lids.

109 Before the drums were stuffed into the 40ft container outside Excellent Mining’s Premises, the drums were checked by RMB’s certification officer and at the time of stuffing, representatives from ASIR and RMB did not notice any sign of tampering or theft (see [9(f)] above). Thereafter, the 40ft container with a temporary seal affixed would be transported to the Bonded Warehouse in Kigali, a 45 to 60 minutes’ drive away, where customs was cleared. Significantly, the documents adduced by Sizer from the RRA demonstrate that the drums containing tin concentrate remained intact after customs had been cleared at the Bonded Warehouse. We turn to discuss the pertinent evidence in this regard.

110 On 31 October 2019, Sizer wrote to RRA requesting copies of the inspection certificates in respect of the Sixth to Ninth Shipments. On 1 November 2019, RRA replied to Sizer by way of letter attaching the relevant inspection certificates. On 13 November 2019, RRA sent a further letter stating that: ¹²

111 This correspondence was included in the parties’ Agreed Bundle, and even though Chubb argues that it had only agreed to the authenticity but not the content of the various documents, a perusal of the record shows no objection to the contents of these documents were taken in the course of the trial below. We accept that formal proof of the documents is not the same as accepting the truth of their contents into evidence. As stated by the Court of Appeal in Jet Holding Ltd and others v Cooper Cameron (Singapore) Pte Ltd and another and other appeals [2006] 3 SLR(R) 769 (“Jet Holding”) at [44]:

112 That being said, the Court of Appeal when on to hold (at [51]):

113 Not only did Chubb not object to the content of the documents during trial, it also accepted the truth of the contents by relying on them to support its own arguments and defence. In Chubb’s Defence at para 13(d), it pleaded as follows: ¹³

114 Indeed, in Chubb’s Closing Statement it maintained the position that there was no evidence of tampering of the seals on the drums as well as the containers after customs had been cleared at the Bonded Warehouse. Chubb even noted that its own witness, Mr Sim, had testified to the same effect by stating that throughout the entire duration that the drums were in the 40ft and 20ft containers, there was no evidence of any tampering. Thus, Chubb’s position is entirely consistent with the contents of the RRA documents which stated that after documentary and physical verification of the drums by customs, the seals and the drums were found intact at the time of customs clearance.

115 Second, the Stuffing Reports issued by Bureau Veritas subsequently also recorded that all the drums in the Sixth, Eighth, and Ninth Shipments were in good condition when they were transferred from the 40ft containers to the 20ft containers at the port in Dar es Salaam after the overland journey from the Bonded Warehouse. As for the Seventh Shipment, it was recorded that two out of the 30 drums shipped had been damaged by stevedores. However, nothing turns on this as neither party relied on this fact.

116 In its Closing Statement in the proceedings below, Chubb highlighted that if representatives from Bureau Veritas had been called to give evidence, they would surely have testified that “factually, there was nothing illegal, improper and/or suspicious that took place ‘on the ground’ in respect of the 40 foot containers, the 20 foot containers, the seals of all the containers and the contents of the drums” [emphasis in original omitted]. In short, Chubb’s position was that the representatives would have given evidence that “no theft occurred at [Dar es Salaam]”. ¹⁴ It is beyond peradventure that despite Chubb’s initial reservation to agree to the contents of the Stuffing Reports in the Agreed Bundle, its subsequent conduct in relying on the truth of their contents during cross-examination and in its final submissions demonstrates that it had lifted any prior reservation or qualification as to the truth of their contents. The result of this is that the documents were included in the Agreed Bundle without qualification and therefore stood as evidence in the case before the court (see Jeffrey Pinsler, Evidence and the Litigation Process (LexisNexis, 7th Ed, 2020) at paras 17.044–17.045).

117 However, an additional question that arises pertains to the admissibility of these documents, specifically with regard to their nature as hearsay evidence. This may be answered shortly. With the amendments to the Evidence Act (Cap 97, 1997 Rev Ed) (“EA”) in 2012, s 32(1)(k) was introduced to admit hearsay evidence by agreement. Unlike in Goldrich Ventures Pte Ltd and another v Halcyon Offshore Pte Ltd [2015] 3 SLR 990 at [129], where s 32(1)(k) of the EA was not satisfied because the parties’ agreed bundle of documents limited the scope of their agreement to the authenticity of the documents, and not to the truth of their contents. Chubb’s reliance on the truth of the contents of these documents effectively served to withdraw any earlier limitation placed on the scope of their agreement as to the veracity of their contents. By Chubb’s conduct at trial, these documents were accepted and admissible as evidence having satisfied s 32(1)(k) of the EA.

118 Having regard to all the above matters, we are of the view that this points toward the Four Shipments being prima facie intact after customs clearance (ie, after commencement of the Transit Period and during the period of insurance). The evidential burden of proof thus shifts to Chubb to rebut Sizer’s evidence or to adduce evidence to support its counter theory that the loss had occurred in Excellent Mining’s premises after the drums were filled with tin concentrate and sealed.

119 The Minority Judgment again emphasises that it is immaterial that no tampering of the seals on the drums and the containers were recorded in the above documents (see [152]–[157] below). These documents, however, do not simply demonstrate the lack of tampering of the seals. They go further to show that no anomalies about the physical condition of the drums or the containers were raised despite physical verification of the cargo by customs and Bureau Veritas. In fact, where there was damage, this was duly recorded as seen in the case of the Seventh Shipment where Bureau Veritas recorded that two out of the 30 drums shipped had been damaged by stevedores (see [115] above).

120 In the court below, Chubb led evidence to support its counter theory that the thefts had occurred at Excellent Mining’s premises before the Transit Period and thus before the risk attached under the Policy. In doing so, Chubb at the same time sought to discredit Sizer’s evidence pointing to the occurrence of the theft during the Transit Period. It bears repeating that at the trial, Chubb did not take the position that the thefts could have occurred at MSC’s premises in Penang (ie, after the end of the Transit Period) (see [77]-–[83] above). We agree with the Judge that it was unlikely that the thefts had occurred at Excellent Mining’s premises. Chubb did not discharge its evidential burden of proof with the result that Sizer’s evidence stood unrebutted. We will elaborate on this view later after discussing Chubb’s counter theory in light of the scope of this appeal.

121 We digress briefly also to address the minority’s suggestion that the swap could have occurred outside Excellent Mining’s premises but before the cargo eventually left for the Bonded Warehouse for each shipment (see [194]–[195] below). Despite acknowledging that this was not Chubb’s case theory, the minority notes that the burden was not on Chubb to prove a positive case if there was no shift in the evidential burden. However, given our analysis above, we are of the view that the evidential burden shifted to Chubb as Sizer had discharged its legal burden on a prima facie basis. In any event, we make the following observations in regard to the suggestion that the swap could have occurred outside Excellent Mining’s premises. First, the undisputed evidence is that the drums were stored in Excellent Mining’s premises pending receipt of the export documents. Further, another piece of undisputed evidence is that the drums were loaded into the 40ft container directly from Excellent Mining’s premises in the presence of ASI and RMB representatives. Second, as the minority recognised, Chubb did not run a case that the swap could have occurred outside Excellent Mining’s premises and before the drums’ arrival at the Bonded Warehouse. Thirdly, customs clearance of the four shipments are matters of fact that are undisputed (see [104(b)] above).

122 Chubb’s counter theory that the thefts occurred in Excellent Mining’s premises is very much bound up with Dr Petrone’s case theory on the modus operandi of the theft. A case theory is not a fact and every fact in support of a case theory must be proved; any inferences of fact can only be drawn from proven facts. Earlier, we set out at [34]–[42] above the Judge’s reasons for rejecting the bases of Dr Petrone’s case theory, with the result that no inferences can be drawn from the rejected facts. Plainly, the Judge was not persuaded by the factual aspects of Dr Petrone’s case theory. One of Chubb’s main difficulties in establishing the bases of Dr Petrone’s case theory lay in his lack of credibility as an expert witness. In this regard, we noted at [42] above that the Judge found Dr Petrone’s investigation lacked objectivity and was “unsatisfactory, superficial and substandard”. In any event, as mentioned at [54] above, Chubb does not challenge the Judge’s findings on the 3Ts minerals and the identity of the white alkyd paint on the drums’ lids in this appeal. In any case, we see no reason to interfere with the Judge’s findings at [34]–[42] above.

123 Therefore, Chubb’s primary argument in this appeal is that the Judge impermissibly reversed the burden of proof, by adopting a process of elimination to find that Sizer had discharged its burden of proving that the thefts had occurred during the Transit Period by virtue of his disagreement with Chubb that the thefts had occurred at Excellent Mining’s premises. In so doing, the Judge failed to consider the possibility that the thefts had occurred at MSC’s premises in Penang. In the alternative, Chubb argues that there was insufficient evidence to support the Judge’s finding that the thefts had occurred during the overland transport of the drums from Kigali to Dar es Salaam. For the reasons explained above at [60]–[65], we are unable to agree that the Judge had in any way reversed the burden of proof. Moreover, we are also of the view that the Judge was wholly justified in not considering Chubb’s argument that the thefts could have occurred in Penang (see [77]–[83] above).

124 To be clear, the inquiry here is whether the insurance period had commenced at the time the thefts of the tin concentrate had occurred. Based on the evidence led at the trial below and as accepted by the Judge, it was logistically impossible for the thefts to have had happened at Excellent Mining’s premises. However, in this appeal not only does Chubb agree with the Judge as to the complexity of the logistics involved in the thefts, its only argument as to the Judge’s finding of logistical impossibility of the thefts happening at Excellent Mining’s premises was that the Judge should have applied the same difficulties to the overland transport of the drums from Kigali to Dar es Salaam. In short, having accepted that it was logistically impossible for the thefts to have happened at Excellent Mining’s premises, it is no answer for Chubb to argue that the same difficulties applied to another segment of the Transit Period. It was incumbent on Chubb to either show why the Judge had erred in his reasoning or put forward a positive case that the thefts had happened outside of the Transit Period.

125 In this appeal, Chubb did not adequately challenge the evidence led by Sizer in the trial below, which had been accepted by the Judge. To recapitulate, Sizer’s evidence at trial was that the thefts could not have happened at Excellent Mining’s premises due to logistical difficulties, time and space constraints, the location, and the security measures present (see [16] above). Having considered the evidence from the various witnesses, the Judge agreed that it was “clearly logistically impossible for the thefts to have taken place at Excellent Mining’s premises” (see Judgment at [68]). We have already detailed the Judge’s reasons in reaching this conclusion at [24] above. As such, the only way for the thefts to have occurred in Excellent Mining’s premises was with the complicity of Excellent Mining’s staff, for which there was no evidence (Judgment at [65]–[66] and [68]). -

126 Further, Chubb points to the Judge’s decision not to disregard the NPPA Report as an example of the Judge’s reversal of the burden of proof. Notably, although the Judge acknowledged that it was unfortunate that Sizer had failed to call the official from the NPAA in charge of the investigation to testify in court, he nonetheless relied on the NPAA Report’s finding that the thefts did not occur in Rwanda as Chubb had not contested its admissibility or produced reasonable or credible evidence before the court to justify jettisoning the findings therein (Judgment at [97]). With respect, we disagree with the Judge’s approach. As stated by Judith Prakash J (as she then was) in Keimfarben GmbH & Co KG v Soo Nam Yuen [2004] 3 SLR(R) 534 at [17], the admissibility of hearsay evidence is governed by the EA and as the provisions of the EA are mandatory, hearsay evidence cannot be admitted even if parties had failed to object to it. On the facts of the present case, as the maker of the report was not called to testify as to its contents, the NPPA Report should have been considered inadmissible hearsay evidence.

127 In any case, we note that the Judge did however express that the NPPA Report should be treated with caution and that its effect was “limited to whether Excellent Mining and Bolloré Logistics were complicit in the thefts in Rwanda” [emphasis in original] (Judgment at [97]).

128 Chubb’s counter theory is that the thefts had occurred at Excellent Mining’s premises in accordance with the modus operandi suggested by Dr Petrone. Ultimately, the question is whether such a counter theory is viable given the magnitude of the task of substituting and replacing a large quantity of tin concentrate in drums with iron oxide. To prove its counter theory and discharge its evidential burden, facts in support must be identified and established. This should be viewed through a pragmatic lens and with a healthy dose of common sense. In our view, Chubb’s counter theory simply does not cohere with common sense. The modus operandi itself is bizarre considering the mammoth task involved. It would make more sense to drive away with the container load of tin concentrate. Chubb may have suspected insurance fraud but, without proof, it has not pleaded this as a defence. Similarly, as it is not Chubb’s defence that the drums were filled with iron oxide from the beginning, Chubb resorted to defending the claim saddled with Dr Petrone’s case theory. That said, as emphasised before at [56]–[58] above, in the case of an ICC(A) “all risks” cover, bearing in mind the observations in British & Foreign Marine Insurance Co Ltd, there is no need to inquire into how the substitution and replacement of tin concentrate with iron oxide occurred. Importantly, it is an agreed fact that the loss was fortuitous. We are of the view that Chubb has not discharged its evidential burden. It has not succeeded in discrediting Sizer’s evidence or managed to prove its counter theory that the thefts had occurred at Excellent Mining’s premises.

129 We turn to address Chubb’s arguments that the Judge erred in concluding on the evidence that the thefts had occurred during the overland transport of the drums. Chubb raises three reasons. First, the Judge should not have accepted Mr Wheeler’s expert evidence that the thefts had taken place during the inland transit between Kigali and Dar es Salaam. Second, there was no direct and/or credible evidence to support the Judge’s finding that the drivers and their assistants transporting the Sixth to Ninth Shipments had been compromised, and that the security measures during this journey “could not have been inadequate”. In this regard, Chubb asserts that the Judge’s findings contradicted facts that he had accepted. Third, the Judge did not consider that the same logistical difficulties associated with the thefts occurring at Excellent Mining’s premises would similarly have had applied during the overland journey.

130 To recapitulate, the Judge found that it was more probable than not that the thefts had occurred sometime after the containers cleared customs at the Bonded Warehouse in Kigali during the land transport leg from Kigali to Dar es Salaam (see [84] and [89] of the Judgment). The Judge gave the following reasons for concluding as such:

131 The Judge also accepted Mr Wheeler’s evidence that there were significant logistical difficulties in swapping the tin concentrate with iron oxide at Excellent Mining’s premises (see [27] above).

132 We have dealt with [130(b)] at [126]–[127] above.

133 It also bears repeating that in this appeal, Chubb does not challenge the Judge’s findings in [130(c)]–[130(d)] above. However, given that the Minority Judgment expresses some views on this, we consider these points briefly. In essence, although the Minority Judgment expresses agreement with the Judge that both the presence of the 3Ts minerals in the swapped iron oxide and the use of the white alkyd paint on the drums’ lids do not conclusively demonstrate that the thefts had occurred at Excellent Mining’s premises, the minority is of the view that these factors reinforce the likelihood of complicity (see [183]–[188] below). Quite apart from the fact that complicity was not pleaded by the parties, given that the Minority Judgment agrees with the Judge that neither of these points conclusively point to the thefts having occurred at Excellent Mining’s premises, it is in our view, somewhat speculative to conclude that they reinforce the likelihood of complicity. As stated in [100] above, complicity is fact-specific in that an allegation of complicity must involve others in the unlawful activity. Thus, for such a conclusion to be meaningful, it is necessary to link the presence of the 3Ts minerals and the use of white alkyd paint to an alleged complicity.

134 As for [130(a)] above, we agree with the Judge that with or without security (ie, either the security guards or the presence of CCTVs at Excellent Mining’s premises), the swap of tin concentrate for iron oxide contained in the same drums could not have been a clandestine operation given the huge quantity of tin concentrate and iron oxide involved in respect of each of the affected shipments. As such, the minority’s reservations (at [160]–[174] below) regarding the presence and efficacy of these security features are not material. Excellent Mining’s director who lived next to Excellent Mining’s premises would have noticed the commotion associated with an instance of theft, not to mention theft over four separate occasions. This, coupled with the significant logistical difficulties in swapping the tin concentrate with iron oxide at Excellent Mining’s premises – which finding has not been challenged by Chubb – quite clearly eliminates Excellent Mining’s premises as the location where the thefts had taken place. We also agree with the Judge that there was no evidence to prove, on a balance of probabilities, that Excellent Mining’s staff were complicit in the thefts. Indeed, complicity on the part of Excellent Mining or its staff was not pleaded by either of the parties, we repeat our views expressed at [97]–[100] above.

135 Even if the Judge erred in concluding that security was weakest during the inland transit and therefore the thefts were most likely to have had occurred there, this is not fatal. Sizer’s legal burden of proof was to prove that the thefts had occurred during the period of insurance cover. As we have established above, the inferences from the four undisputed facts (see [104] above) and the documentation for the Sixth to Ninth Shipments showed on a prima facie basis that it was more likely than not that the thefts occurred during the Transit Period. In summary, iron oxide instead of tin concentrate was discovered in the drums at MSC’s premises in Penang, and this was agreed to have been caused by theft, a fortuitous casualty. The drums had cleared customs without incident and the first five shipments under the First Contract which followed the same journey had been delivered without incident (see [105] above). Documentation in the form of RRA’s letter to Sizer indicated that after documentary and physical verification of the drums by customs, the seals and the drums were found intact at the time of customs clearance (see [114] above). Further, the Stuffing Reports issued by Bureau Veritas recorded that all the drums in the Sixth, Eighth and Ninth Shipments were in good condition when they were transferred from the 40ft containers to the 20ft containers at the port in Dar es Salaam after the overland journey from the Bonded Warehouse. Whereas for the Seventh Shipment, it was recorded that two of the 30 drums shipped had been damaged by stevedores (see [115] above). Thus, given our conclusion above that Chubb had not discharged its evidential burden to show that the thefts had occurred prior to the commencement of the Transit Period (ie, at Excellent Mining’s premises) and the fact that it was not argued by Chubb that the thefts occurred after the Transit Period (ie, at MSC’s premises in Penang), it stands to reason that the thefts occurred during the Transit Period.

136 In addition, as explained at [58] above, Sizer is not required under an “all risks” cover to prove how the thefts had occurred. It was therefore not necessary for the Judge to comment on how the thefts could have had occurred, eg, with the complicity of the truck drivers and their assistants. Simply put, the losses of the tin concentrate were agreed. It was also agreed that they had been occasioned due to theft, a fortuitous casualty.

137 The Minority Judgment stresses that the issue at the heart of this case is more likely than not an issue of complicity (see [178]–[182] below), and that this is most apparent by the thieves’ painstaking efforts to avoid detection by replacing the contents of the drums, re-welding and re-painting each drum lid, etc. With respect, we are unable to agree with the minority that the issue of complicity is material to the resolution of this appeal. The minority’s observations on the issue of complicity in the context of the modus operandi of the thefts are entirely based on Dr Petrone’s case theory. We expressed our doubts at [128] above on the viability of this case theory. The facts to build such a case theory were not proved. Accordingly, any discussion on the active concealment and avoidance of detection by the thieves premised on such a case theory is but speculative. We also repeat that there is no need for Sizer to prove more than what it has done since theft was agreed, which satisfied the requirement of a fortuitous casualty under the Policy. It was for Chubb to raise the issue of complicity if necessary to support its own defence (see [97] above). The Judge in the court below rightly dealt with Sizer’s claim based on the contours of the parties’ own cases, in particular the ASOF 4 and agreed issues. As we explained earlier, Chubb did not run a case premised on insurance fraud (see [98] above), neither did Chubb avail itself of an alternative defence based on the non-existence of insurable subject matter (see [91] above). Instead, Chubb elected to explain the presence of iron oxide in the drums upon their arrival in Penang based on Dr Petrone’s case theory which, in our view, defies common sense. While Dr Petrone conceded that his case theory ultimately depended on the complicity of Excellent Mining’s staff, this was not pleaded, and no evidence was adduced in support of this. We reiterate that as complicity involves deliberate and intentional conduct, who the participants are, and each person’s involvement are crucial material facts that ought to have been pleaded (see [98] above). In any case, complicity itself goes toward the modus operandi of the thefts which is strictly unnecessary for the court to determine in order to arrive at a conclusion on recovery under the Policy.

138 Chubb argues that the Judge should not have relied on the evidence of Mr Wheeler to arrive at the opinion that the theft had likely occurred during the overland phase of the Transit Period. At the beginning, the position Mr Wheeler adopted in his expert report dated 16 November 2020 was that the thefts had likely occurred during the overland transport of the drums. However, in his oral evidence in court, Mr Wheeler stated on further cross-examination that his opinion was that the thefts likely occurred at the port in Dar es Salaam. Although the Judge eventually rejected Mr Wheeler’s oral testimony that the thefts were likely to have had occurred at the port in Dar es Salaam, he preferred and accepted Mr Wheeler’s written expert opinion that it was more likely than not that the thefts occurred during the inland transit. In our view, the thrust of Mr Wheeler’s expert evidence is that the opportunity for tampering of the shipments was after customs clearance at the Bonded Warehouse.

139 For the sake of argument, even if the Judge erred in his assessment of the evidence on the security at Excellent Mining’s premises (see [24] above) and erred in placing too much weight on NPPA Report (see [126] above), the undisputed facts and the inferences to be drawn from them (see [104] above) and the unchallenged documentary evidence and the inferences to be drawn from them (see [106]–[118] above) provide a preponderance of evidence establishing on the balance of probabilities that the thefts had occurred after customs clearance at the Bonded Warehouse and during the Transit Period.

140  Having considered all the evidence, we see no reason to set aside the Judge’s decision in favour of Sizer. We therefore dismiss the appeal. We order Chubb to pay costs of $80,000 (all in) to Sizer. The usual consequential orders apply.

141 The majority of this court has decided to dismiss the present appeal by Chubb against the Judge’s decision to allow Sizer’s claim for the undisputed sum of US$1,154,508.94 for the loss of the Four Shipments under the Policy. I have read the Majority Judgment in draft and agree with it with regard to the conclusion that the Judge did not err when he applied a process of elimination (see [76] above) and where he did not consider the possibility that the thefts of the tin concentrate might have occurred at MSC’s premises in Penang (see [77]–[83] above). However, with regard to the substantive decision of whether the Judge erred in considering the evidence, I have reached a different conclusion from the majority. In my judgment, I shall only address the latter issue and unless otherwise indicated, use the same abbreviated terms as those in the Majority Judgment.

142 The background facts and history of the proceedings between the parties are set out in the Majority Judgment at [4]–[21] above. In gist, the Judge was of the view that Sizer had proved (Judgment at [139]):

It is my view that in coming to the conclusion above, the Judge erred in his assessment of the evidence for the reasons which I elaborate later below.

143 First, I address the Majority Judgment which concludes that there was prima facie proof that the tin concentrate was swapped with iron oxide during the Transit Period and the evidential burden was then shifted to Chubb to prove otherwise. This was not the approach of the Judge or Sizer as neither suggested that there was such prima facie proof.

144 The Majority Judgment identifies four undisputed facts which cumulatively give rise to prima facie proof that the thefts occurred during the Transit Period (see [104] above).

145 The first undisputed fact is that iron oxide instead of tin concentrate was delivered to the end receiver in Penang (see [104(a)] above). While I agree, this is not material to the question whether the thefts occurred during the Transit Period (“the Question”).

146 The third undisputed fact is that there was a casualty that occurred which caused the loss (see [104(c)] above). But this is also not material to the Question.

147 I come to the second undisputed fact which is that the tin concentrate was substituted and replaced (see [104(b)] above). This suggests that there was opportunity for tampering with the contents after the sealing of the drums and containers after clearance at the Bonded Warehouse in Rwanda.

148 I am of the view that while there was opportunity to tamper after such sealing, this is neither here nor there. There was also opportunity for the theft to have occurred before each shipment left the premises of Excellent Mining. That Chubb accepted that the checks by customs at the Rwanda-Tanzania border are stringent is also neither here nor there. That any discrepancy between the container seal numbers and the numbers stated in the cargo documents would result in the cargo being seized by the relevant authorities is also not material. There was no suggestion that any of the temporary seals affixed on the doors of the 40ft containers at Excellent Mining’s premises were found to be tampered with when the containers reached the Bonded Warehouse.

149 In addition, the parties proceeded on the premise that the thefts had occurred without tampering of the Precintia clips spot welded on the sealed bungholes of the drums. This was supported by Mr Yeoh’s observations on the same after inspecting the drums upon their arrival in Penang (see [32] above). Thus, non-tampering of the Precintia clips does not mean there was no theft and it still leaves open the question as to when the theft took place.

150 The fourth undisputed fact is that the first five shipments under the First Contract were received without incident (see [104(d)] above). These five shipments had followed the same journey as the four later shipments where losses were sustained. The Majority Judgment says that comparing these two similar journeys, it stands to reason that since the Four Shipments had cleared customs without incident, the theft could only have occurred during the Transit Period, ie, meaning that it could only have occurred after clearing customs at the border.

151 With respect, the comparison with the first five shipments for the purpose of determining the Question is neither here nor there. There was no theft of those shipments so they are not material to the Question as such.

152 I next address the documents which the Majority Judgment relies on to conclude that the tin concentrate was stolen during the Transit Period (see [106]–[117] above). The Majority Judgment (at [110]) refers to a letter sent by RRA which states: ¹⁵

153 The Majority Judgment notes that Chubb pleaded in its Defence at para 13(d) that: (a) checks by customs at the Rwanda-Tanzania border are stringent and any discrepancy between the numbers of the actual container seals and those stated in the cargo documents would result in the container trucks being seized by the relevant authorities; and (b) container seals in respect of the relevant containers were found to be intact during cross-stuffing of the cargo into the respective 20ft containers at Dar es Salaam (see [113] above).

154 As I have mentioned above, the fact that the container seals were not broken is not material. There was no evidence of any discrepancy at any time, whether before or after the journey to the Bonded Warehouse, between the numbers on the temporary or permanent container seals and those in any document. There was also no evidence of tampering of the Precintia clips on the drums at any time before or after the journey to the Bonded Warehouse.

155 To the extent that the RRA letter started that the seals of all the tin concentrate drums were found intact at the border, this is neither here nor there because, as mentioned above at [149], the Precintia clips on the drums were still found to be intact in Penang.

156 In so far as the letter from the RRA mentioned physical verification of the cargo it is unclear what this meant. There is no suggestion that the RRA opened the drums to examine the contents. If they had done so, the Precintia clips on the drums would have been broken. There is also no evidence that they had opened the drums without breaking the clips.

157 The Majority Judgment relied on the Stuffing Report issued by Bureau Veritas which recorded that all the drums were in good condition when they were transferred from the 40ft containers to the 20ft containers at the port in Dar es Salaam after the overland journey from the Bonded Warehouse (see [115]–[117] above). In my view, this would then suggest that the thefts did not occur during that overland journey and might have occurred even before the cargo reached the Bonded Warehouse. It appears that no tampering of the drums was noticed either when the cargo reached the Bonded Warehouse at the border or at the port. Hence, the absence of noticeable tampering of the drums is neither here nor there.

158 In the circumstances, I am not persuaded that a prima facie case has been established that the thefts had occurred during the Transit Period. This is important as the evidential burden then still remains with Sizer. As mentioned, the Judge was of the view that Sizer had discharged its burden. With respect, I reach a different conclusion for reasons stated below.

159  The Judge concluded that the “security was at its highest at Excellent Mining’s premises” (Judgment at [138(a)]). The Judge appeared to have drawn his conclusion from the evidence summarised at [24] above. For present purposes, I highlight five pieces of evidence that are pertinent:

(a) First, the compound walls, thorned fences and a locked metal gate of Excellent Mining’s premises (Judgment at [56]).

(b) Second, the presence of CCTV cameras at Excellent Mining’s premises (Judgment at [56]).

(c) Third, the four security guards patrolling the premises around the clock in shifts (Judgment at [57]).

(d) Fourth, the fact that Mr Sindikubwadbo lived with his family in a house adjacent to Excellent Mining’s premises (Judgment at [57] and [68]).

(e) Fifth, the fact that the MFO (from RMB) who was stationed at Excellent Mining’s premises, along with ITSCI representatives, would conduct routine spot checks throughout the day on the drums of tin concentrate at Excellent Mining’s premises (Judgment at [59]–[60]).

I shall address each of these pieces of evidence in seriatim.

160 In Mr Sindikubwadbo’s AEIC filed on 5 August 2021 (“Mr Sindikubwadbo’s AEIC”), he stated that Excellent Mining’s premises are “surrounded by compound walls and are securely fenced up…[and] [t]he compound walls were fitted with thorn fences to prevent unauthorised access”. Further, there was also a “metal gate at the entrance of the Premises which would be locked at the end of every day”. ¹⁶ Although, the photographs annexed to Mr Sindikubwadbo’s AEIC did not exhibit any “thorn fences”, on balance, it could not be said that the Judge’s finding that there were compound walls and fences surrounding Excellent Mining’s premises was against the weight of the evidence.

161 Turning next to the Judge’s finding that there were CCTVs at Excellent Mining’s premises, in my view this finding should be treated with caution for the following reasons.

162 First, in Mr Sindikubwadbo’s AEIC, he does not even mention the presence of CCTVs in Excellent Mining’s premises, neither did any of the photographs appended to the AEIC show the presence of CCTV cameras.

163 Second, under cross-examination, Mr Sindikubwadbo conceded that the pictures taken of Excellent Mining’s premises did not show any CCTV cameras covering the area where the tin concentrate was processed. Mr Sindikubwadbo also conceded that the pictures taken of the area where the drums were stored did not exhibit any CCTV cameras, although he claimed that the photographs simply did not capture the CCTV cameras. While Mr Sindikubwadbo argued that the cameras in those areas were not captured in the photos taken of the premises, this was nothing more than a bare allegation.

164 In fact, the only evidence of a CCTV camera present on the premises was provided by Dr Petrone, who had taken a photo of one CCTV camera and documented in his report that it did not even cover the area where the drums were stored. In this regard, I note that Dr Petrone also requested for recordings from the CCTV but was told that no camera was in place or functioning at that time.

165 Further, Mr Sindikubwadbo also conceded under cross-examination that there was no video footage of the shipments captured on the CCTV. According to Mr Sindikubwadbo, this was because the cameras only stored the video footage for a period of ten days: ¹⁷

166 In summary, apart from the oral evidence of Mr Sindikubwadbo, there was no evidence that any CCTV cameras covered the area in which the drums were stored or processed. The only unrebutted evidence available, was provided by Chubb’s witness, Dr Petrone, which showed a single CCTV camera that would not have been able to provide any form of security or surveillance over the drums while they were stored at Excellent Mining’s premises before their onward transport to the Bonded Warehouse.

167 As for Mr Sindikubwadbo’s claim that he had hired four security guards to patrol Excellent Mining’s premises in 12-hour shifts comprising of two guards at any time, this was not backed up by any physical records. No records were made of the guards’ movement, of which guards had been working at the relevant time, or any other documentary evidence that the guards were employed.

168 Notably in Dr Petrone’s report, no mention was made of the presence of security guards at Excellent Mining’s premises when he visited it in October 2018.

169 In this regard, it is interesting to note that Dr Petrone also provided documentary evidence from the ITSCI that after an unrelated theft of moneys at Excellent Mining’s premises in 2019, Excellent Mining was advised to beef up its security measures, which was done subsequently in March 2019 by the hiring of security agents and installation of cameras. In Dr Petrone’s re-examination, he stated that it was his conclusion that the CCTV cameras and security guards were only instituted after this incident of theft in 2019. That being said, the ITSCI documentation may not be admissible as it constitutes hearsay evidence, and in any event, the Judge noted that Dr Petrone had not spoken to any representative from the ITSCI and thus was not able to provide any direct evidence (Judgment at [134(c)]).

170 In my view, there are significant doubts as to whether there were even security guards present at Excellent Mining’s premises at the time the tin concentrate was swapped out for iron oxide, and it cannot be said that Sizer had properly discharged its burden to show that this was a factor which would have made the theft at Excellent Mining’s premises improbable.

171 While it is not disputed that Mr Sindikubwadbo’s residence was located adjacent to Excellent Mining’s premises at the material time, for reasons which I will discuss below, this would not have been an important factor if one considers that Mr Sindikubwadbo might have been complicit in the thefts of the tin concentrate.

172 In his judgment, the Judge appeared to place significant weight on the fact that the MFO who was stationed at Excellent Mining’s premises during working hours, had made daily checks on the drums to ensure that they were not tampered with (Judgment at [60]), and the Judge had categorised Dr Petrone’s failure to speak to any representative or MFO from RMB as an omission of an important task (Judgment at [134(a)]).

173 In my view, the Judge erred in placing significance on the presence of the MFO at Excellent Mining’s premises. First, the evidence of the MFO having conducted the purported checks on the drums at the Excellent Mining’s premises was provided by Mr Nkomeje who was the colleague of the MFO, Ms Laurence. Ms Laurence herself was not called as a witness, neither did she provide any statements putting on record what she had done as part of her duties, nor was any explanation given for her absence in the proceedings below. It should also be noted that evidence of Ms Laurence checking the drums daily was not found in Mr Nkomeje’s AEIC, and was only stated for the first time on the witness stand. Further, Mr Nkomeje was testifying as to what the MFO would have done but did not state that this was what he knew Ms Laurence had actually done. Second, if it was Sizer’s case that because the drums were checked regularly for tampering at Excellent Mining’s premises daily by the MFO and therefore that the thefts could not have happened there, the onus would have been on Sizer to provide evidence that such daily checks had taken place.

174 In summary, the evidence does not support the Judge’s conclusion that the security at Excellent Mining’s premises was at the highest amongst all the other possible locations and that therefore the thefts were unlikely to have occurred before the cargo left the premises for the Bonded Warehouse. Even if the security at the premises was relatively higher than during the route to the Bonded Warehouse or to the port at Dar es Salaam, there was also the important question of complicity which I will come to later.

175 The Judge also placed significant reliance on the NPPA Report. In this regard, the Judge held that the parties agreed that such a report was made (Judgment at [16] and [90]), and that even though it was unfortunate that the official in charge of the investigation was not called to testify as to the contents of the NPPA Report, Chubb had not produced “any reasonable or credible evidence” for the findings made in the NPPA Report to be rejected (Judgment at [97]).

176 At the outset, it should be clearly stated that even though RIB was the maker of the NPPA Report, a representative from RIB was not called to testify as to its contents. The substance of the NPPA Report was hearsay evidence. Even if Chubb had agreed that the NPPA Report was made (ie, that it was authentic), this was not the same as to say that Chub had agreed to its contents. On this point, I am in agreement with the Majority Judgment (see [126] above). Since the contents of the NPPA Report are objectionable as hearsay evidence, Chubb was entitled to argue that no weight should be given to the substance of the NPPA Report even if technically it was admissible as evidence. The Judge’s observation that “some caution” was warranted when relying on the NPPA Report was not adequate. I also note that the Judge faulted Chubb for not having called the official from NPPA to give evidence or make further inquiry into the extent of NPPA’s investigations (Judgment at [96]–[97]). In my view, this was an error as it was not for Chubb to show that no weight should be given to the NPPA Report but for Sizer to show that weight should be given to the NPPA Report.

177 Further, it was for Sizer to show how extensive the NPPA’s investigations were, which it did not.

178 In his decision, the Judge held that it was “clearly logistically impossible for the thefts to have taken place at Excellent Mining’s premises” given the steps involved in swapping the contents of the drums, unless Excellent Mining’s staff were complicit in the thefts (Judgment at [68]–[69]). Given that parties were agreed as to the method in which the thefts had occurred, I am in agreement with the Judge’s finding that the thefts would have entailed significant logistics, which in turn would have resulted in much work and noise to swap the tin concentrate in the drums with iron oxide.

179 However, I am of the view that the Judge erred in rejecting the evidence of Chubb’s expert, Dr Petrone, who had suggested complicity on the part of Excellent Mining’s staff, solely on the basis that there was no direct evidence of it (Judgment at [65]). On the other hand, the Judge was prepared to consider that there might have been complicity by each container’s driver and his assistant solely on the basis of Mr Sim’s evidence that “no security was provided” to the drivers (Judgment at [79]). In my view, there was no basis for this distinction.

180 Ultimately, the Judge erred because he had approached the issue from the perspective as to which aspect of the process or journey was weakest in terms of security. However, this was not a case of weak security but rather one of complicity. The Judge had implicitly recognised this when he suggested that the drivers and assistants might have been complicit but did not follow through on this vis-à-vis complicity on the part of Excellent Mining’s staff.

181 In my view, the issue at the heart of this case is more likely than not an issue of complicity for the simple reason that the thieves did not simply take the cargo but replaced it so as to avoid detection for as long a time as possible. This was not a snatch and grab job. The thieves had taken the effort to replace the contents of the drums to match the exact weight of the tin concentrate. They had painstakingly re-welded each drum and re-painted each drum lid without damaging the Precintia clips spot welded on the sealed bungholes of the drums. They did their job well such that no tampering of the drums was noticed at any stage, ie, before they left and after they left the premises of Excellent Mining. As the Judge himself put it “the swaps were not done by amateur thieves but by a professional and well-organised gang of thieves” (Judgment at [76]).

182 The next question is why the thieves went through all that trouble. The simple answer is that it was to avoid a quick detection. This was not just to facilitate a quick get-away but so that the theft could be repeated another three times. This in turn suggests that the thieves had inside knowledge that there would be other shipments of tin concentrate to the same buyer, using the same purported route (regardless of how and where the thefts actually occurred), and that by the time the first theft was discovered, it would have been too late for the other thefts to be prevented. The Judge had not considered any of this.

183 On the question of the white alkyd paint on the drum lids, Chubb has chosen not to raise this as part of its argument on appeal. This is an unusual concession for the reasons stated below.

184 To recapitulate, the argument in the court below was that the same white alkyd paint available at Excellent Mining’s premises was used to conceal the weld marks after the thefts had occurred. In essence, this argument consisted of two prongs. The first was Dr Petrone’s opinion that the photographs of the drums as they were transported out of Excellent Mining’s premises showed a similar layer of thick white paint at the rim of the drums as the photographs taken when the drums arrived in Penang. The second was a chemical analysis of the white paint on the lid of the drums (ie, the original paint on the drums) and paint flakes taken from the circumference of the drums’ lids (ie, where the paint was applied to conceal the weld marks) performed by Socotec, which purportedly showed a strong correlation between the two paint samples.

185  I turn to the Judge’s findings regarding the chemical analysis of the paint. While the Judge agreed with the Socotec Report that the two samples showed a “very strong correlation” he held that because there was no complete match the Socotec Report was ultimately inconclusive (Judgment at [116]). Again, the Judge appears to have missed the primary contention raised by Chubb, which was that the thieves had used paint of the same colour and similar composition as the original paint, which strongly suggests some level of complicity. In other words, the thieves knew what paint to use which suggested a vested interest in ensuring that the thefts were discovered as late in the day as possible. Thus, while I do not disagree with the Judge’s holding that the analysis of the paint samples did not necessarily show that the thefts had occurred at Excellent Mining’s premises (Judgment at [131]), the Judge missed the point made by Chubb, which was that the evidence concerning the white alkyd paint used strongly suggested inside knowledge and therefore complicity.

186 However, I will not place any weight on the point since Chubb is not relying on the point for its appeal.

187 Similarly, Chubb does not rely on the presence of 3Ts minerals in the drums for this appeal. Chubb contended at trial that as the 3Ts minerals were found in the drums mixed with the iron oxide this suggested that the swap must have occurred in Rwanda (ie, at Excellent Mining’s premises). However, the Judge held that this was “not conclusive but speculative” (Judgment at [109]). In this regard, I agree with the Judge that the evidence of the 3Ts minerals was not conclusive. However, it was not entirely irrelevant as it suggests that the thefts might have occurred in Excellent Mining’s premises. Furthermore, in rejecting Chubb’s argument, the Judge had come up with his own explanation as to how the minerals came to be mixed with the iron oxide, an explanation that was not supported by any evidence whatsoever (Judgment at [110]). That said, I will not place any weight on the presence of the 3Ts minerals since Chubb did not appeal the Judge’s findings on it.

188 In any event, the evidence concerning the white alkyd paint and the 3Ts minerals merely reinforced the likelihood of complicity. Without them, complicity was still likely and the thefts might have occurred before each container truck left Excellent Mining’s premises for the Bonded Warehouse.

189 It should be noted that if the thefts did take place after the tin concentrate had left Excellent Mining’s premises, it would mean that the thieves would have had to undertake the additional task of removing the seals on the container doors without breaking them or removing the doors themselves in order to remove the drums and swap the contents of the drums. There was no good reason for the thieves to undertake this additional task if there was indeed complicity and this in turn suggests that the thefts might have occurred before the drums left Excellent Mining’s premises.

190 The Majority Judgment notes that Chubb did not plead complicity. In this regard, the majority observes that Chubb did not run its case on the basis that losses were brought about by Sizer’s wilful misconduct which is an express exclusion in cl 4.1 of the ICC(A). Also, the Judge noted that Dr Petrone had belatedly sought to argue at trial that Excellent Mining was complicit in the thefts and that this was an afterthought which undermined Dr Petrone’s credibility (see [97] above).

191 I address the pleading point first. Chubb was not relying on Sizer’s complicity but on the complicity of Excellent Mining. Hence Chubb was not relying on cl 4.1. It was also not necessary for Chubb to plead complicity as a defence so long as it did dispute that the thefts had occurred during the Transit Period. It was not necessary for Chubb to plead how the thefts had occurred as the legal burden of proof was on Sizer.

192 As regards the point that complicity was raised late in the day by Dr Petrone, the lateness was only one factor to be taken into account. The more important point was whether complicity was a logical inference to be drawn based on the facts and this did not depend on Dr Petrone’s credibility. As mentioned above, the Judge himself thought that there was complicity, but he confined it to that of the driver and his assistant of each truck. I have already mentioned that there was no reason why it should be so confined. Indeed, Sizer itself did not raise such complicity.

193 The Majority Judgment reasons that it was not necessary for the Judge to comment that the thefts had occurred with the complicity of the driver and his assistant (see [136135] above). However, that is because the majority is of the view that there was prima facie proof that the thefts had occurred during the Transit Period. If that premise is removed, then the question of complicity becomes more important.

194 In so far as logistics is concerned, I agree that the evidence suggests that it would be near impossible for the thefts to have occurred at Excellent Mining’s premises without complicity. In any event, with complicity, the swap could have occurred outside Excellent Mining’s premises but before the cargo eventually left Excellent Mining’s premises for the Bonded Warehouse each time. There was some interval between the time when: (a) the drums were sealed; and (b) the time when the cargo officially left Excellent Mining’s premises, to commit the theft. Indeed, it was not Sizer’s case that the interval was too short (for each shipment) for the theft to take place.

195 I acknowledge that it was Chubb’s case theory that the thefts took place at Excellent Mining’s premises and not outside Excellent Mining’s premises. But again, it is worth repeating that the burden was not on Chubb to prove a positive case if there was no shift in the evidential burden. It was for Sizer to prove that the thefts took place during the Transit Period. Thus, for example, if there was sufficient evidence to establish that the thefts had occurred outside Excellent Mining’s premises but before the Transit Period began, Sizer would have failed in its claim. It would not succeed just because the thefts had occurred outside its premises instead of at its premises as the question is not so much where but when the theft occurred.

196 There are too many unknowns. I do not say that the thefts did occur at Excellent Mining’s premises or outside it before the Transit Period. Rather, that Sizer has not proved that the thefts had occurred during the Transit Period.

197 In summary, if the Judge had properly considered the question of complicity, then the arguments from Sizer about security would be less important. This would also mean that it was equally likely that the thefts had occurred before or after the cargo left Excellent Mining’s premises. Accordingly, Sizer would not have discharged its burden of proof of showing that the thefts had occurred after the insurer’s risk had passed to Chubb, ie, during the Transit Period. For the foregoing reasons, I would have allowed Chubb’s appeal.