FEDERAL COURT OF AUSTRALIA Dalby Bio-Refinery Ltd v Allianz Australia Insurance Limited [2019] FCAFC 85 Appeal from: Dalby Bio-Refinery Ltd v Allianz Australia Insurance Limited [2018] FCA 1806 File number: VID 1596 of 2018 Judges: ALLSOP CJ, BEACH AND ANASTASSIOU JJ Date of judgment: 24 May 2019 Catchwords: INSURANCE – where damage found to stockpiles of dry distillers' grain and solubles – construction of perils exclusion clause in industrial special risks insurance policy – whether "heating" in exclusion clause qualified by the word "spontaneous" – interpretation of the phrase "spontaneous heating" – whether explanation required for cause of spontaneous heating – appeal dismissed with costs Cases cited: Darlington Futures Ltd v Delco Australia Pty Ltd [1986] HCA 82; 161 CLR 500 Electricity Generation Corporation v Woodside Energy Ltd [2014] HCA 7; 251 CLR 640 Gunns Forest Products Ltd v North Insurances Pty Ltd [2006] VSCA 105; 14 ANZ Insurance Cases 61-691 Johnson v American Home Assurance Company [1998] HCA 14; 192 CLR 266 McCann v Switzerland Insurance Australia Ltd [2000] HCA 65; 203 CLR 579 Mount Bruce Mining Pty Ltd v Wright Prospecting Pty Ltd [2015] HCA 37; 256 CLR 104 Selected Seeds Pty Ltd v QBEMM Pty Ltd [2010] HCA 37; 242 CLR 336 Wilkie v Gordian Runoff Ltd [2005] HCA 17; 221 CLR 522 Date of hearing: 14 May 2019 Registry: Victoria Division: General Division National Practice Area: Commercial and Corporations Sub-area: Commercial Contracts, Banking, Finance and Insurance Category: Catchwords Number of paragraphs: 39 Counsel for the Appellant: Mr M D Wyles QC with Mr B J Murphy Solicitor for the Appellant: McCullough Robertson Counsel for the Respondents: Mr G G McArthur QC with Ms K E Foley Solicitor for the Respondents: Lander & Rogers ORDERS VID 1596 of 2018 BETWEEN: DALBY BIO-REFINERY LTD Appellant AND: ALLIANZ AUSTRALIA INSURANCE LIMITED First Respondent CHUBB INSURANCE AUSTRALIA LIMITED Second Respondent ZURICH AUSTRALIAN INSURANCE LIMITED Third Respondent JUDGES: ALLSOP CJ, BEACH AND ANASTASSIOU JJ DATE OF ORDER: 24 May 2019 THE COURT ORDERS THAT: 1. The appeal be dismissed with costs. Note: Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011. REASONS FOR JUDGMENT THE COURT: 1 This is an appeal from orders made by a judge of the Court in the Insurance List dismissing an application by an insured to enforce a claim for material damage under an Industrial Special Risks Policy issued by the respondent insurer. 2 The appellant carries on business as a manufacturer and distributor of ethanol. In carrying out that business it carries on a bio-refinery at Dalby in South East Queensland. 3 In the early hours of 2 March 2016, fire and emergency services attended the premises at Dalby in respect of damage that had occurred and was occurring to stockpiles of dry distillers' grain and solubles in four bays (numbers 1, 2, 3 and 4). The primary judge summarised the matter at [3] of his reasons: At approximately 6.30 am on 2 March 2016, smoke was detected in Bay 2. Queensland Fire and Emergency Services attended the site and conducted tests on small samples which indicated that the Product was unlikely to develop into large scale flaming combustion. When the stockpile of Product in Bay 2 was opened, there was significant discolouration below the surface, while Product near the surface remained pale. The Product had a "burnt smell" and, as a result, the entire stockpile of the Product in Bay 2 was discarded. Subsequent inspection of the Product in Bays 1 and 3 uncovered significant damage and the stockpiles were also discarded. Bay 4 was in a separate part of the storage shed, was only partially filled at the time, and the Product showed no sign of damage. 4 The issue in the case and on appeal was the operation of a perils exclusion in the policy. There was no argument that the policy otherwise would respond to the relevant physical damage. The terms of the perils exclusion were as follows: Perils Exclusions: The Insurer(s) shall not be liable … in respect of:- … 6. physical loss, destruction or damage occasioned by or happening through:- … (c) (i) spontaneous combustion (ii) spontaneous fermentation or heating or any process involving the direct application of heat Provided that Perils Exclusions 6 (c)(i) and 6(c)(ii) shall be limited to the item or items immediately affected and shall not extend to other property damaged as a result of such spontaneous combustion, fermentation or heating or process involving the direct application of heat. 5 The circumstances that gave rise to the damage were referred to a referee for a report. The referee (Rodney Weber B.Sc. (Melb.) and Ph.D. (Tas.)) prepared a report and supplementary report which (subject to questions of the construction of the policy) were agreed by the parties to be adopted. 6 It is convenient to recite [9]–[12] of the primary judge's reasons to understand the essence of the referee's views that were adopted: 9 On 27 July 2018 the referee was asked to address two questions. They are worth setting out in full: 1 Whether, in the opinion of the referee, it is more likely than not, that the damage to the Applicant's product (stockpiles of dry distillers grain and solubles) was occasioned by or happened through: (i) spontaneous combustion of the product; (ii) or spontaneous fermentation; (iii) or heating; (iv) or any process involving the direct application of heat. 2 Whether, in the opinion of the referee, it it is more likely than not, that the damage to the Applicant's product was occasioned by or happened through some other process and, if so, what caused the damage. 10 Following the filing (but prior to the adoption) of the Referee Report, Dalby submitted that the report was not pellucid in that it referred to the term "self-heating". Reference was made to earlier expert reports (not in evidence on the final hearing) to suggest that there was an important scientific distinction between "self-heating" and "spontaneous combustion", and that the Referee Report did not explain what led to the combustion and damage to the Product. It was in these circumstances that an application was made for the referee to answer further questions aimed at clarifying this issue. This occurred by direction being made pursuant to FCR 28.67(1)(b) that the referee explain by further report: (a) whether his conclusions that the damage to the Applicant's product was occasioned by or happened through self-heating means that damage was occasioned by or happened through spontaneous heating; (b) if the answer to question (a) is either yes or no, please explain; and (c) if the answer to question (a) is not necessarily, please explain. 11 During the course of oral submissions today, it was accepted by Mr Horgan QC (who appeared with Mr Murphy on behalf of Dalby) that the referee's opinion, as demonstrated by the Reports, was that it was more likely than not that the damage to the Product was occasioned by or happened through the process of self-heating. This was plainly correct. If one looks at the Referee Report, this conclusion is expressed on five separate occasions with the referee noting: (i) In my opinion the damage to the [Product] was most likely occasioned by or happened through self-heating. (ii) The most probable cause of the damage to the [Product] is self-heating. (iii) The observations that: • product in the first Bay to be filled; namely Bay 2, which was where the smoke and localised flaming was first noticed, had reached a temperature in excess of 65 degrees Celsius, had a burnt smell and was found to be damaged within the stockpile; • product in the next Bay to be filled; namely Bay 1, had also reached a temperature of 65 degrees Celsius and was found to have similar damage within the stockpile; and • the last Bay to [be] filled; namely Bay 3, was also found to have product damage (no temperature measurements given) but deeper within the stockpile; lead me to conclude that the most probable cause of the damage to the [Product] was self-heating. (iv) The most probable cause of the bulk of the damage to the [Product] is self-heating. (v) One cannot be absolutely certain of the most probable cause of damage to the [Product]. Unfortunately, based upon the information and materials available, it is not possible to identify a single cause for the self-heating which probably led to combustion and damage of the [Product]. It may well be that all of the factors identified above acted together to cause the damage, but without conducting large scale tests, it is likely that detailed causal relationships will remain unknown. 12 Further, in the Supplementary Report, the referee noted: Unfortunately, it is not possible to be certain of all the factors that contributed to the product damage in the current case. In particular, the cause of the observed flaming combustion on 2 March 2016 cannot be determined with certainty; it may have been self-heating reaching criticality, but it may have been the result of a more complex mix of factors. In my opinion the most probable cause of damage to the [Product] was self-heating and the observations on the 2nd March 2016 were an alert that product damage had occurred. 7 The circumstances of the loss were set out by the referee in a document called "Background Information Relevant to Report". This is annexed and marked as A. 8 The referee sought to explain why the self-heating had happened. He said that there was more than one mechanism that can cause self-heating. These were not causes of the damage distinct from self-heating, but the mechanisms that brought about the self-heating. They were: ambient temperature and stockpile size, initial temperature and stockpile size, wetting, humidity and spontaneous fermentation. The last factor (spontaneous fermentation) was ruled out because of the insufficiency of yeast within the stockpile. The referee said: The self heating may be the result of the manner in which the stockpiles were constructed, including height of these stockpiles, combined with ambient conditions of high temperature and at times high humidity, and possibly exacerbated by moisture adsorption from a humid atmosphere and water ingress prior to 2nd March 2016. Whether the self heating actually led to the observed combustion in Bay 2, or whether the combustion was locally initiated by a more complex combination of self heating and water ingress cannot be conclusively determined based upon the information available. 9 The question of the effect of wetting was then referred to in the answer to question 2, as set out at [6] above: In my opinion water ingress would have contributed to discolouration of the product through the oxidation reaction referred to above. However, it seems unlikely to have been the sole reason for any temperature increase, so in my opinion, the most probable cause of the bulk of the damage to the Applicant's product (DDGS) is self heating. 10 The second report answers the further questions as follows: In my opinion self-heating, an increase in temperature due to internal exothermic reactions, and spontaneous heating are not quite synonymous terms. Spontaneous suggests an event or a suddenness. Self-heating can be a very slow process and if conditions are such that a stockpile is close to critical, it can take months for the temperature in the stockpile to either reach equilibrium, or to reach criticality. It is for this reason that the process is described as self-heating rather than spontaneous heating. If the stockpile reached criticality, then the related ignition event would be described as spontaneous. Additionally, because the self-heating can be a slow process, this allows for the possibility that other factors could influence the final outcome; particularly if it is a large stockpile of product. Unfortunately, it is not possible to be certain of all the factors that contributed to the product damage in the current case. In particular, the cause of the observed flaming combustion on 2nd March 2016 cannot be determined with certainty; it may have been self-heating reaching criticality, but it may have been the result of a more complex mix of factors. In my opinion the most probable cause of damage to the Applicant's product (DDGS) was self-heating and the observations on the 2nd March 2016 were an alert that product damage had occurred. 11 From these reports it is clear that the referee concluded that the damage was caused by self-heating, though he could not be precise about a cause or mechanism that brought about the self-heating, which may have included wetting from rain. His view about a difference between spontaneous heating and self-heating was explained in his second report as based on his understanding of the English meaning of the word "spontaneous". The approach of the primary judge 12 The primary judge accepted the submission of the insurer that "heating" in exclusion 6(c)(ii) was not qualified by the word "spontaneous". The reasons for this conclusion were set out at [32]–[37]: Conclusions 32 It seems to me it is tolerably plain, from both the text of the exclusion and having regard to the relevant context, that the Insurers were not prepared to accept a risk of damage to the Product occasioned by or happening through heating of any type. I accept the submissions of the Insurers that there are a number of textual matters pointing against the constructions agitated by Dalby. 33 The first and most obvious is that the word heating is not qualified or modified by the word "spontaneous" as is the case with "combustion" or "fermentation". The notion that spontaneity qualifies all of the perils in cl 6(c)(ii) cannot be sustained in circumstances where one of the matters referred to in that clause involves what is clearly an external process (being the application of heat, and is itself qualified by the word "any"). 34 Secondly, there is a difference between damage occasioned by spontaneous combustion and damage that would be occasioned by any form of heating (including self-heating); it follows that "heating" has work to do in the exclusion clause properly construed. Indeed the circumstances of this case make it clear that there could be damage to the Product occasioned by heating which fell short of any combustion and it seems to me that this is precisely the sort of eventuality to which perils exclusion 6(c)(ii) is directed. 35 Thirdly, contrary to the submissions of Dalby, I do not believe that the proviso assists. The proviso refers to "property damage as a result of such spontaneous combustion, fermentation or heating or process involving the direct application of heat". 36 Plainly, given the comma between "combustion" and "fermentation" prior to the disjunctive appearing before the word "heating", "spontaneous" is qualifying "combustion" and "fermentation". What is also evident is that the insertion of the disjunctive before "heating" points to the fact that "heating" is embracing a broader causal concept as is "any other process involving the direct application of heat". For these reasons, I do not consider there to be any ambiguity and the Insurers' submissions as to the proper construction of the perils exclusion are to be preferred. 37 It also follows that the argument of Dalby based on the application of the contra proferentem rule, such as it is, must be rejected. 13 The primary judge then dealt with "spontaneous heating" should he be wrong about the word heating being unqualified by "spontaneous". It is at this point that the referee's second report becomes relevant: whether "self-heating" was a synonymous expression for "spontaneous heating". At [39]–[41] of his reasons the primary judge said: 39 In determining this issue, I was referred by the parties to various dictionary definitions. The Macquarie Concise Dictionary, 5th ed, provides a definition of "spontaneous" and a definition of "spontaneous combustion" in the following terms: spontaneous / adj. 1. Proceeding from a natural personal impulse, without effort or premeditation; natural and unconstrained; a spontaneous action; a spontaneous remark. 2. (of impulses, motion, activity, natural processes, etc.) arising from internal forces or causes, or independent of external agencies. 3. growing naturally or without cultivation, as plants, fruits, etc. 4. produced by natural process. spontaneous combustion / n. the ignition of a substance or body from the rapid oxidation of its own constituents, without heat from any external source. 40 The Oxford Encyclopaedic English Dictionary sets out definitions of the same terms as follows: spontaneous / adj. 1 acting or done or occurring without external cause. 2 voluntary, without external incitement (made a spontaneous offer of his services). 3 Biol. (of structural changes in plants and muscular activity esp. in young animals) instinctive, automatic, prompted by no motive. 4 (of bodily movement, literary style, etc.) gracefully natural and unconstrained. 5 (of sudden movement etc.) involuntary, not due to conscious volition. 6 growing naturally without cultivation. spontaneous combustion the ignition of a mineral or vegetable substance (e.g. a heap of rags soaked with oil, a mass of wet coal) from heat engendered within itself, usu. By rapid oxidation. 41 The Supplementary Report provides the referee's reasoning process for drawing some meaningful distinction between the concept of "self-heating" and "spontaneous heating". The referee expresses his view that they are "not quite synonymous terms". This is because, according to the referee, spontaneous suggests "an event or a suddenness" and self-heating can be a "very slow process" which can take months to reach "criticality". It is for this reason that the referee described the process as self-heating rather than spontaneous heating. Both parties agreed that it was for the Court to determine the content to be given to the word "spontaneous" as used in the Policy to the extent that it was relevant. It seems to me that the ordinary, usual and relevant meaning of the word "spontaneous", is that it describes the occurrence of something without external cause. Although spontaneous combustion is usually related to something that occurs with some rapidity, when talking about spontaneous in the sense of qualifying the causal concept of heating, it simply means that it occurs without external incitement or factors, rather it is something which occurs by reason of internal processes. 14 Thus, the primary judge concluded that there had been shown to be "spontaneous heating", if the word "spontaneous" qualified the word "heating". 15 The third question dealt with by the primary judge was the argument of the appellant encapsulated by the primary judge at [42] of his reasons: The above is not sufficient to determine the case because Dalby argues that the Insurers have failed to discharge their evidentiary and persuasive burden of identifying that heating (or more precisely, self-heating) is the proximate cause of the damage. Reliance is placed by Dalby on the cautions, or more specifically, the "caveat" expressed by the referee. In the Referee Report, under the heading "Overall Caveat", the following appears: Without detailed records of the manner in which Bays 2, 1 and 3 were stacked with product (DDGS); including the quantities delivered, the dates of delivery, the temperature of each delivery and the ambient temperature and relative humidity at each delivery, and without further information of the physical properties of the product (DDGS), one cannot be absolutely certain of the most probable cause of damage to the Applicant's product. Unfortunately, based upon the information and materials available, it is not possible to identify a single cause for the self-heating which probably led to combustion and damage of the Applicant's product. It may well be that all of the factors identified above acted together to cause the damage, but without conducting large scale tests, it is likely that detailed causal relationships will remain unknown. 16 At [44]–[48] the primary judge examined the reasons of the referee as follows: 44 What the referee was relevantly reporting upon was his opinion as to what was the likely cause of damage to the Product. As both parties accepted on the application for adoption of the Referee's Report, the referee properly attended to his task and his process of reasoning set out in the Referee's Report is relatively clear. He starts by identifying that the damage to the Product "was most likely occasioned by or happened through self-heating". He then proceeds to note that "as there is more than one mechanism that can cause self-heating, each of these mechanisms need to be considered in turn". The reference to "cause" in this context needs to be examined more closely. Although the referee was expressing a view that damage was caused through self-heating, the referee was also directing himself to the anterior issue as to the mechanism which itself caused the self-heating. 45 The referee then considers various "mechanisms" which could cause self-heating, being: ambient temperature and stockpile size, initial temperature and stockpile size, wetting, and spontaneous fermentation. Dealing with each of these in turn, the referee notes that the ambient temperature and stockpile size could only have caused self-heating leading to spontaneous combustion if the stockpile size was significantly larger than the stockpile size that had been referred to in the reports provided to him. As to initial temperature and stockpile size, the referee considered it was difficult to estimate its contribution as there was, apparently, insufficient material available to make an estimate of the critical stockpiling sizes or stacking temperatures at the relevant times. 46 As to wetting, while there was evidence that ingress of water caused localised wetting, the referee considered the amount of water appeared to be quite small in comparison to the overall size of the stockpile. This factor, and other indications, suggest that wetting would have had little effect on heating, although it could not be conclusively ruled out. The conclusion was formed that it seemed unlikely to be a singular cause of the self-heating. 47 Finally, the referee ruled out spontaneous fermentation as all investigators had concluded that there were insufficient quantities or sources of yeast within the stockpile. 48 In this way, what can be observed is that the referee was carefully going through the various factors which either individually or together might have been the basis for self-heating to occur. The way in which Dalby puts it, is that it could not sensibly be said that self-heating was the real and proximate cause of the damage, but rather there needs to be recognition of the fact that the referee could not identify what, in fact, caused the loss and damage to the Product. 17 After referring to Gunns Forest Products Ltd v North Insurances Pty Ltd [2006] VSCA 105; 14 ANZ Insurance Cases 61-691, the primary judge concluded that the referee's inability to explain why the self-heating took place did not undermine his conclusion that, whatever the causative mechanism, the appellant was seeking to make the insurer liable for damage occasioned by or happening through self-heating and so (spontaneous) heating, within perils exclusion 6(c)(ii) of the policy. The arguments on appeal and their resolution 18 There was no disagreement between the parties as to the principles of construction of insurance policies. The proper approach to construction of insurance policies, and in particular the resolution of the meaning of exclusion clauses, as a species of commercial contract is to be found in decisions of the High Court: Darlington Futures Ltd v Delco Australia Pty Ltd [1986] HCA 82; 161 CLR 500 at 510; Johnson v American Home Assurance Company [1998] HCA 14; 192 CLR 266 at 272–276 [19] (Kirby J, albeit in a dissenting judgment); McCann v Switzerland Insurance Australia Ltd [2000] HCA 65; 203 CLR 579 at 589 [22]; Wilkie v Gordian Runoff Ltd [2005] HCA 17; 221 CLR 522 at 528–529 [15]–[16]; Selected Seeds Pty Ltd v QBEMM Pty Ltd [2010] HCA 37; 242 CLR 336 at 344 [29]; Mount Bruce Mining Pty Ltd v Wright Prospecting Pty Ltd [2015] HCA 37; 256 CLR 104 at 116–117 [46]–[52]; and Electricity Generation Corporation v Woodside Energy Ltd [2014] HCA 7; 251 CLR 640 at 656–657 [35]. 19 Complaint was made by the appellant that the primary judge had failed to consider the whole of the policy in coming to his preferred construction of the exclusion. In support of that submission, the appellant referred to the structure and various parts of the policy. It is unnecessary to deal with every refinement of this submission. 20 The policy had clearly been negotiated by important changes to the operation of the cover, the property exclusions and the perils exclusions. None of the terms of the insuring clause, the level of the deductibles, or the property insured (subject to property exclusions) or the basis of settlement clause assists in ascribing meaning to the word "heating" in perils exclusion 6(c)(ii). 21 It was submitted that the changes to property exclusion 16 dealing with machinery breakdown, to property exclusion 17 dealing with boiler and pressure vessel explosion, the extension of cover for computer and electronic equipment breakdown and the extension of cover for fusion assisted in understanding that the word "spontaneous" qualified the word "heating" in perils exclusion 6(c)(ii). None does so. Property exclusion 16 was replaced in the endorsement by a completely new regime for indemnification and exclusion (property and perils) for "sudden and unforeseen physical loss of or damage to Machinery [as defined] occasioned by or happening through any mechanical, electrical, electromechanical, electronic or hydraulic malfunction, failure, derangement, breakdown or non-operation of any kind except as otherwise specifically excluded". A regime of property and perils exclusions that took precedence over the exclusions applicable generally was then set out. It was submitted that the indemnity in this new regime covered heating. So much can be accepted, but the operation of this clearly separated and coherently drafted regime of coverage for machinery breakdown does not assist the understanding of perils exclusion 6(c)(ii). 22 The computer and electronic equipment breakdown extension of cover in the endorsement excluded perils exclusion 6(c). It was submitted that this showed that "heating" in perils exclusion 6(c)(ii) was "spontaneous heating". That does not follow at all, other than by bare assertion. 23 The new regime for boiler and pressure vessel explosion by the replacement of property exclusion 17 and the extension of cover for fusion in electrical equipment by the endorsement likewise do not assist. 24 The primary judge displayed no error in not lengthening his reasons for judgment by referring to parts of the policy of no assistance. 25 The proper meaning to be given to perils exclusion 6(c)(ii) is a businesslike construction, by reference to what a reasonable business person would have understood the words in their commercial context to mean. 26 Here, the policy is carefully drawn and proper regard should be made to the language used, including usual grammatical and syntactical usage, without overly dwelling on choices of structure, syntax or punctuation that are open to different nuances. 27 The reasons of the primary judge for the lack of qualification to the word "heating" by the word "spontaneous" set out at [12] above have force. They are supported, as was submitted by the insurer, by the repetition elsewhere (such as in perils exclusion 4(e) of a relevant modifier: "faulty materials or faulty workmanship"). 28 If anything, the policy is marked by a degree of precision and economy of language. For instance, in perils exclusions 1–7 the phrase "physical loss, destruction of or damage to the Property" does not require the repetition of "physical" before "destruction" on "damage" for the word "physical" to pervade the expression. 29 The policy is to be understood in a context where the insured stored large quantities of grain. Perils exclusion 6(c) can be seen to be removing certain matters from the risks that the insurer was willing to accept and that were to be placed at the risk of the insured. Perils exclusion 6(c)(i), "spontaneous combustion", is a phrase encompassing actual "combustion". Its spontaneous character may manifest itself in a sudden event of fire or ignition, but spontaneous does not mean sudden; rather, as the dictionaries show, it means self-generated or from within. That is, the thing itself, by reason of its particular environmental or internal conditions will oxidise and heat and through a feedback mechanism ultimately raise the temperature leading to the point of ignition or fire, which a layman would consider to be combustion; a more technical viewpoint, though, may consider combustion to be merely a synonym of oxidation. The first part of perils exclusion 6(c)(ii), "spontaneous fermentation", occurs in the process of sugar being converted to alcohol in the presence of and by yeast. Again, this is a self-generated process. The second part of perils exclusion 6(c)(ii), "heating", is explicable in the context where perils exclusion 6(c)(i),"spontaneous combustion", as understood by a layman, is only the endpoint of and may not capture the self-generated heating prior to that endpoint , which self-generated heating may (as here) damage property such as grain, hay and the like. Accordingly the context makes it apparent that "heating" is to be qualified by "spontaneous" to read "spontaneous heating". And that is an open grammatical construction, with the word "spontaneous" in the phrase "spontaneous fermentation and heating" being applied distributively. Now up to this point these perils exclusions are dealing with internally generated processes. Contrastingly, the balance of (ii): "or any process involving the direct application of heat" involves the external application of heat. Indeed, if "heating" were not qualified by the idea of self-generation in the meaning of "spontaneous" there would be little need for the balance of (ii). 30 Also the structure of the proviso – "such spontaneous combustion, fermentation or heating" –tends to confirm that spontaneous in the sense of self-generated heating was intended to qualify heating. 31 On this view, the clause would cover self-generated combustion, fermentation and heating as well as the external application of heat. 32 This difference of view with the primary judge is as much, if not more, a conclusion drawn from impression of meaning as it is from any logical reasoning. Further, though one needs to be careful with reliance on the contra proferentem rule, especially when there has been an evident degree of negotiation of the policy, if there are two genuinely available alternatives preference should be given to one that limits rather than expands the exclusion. That is not to approach the matter other than as dictated by the Court in Darlington v Delco 161 CLR at 510. 33 Thus, we respectfully disagree with the primary judge on the qualification of the word "heating" with the adjective "spontaneous". 34 We agree, however, with the alternative reasoning of the primary judge that the referee's report and the ordinary and proper meaning of the phrase "spontaneous heating" that should be given objectively as that which would commend itself to business people lead to the conclusion that the self-heating, caused by whichever mechanism or combination of mechanisms, was spontaneous heating. 35 The appellant accepted that the meaning of "spontaneous" included the notion of self-generation or arising from internal forces or causes. It urged, however, a meaning that was more colloquial that included the notion of an event or some suddenness. We do not see that as being a necessary part of the meaning of spontaneous. It may be within the meaning of the composite phrase "spontaneous combustion" (if one conceives of combustion as immediate or sudden). It may, in some contexts (such as a spontaneous human gesture), sometimes carry a sense of immediacy. But its meaning is broader, as the dictionaries show. Further, there is no reason why "spontaneous fermentation" would be understood as in any way sudden. Fermentation is a gradual process. Thus, we agree with how, in [41] of his reasons, the primary judge dealt with the referee's views that "self-heating" and "spontaneous heating" were "not quite synonymous". There is no basis to conclude that the referee's second report was other than his understanding of the word "spontaneous". It was not put by him as a term of scientific art or as a matter of science. The primary judge was entitled to approach the meaning of the word as a matter for him. He was correct in his analysis and conclusion. 36 The appellant also submitted that there was required to be proved by the insurer an absence of an independent cause, which it was said had not occurred, because of the doubts of the referee as to the cause of the self-heating and the possibility of wetting (as an external cause, independent of self-heating, it was submitted). 37 Once again we agree with the way the primary judge dealt with the issue. The perils exclusion dealt with damage occasioned by or happening through (a process of) spontaneous heating. Spontaneous heating, that is heating generated internally (as opposed to externally), involves environmental factors providing oxygen, whether through air or water, for a process of oxidation and heating. The circumstances that provided the suitable conditions for the self-generated heating through oxidation explain why the process of spontaneous heating occurred. An inability to be precise or definite about why the process occurred is not doubt that the process occurred. 38 The sense of this approach can be seen if one appreciates that the circumstances and control of the environment which may lead to spontaneous heating or combustion or fermentation are likely to be, in significant part, in the control of the insured: the exposure of the property likely to be subject to such processes to conducive environments of moisture, heat and oxygen. In these circumstances, the enquiry under the perils exclusion ceases at the finding of the answer to the question whether the damage was occasioned or happened through (the process of) spontaneous heating, and does not go further to seeking to explain why the (process of) spontaneous heating came about. 39 For these reasons the appeal should be dismissed with costs. I certify that the preceding thirty-nine (39) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Chief Justice Allsop, and Justices Beach and Anastassiou. Associate: Dated: 24 May 2019 Annexure A