1  This is Mr Leong Sow Hon’s appeal against his sentence of six months’ imprisonment after having pleaded guilty to an offence under s 18(1) punishable under s 18(3) of the Building Control Act (Cap 29, 1999 Rev Ed) (“BC Act”), for failing to evaluate, analyse, and review the structural design in respect of, and perform independent calculations for, a number of key structural elements for a viaduct from the Tampines Expressway to the Pan-Island Expressway (Westbound) and Upper Changi Road East (the “viaduct”). ¹ One further charge under s 43A(a) punishable under s 43A of the BC Act for falsely certifying that he had evaluated, analysed, and reviewed the structural plans in relation to the viaduct construction was taken into consideration for the purpose of sentencing. ² The Prosecution had initially filed a cross-appeal, but that has since been withdrawn.

2 Mr Leong (the “appellant”) was the accredited checker for the construction of the viaduct. As outlined in the statement of facts, which the appellant admitted to without qualification, he was appointed in June 2016 pursuant to s 8 of the BC Act as an accredited checker by the Land Transport Authority (the “Developer”) for the building works related to the construction and completion of the viaduct. ³ At the material time, the other parties involved in the construction of the viaduct were one Robert Arianto Tjandra, the qualified person (“QP”) appointed under s 11 of the BC Act to design the building works of the viaduct and the QP appointed under s 8 of the Act to supervise the building works of the viaduct, and Or Kim Peow Contractors (Private) Limited, the builder of the viaduct.

3 The viaduct itself is, as described in the statement of facts, approximately 1.8 kilometres long. It consists of eight flyovers. Each flyover is a monolithic structure supported by five to nine columns which are integrated with the flyover structure. Each flyover has expansion joints at each end.

4 Each end of a flyover rests on a permanent corbel, a short reinforced concrete projection from the crosshead at an expansion joint. Part of the vehicular load on a flyover would have been transferred to an independent column through the permanent corbel. The permanent corbels are a key structural element of the viaduct as they are essential for the support and overall structural stability of the viaduct.

5 The appellant accepted that under s 18(1) of the BC Act read with paragraph 7(1) of the Building Control (Accredited Checkers and Accredited Checking Organisations) Regulations (Cap 29, Rg 2, 2002 Rev Ed) (the “Regulations”), he was under a duty as the accredited checker to evaluate, analyse, and review the structural design in the plans of any building works and perform such original calculations with a view to determining the adequacy of the key structural elements of the building to be erected or affected by the building works carried out in accordance with those plans. Section 2(1) of the BC Act defines “key structural elements” to mean the foundations, columns, beams, shear cores, structural walls, struts, ground anchors, and such other parts of a building which are essential for its support and overall structural stability. It is admitted by the appellant that the permanent corbels were a key structural element for the purposes of the abovementioned duty.

6 On 14 July 2017, the crossheads at two piers of the viaduct, which were temporary structures, gave way. As a result, the precast girders and formwork supporting the casting of the concrete deck slab of the viaduct collapsed, tragically causing the death of one worker involved in the building works at the material time and injuring ten others with varying degrees of injury. Following that collapse, which did not implicate the appellant, the calculations for the whole structure were checked and several of the permanent corbels were found to have been inadequately designed (see [8] below in particular).

7 It was not contested that in the design calculations submitted to the Building and Construction Authority (“BCA”) for the relevant plans of the viaduct works, there were no independent calculations for the permanent corbels of the viaduct prepared or submitted by the appellant. Further, the appellant, as the accredited checker, acknowledged that he had failed to evaluate, analyse, and review the structural design in the relevant plans and perform original calculations for the permanent corbels of the viaduct, as was his statutory duty under the BC Act.

8 On 26 July 2017, in the midst of the BCA’s investigations, the appellant initially claimed that he had performed original calculations, checked the adequacy of the permanent corbels, and found them to be adequate. The appellant was then asked to provide evidence of such original calculations, but was unable to do so. He admitted on 21 September 2017 to having been untruthful in his initial claim, and conceded that no calculations had been done on his part at all. Subsequently, investigations revealed that corbels at eight out of the 10 piers with permanent corbels were inadequately designed, with five piers being unable to support their intended weight during the construction stage. These five permanent corbels would have collapsed during the casting of the slab at the construction stage. As for the remaining three piers, while they may have supported the requisite weight during the construction stage, the corbels would have showed significant structural cracks upon the viaduct being opened to a full traffic load, leading to sudden brittle failure and in turn leading to collapse.

9 Apart from the consequences already outlined at [6] and [8] above, the collapse of the viaduct caused the estimated completion date for the construction of the viaduct to be delayed by at least two years. For the avoidance of doubt, however, I reiterate that the collapse of the viaduct on 14 July 2017 was not caused by any of the permanent corbels which formed the subject matter of the proceeded charge against the appellant. Rather, the appellant’s failure to properly evaluate, analyse, and review the structural designs and plans for the permanent corbels was discovered before any actual harm was caused.

10 The appellant was thereafter charged. The statutorily-prescribed penalty for the offence under s 18(1) of the BC Act is a fine not exceeding S$100,000, or imprisonment not exceeding two years, or both. The Prosecution, emphasising that general deterrence ought to be the dominant sentencing principle where the offence in question affects public safety, argued that the appellant ought to be sentenced to at least nine months’ imprisonment. ⁴ A sentencing framework was also put forward, which the District Judge (“DJ”) accepted. ⁵ By contrast, the Defence argued that the accused, being a first offender who acted only with inadvertence and whose acts did not directly cause any actual harm, ought to be sentenced only to a fine of S$25,000. ⁶

11 The DJ accepted that the custodial threshold had been crossed. ⁷ She observed that offences under s 18(3) of the BC Act involved considerations of public safety, and also considered the fact that the penalties for offences under s 18 of the BC Act had been increased in 2008. In addition, the DJ made reference to extracts from the Parliamentary Debates concerning building control legislation which underscored the critical role played by the accredited checker in checking a building’s plans and structural integrity.

12 Bearing the abovementioned considerations in mind, the DJ accepted the sentencing framework proposed by the Prosecution. ⁸ She found that the potential harm arising from the appellant’s offence was high, and that the appellant’s culpability was medium. Accordingly, and after weighing the relevant aggravating and mitigating factors, she sentenced the appellant to six months’ imprisonment.

13 On appeal, the appellant argued that the sentence imposed was manifestly excessive, and that the DJ had erred because she had, inter alia:

(a) failed to recognise that the appellant, as an accredited checker, was entitled to rely on other professionals in his accredited checking organisation;

(b) failed to sufficiently appreciate that the checking system prescribed by statute was based on the collective roles of the accredited checker, the QP, the site supervisors, and the builder, and that the failures by the other parties had a significant impact on the appellant’s failure;

(c) failed to recognise that the risk of any potential harm eventuating from the construction of the viaduct would have been re-assessed by the appellant on a continual basis prior to the completion of the works, and would have been detected either by the appellant or one of the other professionals during the building process; and

(d) did not give adequate weight to the applicable mitigating factors. ⁹

The appellant therefore submitted that the appropriate sentence ought to be a fine, or a short detention order (“SDO”).

14 The Prosecution, relying primarily on the legislative history of the BC Act and the central role of the accredited checker in the process, sought to defend the sentence imposed by the DJ. ¹⁰ Specifically, the Prosecution emphasised that the relevant legislative history disclosed that general deterrence was the primary sentencing consideration for offences under s 18(3) of the BC Act, that Parliament had intended for general deterrence to take the form of custodial sentences in the appropriate cases, and that the high harm and medium culpability disclosed on the facts warranted a custodial sentence. It was also asserted that the mitigating factors alleged by the Defence had been given due consideration.

15 Prior to the hearing of the appeal proper, the appellant filed Criminal Motion No 48 of 2019 seeking to admit a report prepared by one Andrew Theodorus van der Meer (the “Report”) as additional evidence in support of his arguments on the appropriate sentence. ¹¹ The broad thrust of the Report is that the defects in the plans approved by the appellant would have been detected prior to any ultimate failure which might have given rise to the collapse of the viaduct arising. It was argued in the Report that since corbels would have behaved in a ductile manner, meaning that they would have exhibited significant cracking and rotation before ultimate failure, it would be reasonable to expect that any cracking to any corbel could have been identified by the supervision team on-site. ¹² Indeed, the builder, Or Kim Peow Contractors (Private) Limited, had been instructed to carry out rectification works to a completed permanent corbel on 4 July 2017.

16 As for the applicable law governing the adducing of fresh evidence on appeal, both the Prosecution and the Defence were largely ad idem. Section 392(1) of the Criminal Procedure Code (Cap 68, 2012 Rev Ed) (“CPC”) provides that an appellate court may, if it deems such additional evidence to be “necessary”, either take such evidence itself or direct that it be taken by the trial court. Fresh evidence sought to be introduced at the appellate stage should satisfy the longstanding Ladd v Marshall conditions, per Iskandar Bin Rahmat v Public Prosecutor and other matters [2017] 1 SLR 505 (“Iskandar”) at [72]. These requirements are that the additional evidence (a) could not have been obtained with reasonable diligence for use at trial; (b) would have had an important influence, though not necessarily a decisive one; and (c) must be apparently credible, even if not incontrovertible.

17 In considering applications made by offenders in criminal proceedings, the Courts have given less weight to the non-availability requirement, and correspondingly more weight to the relevance and credibility of the further evidence to be adduced: Mohammad Zam bin Abdul Rashid v Public Prosecutor [2007] 2 SLR(R) 410 at [6] and Public Prosecutor v Mohd Ariffan bin Mohd Hassan [2018] 1 SLR 544 (“Ariffan”) from [56] to [60]. While I am mindful of the reasons stated from [57] to [60] of Ariffan why the strict Ladd v Marshall requirements should be attenuated in the abovementioned context, this should not be construed as giving carte blanche for the raising of all manner of evidence only on appeal. Similarly, the non-availability requirement ought not to be altogether disregarded. Where the decision not to adduce evidence previously available, or available with reasonable diligence, “call[s] into question the genuineness” of the defence, the Court retains the discretion to reject such applications: Iskandar at [67]. This is particularly so if there appears to have been “drip-feeding” of arguments and evidence by the applicant.

18 On the facts, however, I allowed the appellant’s motion to adduce further evidence. The Report clearly met the requirements of relevance and credibility. The Report was relevant in that it went towards whether the potential harm considered by the DJ in reaching her conclusion on sentence was reasonable, and thus how much potential harm the appellant’s wrong engendered. Given the centrality of considerations of actual and potential harm to sentencing, I accepted that the Report was relevant, even if not necessarily decisive. As for the question of credibility, the Prosecution did not challenge the credibility of the Report, and I had no reason to doubt Mr van der Meer’s credibility and expertise as the author of the Report. ¹³

19 Applying the Report to the facts, and as alluded to above at [15], the appellant sought to persuade me that the DJ had erred in, among other things, determining that the potential harm engendered by the appellant’s acts was “high”. Accordingly, as argued by the appellant, the sentence imposed was manifestly excessive.

20 The Report and arguments by the appellant notwithstanding, I was not satisfied that it could be said that the sentence imposed by the DJ was manifestly excessive.

21 I begin by examining the relevant sentencing considerations which apply to offences such as those in the present case.

22 The question in this regard is whether the accredited checker can point to the responsibility borne by others, and whether his responsibility was shared with them, as argued by the appellant, or whether he had a personal and non-delegable duty.

The appellant’s counsel argued that the appellant could rely on his team of engineers to carry out his duties under the BC Act. Counsel pointed to the provisions concerning the accredited checking organisation under the BC Act and argued that those provisions contemplate reliance being placed by the accredited checker on a team of qualified engineers to discharge his duties. The appellant also relied on there being a work procedure and system. In particular, the appellant highlighted that the project involved the evaluation, analysis, and review of 284 components of key structural elements and the relevant design calculations. It was asserted that the appellant could not have undertaken that work alone given the scale and magnitude of the project. Consequently, the sentence imposed should take into account the context and practicalities of the project. The appellant’s action was not rash, contrary to what had been found by the DJ, given that he was entitled to rely on his team. Collective duties were owed by the appellant together with the accredited checking organisation, the QP, the site supervisors and the builder.

23 The respondent pointed to the legislative history behind the introduction of the accredited checker. That history, it contended, showed that the purpose of the accredited checker was to provide an independent and final check on designs. The accredited checking organisation framework was introduced to deal with accredited checkers dealing with large projects so that there was support, but the duties imposed on the accredited checker remained non-delegable and personal, as was emphasised in the relevant Parliamentary Debates.

24 A number of statutory provisions provide for the responsibility of the accredited checker, which are reproduced (in part) below:

(a) Section 18 of the BC Act provides that:

(b) Regulation 7 of the Regulations provides that:

(c) The First Schedule to the Regulations also prescribes certain forms for certificates to be issued by accredited checkers such as the appellant, and Form A, which is the relevant form for the instant facts, is in itself instructive, as extracted below:

(d) Finally, the Second Schedule to the Regulations set out a list of tasks which must be carried out by accredited checkers:

25 The upshot of the legislative provisions extracted above is clear:

(a) First, the duty of the accredited checker includes not merely evaluating, analysing, and reviewing the structural design in the plans for building works, but also extends to performing “original calculations with a view to determining the adequacy of the key structural elements”. This is evidenced by the conjunctive “and” joining the two clauses in Regulation 7(1)(a) of the Regulations.

(b) Second, specific duties are imposed on the accredited checker, and these duties are separate and distinct from those imposed on the QP. This is clear from the fact that s 9 of the BC Act outlines particular duties imposed on QPs, s 10 provides for duties on site supervisors, and s 11 does the same for builders. Accredited checkers are specifically provided for in s 18 of the BC Act, and the requirement in s 18(1) that the accredited checker “shall” check the detailed structural plans and design calculations of the building works in accordance with the building regulations and “shall” carry out such other duties as may be prescribed by those regulations makes clear the personal, non-negotiable nature of those duties.

26 The Parliamentary Debates provide a useful backdrop against which the broad propositions outlined at [25] above may be understood. In setting out the relevant extracts of the Parliamentary Debates, it is apropos to begin with the genesis of the BC Act, and the Parliamentary Debates surrounding the Building Control Bill (Bill No. 3 of 1988) (the “Building Control Bill”) in the aftermath of the collapse of Hotel New World in March 1986. A Commission of Inquiry chaired by LP Thean J prepared a report on the collapse dated 16 February 1987, and a working group was established to consider and implement the report. ¹⁴ These recommendations, with the relevant modifications, eventually coalesced in the Building Control Bill.

27 During the Second Reading speech for the Building Control Bill on 16 February 1989, Minister for National Development S Dhanabalan stated, in relation to the Building Control Bill, that: ¹⁵

What is clear from the Minister’s speech is the intention that “all” structural plans and calculations are subjected to a series of “independent” checks. All parties are exhorted to play their role. Accordingly, I find that it is simply not envisaged that the accredited checker can abdicate his duties to the other parties involved.

28 On 25 May 1995, Minister for National Development Lim Hng Kiang observed in relation to the Building Control (Amendment) Bill (Bill No. 10 of 1995) that: ¹⁶

This extract highlights that the accredited checker is supposed to provide an “independent” check of not merely high-level plans, but also specific design calculations. Further, the accredited checker is clearly envisaged as potentially being a “final stage” check on plans, underscoring the importance placed on the role.

29 Following the partial collapse of the roof of Compassvale Primary School on in June 1999, which was found to have involved negligence on the part of both the QP and the accredited checker, the BC Act was further amended in 2000. ¹⁷ Those amendments stipulated that an accredited checker undertaking large projects had to be registered and operate within an accredited checking organisation. This was so as to provide the accredited checker with more specialised technical support staff and resources. However, it was nonetheless emphasised in Parliament that the accredited checker’s duty remained personal to him. On 25 August 2000, Minister for National Development Mah Bow Tan observed, in relation to the Building Control (Amendment) Bill (Bill No. 21 of 2000), that: ¹⁸

This extract not only indicates that all structural plans and calculations have to be checked by an independent accredited checker, but that the accredited checker is personally responsible and cannot delegate his responsibility to his assistants.

30 Crucially, in addition, Minister Mah Bow Tan observed that the QP cannot be “lulled into a false sense of complacency that the accredited checker will be there to spot his mistakes”. This logic applies a fortiori to the accredited checker given that the accredited checker is the proverbial “final gate” in the entire system of checks. The intention on the part of Parliament that one part of the system cannot rely on the vigilance of another part to absolve itself of its own responsibility is clear.

31 In sum, the Ministerial speeches in the Parliamentary Debates demonstrate that the framework of the accredited checker regime was to impose personal, non-delegable, responsibility on the person acting as the accredited checker. The provisions governing the establishment of the accredited checking organisation, or the duties on other parties such as the QP and builder, do not remove or even attenuate that personal responsibility.

32 An accredited checking organisation may be established, and it may assist the work of an accredited checker, but the statutory provisions governing the accredited checking organisation do not at all in any way shift the responsibility that is placed on the accredited checker. Instead, the statutory provisions only deal with the establishment, registration, and independence of the accredited checking organisation, and little else; one would have expected express language to distribute or move responsibility away from the accredited checker if Parliament had intended otherwise. The arguments from the appellant’s counsel in this regard thus had to be rejected. Those arguments go up against the plain language of the text imposing personal and non-delegable responsibility on the accredited checker. Further, the express statements from the Parliamentary Debates extracted above clearly militate away from the position advanced by the appellant.

33 In addition, the fact that others may also be responsible for preventing or remedying errors is immaterial in this context of criminal responsibility. A cascade of errors, each seemingly remediable by itself, can lead to disaster. This was amply borne out by the Nicoll Highway collapse on 20 April 2004. The Committee of Inquiry found, in its report, that the collapse began with two critical design errors. ¹⁹ The warning signs that arose from those errors at an early stage were not heeded, and the presence of multiple parties – the builder, site supervisor, QP, and accredited checker – did not prevent the disaster from arising. Each error by the involved parties could have been remediable by itself, but the fact that the collapse eventuated makes manifest the risk of relying on the vigilance of others instead of robustly performing one’s duty in a system as critical as that for ensuring building security. This was recognised in the parliamentary speeches as well. Dr Teo Ho Pin decried the “[l]ack of commitment to design and implement a comprehensive safety management system where independent checks and enforcements are carried out diligently”, ²⁰ while the Minister for National Development Mah Bow Tan expressed a similar view: ²¹

The nature of the offence, and the responsibility imposed on accredited checkers by Parliament, is thus of central importance as a sentencing consideration in this context.

34 The harm that could ensue from breach of the responsibility imposed on the accredited checker is another significant sentencing consideration, and consists of potential as well as actual harm.

35 At [77(c)] of Public Prosecutor v GS Engineering & Construction Corp [2017] 3 SLR 682 (“GS Engineering”), See Kee Oon JC observed that potential harm may be assessed by considering, among other things, the seriousness of the harm risked, and the likelihood of that harm arising. In a similar vein, Chan Seng Onn J pointed to the seriousness of the harm risked, likelihood of that harm arising, and the number of people likely to be exposed to the risk of that harm in considering the level of potential harm which might arise at [86] of Nurun Novi Saydur Rahman v Public Prosecutor and another appeal [2019] 3 SLR 413 (“Nurun Novi”). Potential harm, to my mind, encompasses all harm that could reasonably flow from breach of a duty or an unlawful act. This broad ambit of potential harm remains entirely consistent with the Prosecution’s duty to prove its case beyond reasonable doubt because that duty entails proving, beyond reasonable doubt, the existence of the potential for harm. The Court will then make its own assessment of, relying on the factors elucidated in GS Engineering and Nurun Novi, the precise level of potential harm which exists. If the risk of the particular potential harm eventuating is low or minimal, that will reduce the weight the Court places on that potential harm for the purposes of sentencing.

36 Actual harm consists of such harm actually occurring as is, subject to the rules of causation in criminal law, attributable to the offence committed by the accused. Given that both parties agree that the collapse of the viaduct on the instant facts did not arise from the appellant’s behaviour which is the subject of the proceeded charge, and that no actual harm eventuated from the appellant’s wrongs, I will not comment further at this point on actual harm.

37 Broadly speaking, in the context of civil works, the harm which the accredited checker is supposed to help guard against typically takes the form of engineering failures caused by faulty or defective design. Of course, this should not preclude other forms of harm, actual or potential, from being said to exist. In protecting against such harm, calculations independently done and checked by the accredited checker are an important part of the process (see [27] to [30] above).

38 The determination of the accredited checker’s culpability will depend on the nature of the breach. Negligence in the context of the BC Act may not merit the same type of relatively lighter treatment as compared to in other offences, such as the causing of physical injury. This is because the entire objective of the legislative framework is to ensure an independent system of checks to maintain building safety. The very purpose of the accredited checker is, as the title suggests, to check the work of others. A substantial degree of reliance is thus placed on the accredited checker to discharge his or her duties to the fullest. Simply asserting in the context of offences under s 18 of the BC Act that the accredited checker was merely negligent in, for example, making inaccurate calculations of his own while conducting his checks, and that he therefore ought not to be punished as severely because he was not “rash” or “intentional”, will not suffice. If anything, the presence of rashness or intentionality on the relevant factual matrix will be a significant aggravating factor.

39 At the other end of the spectrum from mere negligence, I acknowledge that one would not typically expect to see an accredited checker deliberately or intentionally causing a structural flaw, though it perhaps cannot be ruled out totally given the vagaries of human nature.

40 The more likely type of situation which involves greater egregiousness than mere negligence and that might be encountered in the context of offences under s 18 of the BC Act is the abandonment or abnegation of the accredited checker’s duty by either (a) signing off on building plans without having these checked at all; or (b) leaving the checking wholly to others. Between the two, the former, (a), would be more reprehensible, since no checking is done at all, as compared to the latter, in which at least some work is done.

41 General deterrence is of especial significance when the offence in question is one which affects public safety. This is clear from the longstanding decision of Public Prosecutor v Law Aik Meng [2007] 2 SLR(R) 814 at [24(d)], and is a principle of general application.

42 The Prosecution submitted that deterrence ought to be given effect as the primary sentencing consideration on the facts through utilising the full range of sentences prescribed, particularly after the latest amendments in 2008, after the Nicoll Highway collapse. I agree. As stated above, the 2008 amendments have increased the maximum punishment provided for under s 18(3) of the BC Act to a fine of S$100,000, or imprisonment of up to two years, or both. These increases represent a doubling of the maximum pre-amendment sentences. It is trite that the courts should consider the full spectrum of sentences available when determining the appropriate sentence in order to give effect to Parliament’s intention in prescribing a range of sentences for a particular offence. The holding of Chao Hick Tin JA in Ong Chee Eng v Public Prosecutor [2012] 3 SLR 776 at [24] is apposite on this point:

In this regard, the maximum, and where relevant, the minimum, sentences should be borne in mind and actively considered in sentencing: Angliss Singapore Pte Ltd v Public Prosecutor [2006] 4 SLR(R) 653 at [84].

43 This position is reinforced by the parliamentary speeches, which have also underscored that the full spectrum of available sentences should be used. During the Parliamentary Debates on the 2008 amendments to the BC Act, Minister of State for National Development Grace Fu, observed that: ²²

It was thus Parliament’s specific intent, in relation to offences under the BC Act, that custodial sentences would be imposed where appropriate. It would therefore be grossly improper to “read-out” or “read-down” the parts of the legislation which provided for custodial sentences by not imposing them altogether, or by being unduly loathe to impose them.

44 All that being said, I am mindful that any deterrent element cannot be pitched so high that suitably qualified individuals decline to offer themselves up as accredited checkers for fear that any breach, no matter how small, would sound in a criminal offence and imprisonment. That is clearly not what Parliament had envisaged in the context of the BC Act.

45 Accordingly, it will be for the Court to strike an appropriate balance between the relevant sentencing considerations. This will undoubtedly be a fact-specific assessment, but in reaching its conclusions, the Court should bear in mind the rationale underpinning the very offence in the first place. The legislative history undergirding the BC Act, as has been briefly summarised above, must inform that calibration. On the instant facts, it cannot be said that the sentence imposed by the DJ was manifestly excessive.

46 The Prosecution argued for the alignment of the framework in cases such as the present with that laid down for offences under the Workplace Safety and Health Act (Cap 354A, 2009 Rev Ed) (“WSHA”) as both seek to deter similar actions and protect public safety.

47 Beyond those arguments, it bears note that Parliament also intended for the sentencing regimes under the WSHA and the BC Act to be aligned. During the Second Reading of the Building Control (Amendment) Bill (Bill No 34 of 2007), Minister of State for National Development Grace Fu observed that: ²³

I therefore accept that the sentencing framework for offences under the BC Act can, in principle, be developed by reference to that for the WSHA.

48 In principle, I find that the two-stage approach, with sentencing bands, adopted in WSHA cases such as Mao Xuezhong v Public Prosecutor and another appeal [2020] SGHC 99 (“Mao Xuezhong”), is appropriate for application to offences under the BC Act. However, some adaptation is needed, especially to address the culpability-increasing factors that are more applicable to offences under the BC Act.

49 In ascertaining the culpability of the accused person, a three-Judge coram of the High Court in Mao Xuezhong adopted the following non-exhaustive list of relevant factors at [64(a)(ii)]:

(a) The nature of the unsafe act;

(b) The number of unsafe acts committed;

(c) The level of deviation from established procedure; and

(d) Other relevant factors such as whether the unsafe acts were motivated by the offender’s desire to save on costs.

50 To the list outlined above, one might add a number of specific considerations arising in the context of the BC Act. Some of the considerations particularly applicable to accredited checkers include whether there has been:

(a) Abandonment or abnegation of the duties imposed, such as by leaving matters wholly to others;

(b) Gross negligence, such as failing to catch errors which would be obvious and detectable with a modicum of care; and

(c) Deliberate omission to check because of corruption or being given other incentives. Such corruption may be the subject matter of a different charge, but here we are concerned with the consequence of the corruption on the duties under the BC Act.

I note for completeness that the considerations applicable to accredited checkers cited above are not exhaustive.

51 These considerations would then operate within a two-stage sentencing framework adapted from that in Mao Xuezhong as follows:

(a) The first stage is establishing the level of harm and the level of culpability in order to derive the indicative starting point according to the matrix set out below at [52].

(i) Harm includes a consideration of the degree of both actual and potential harm caused. As outlined in Mao Xuezhong and alluded to above from [35] to [37], a number of factors such as, inter alia, the seriousness of the harm risked and the likelihood of the harm arising are relevant considerations in such an assessment. Where the harm was likely to be death, serious injury (such as paralysis, loss of a limb, or loss of one of the five senses), or a serious disruption to Singapore’s key infrastructure, it could – depending on the factual circumstances – be deemed to be within the higher ranges of the “high” category even if the harm in question was potential harm that did not eventually materialise.

(ii) The determination of the accused person’s culpability may be assessed by reference to the non-exhaustive factors identified at [49] and [50] above.

(b) The second stage of the framework calls for an adjustment of the starting point according to offender-specific aggravating and mitigating factors that remain unaccounted for. In reaching its conclusions on the appropriate sentence, the Court should bear in mind that all relevant aggravating and mitigating factors should be fully considered and weighed.

52 The proposed sentencing matrix setting out the indicative starting points for the analysis of the first stage outlined above is as follows:

It bears note that these indicative starting points reflect starting points for first-time offenders who have claimed trial. They are also only indicative starting points. The Court must go on to give due weight to the offender-specific aggravating and mitigating factors that have not yet been accounted for.

53 Building on the observations of the Court of Appeal in Mohd Akebal s/o Ghulam Jilani v Public Prosecutor and another appeal [2020] 1 SLR 266 (at [20]) concerning the treatment of sentencing guidelines:

(a) First, guidelines are a means to an end and the relevant end is the derivation of sentences that are just and broadly consistent in cases that are broadly similar.

(b) Second, sentencing guidelines are not meant to yield a mathematically perfect graph that identifies a precise point for the sentencing court to arrive at in each case. Fair sentencing takes into account the facts of each specific case, while striving to achieve consistency with other similar cases. A fixation on precision carries with it a real risk of injustice, potentially ignoring specific circumstances of the case in order to fit the sentence exactly as mathematical precision demands. Accused persons and their lawyers might understandably want great predictability, and insofar as predictability promotes certainty and therefore early resolution, that should be fostered, but not at the expense of the need for judges to adjust, moderate or enhance sentences as required by the justice of the case.

(c) Third, sentencing guidelines are meant to be applied as a matter of common sense in the light of the foregoing observations. Guidelines cannot cater for all eventualities, and sometimes it is the task of the judge to depart from or even entirely ignore the guidelines, if appropriate grounds are made out.

54 Applying the framework identified above, I agree with the DJ that the appellant’s culpability was at least medium, while the harm on the instant facts could be said to be high.

55 In relation to the appellant’s culpability, I accept the Prosecution’s argument that the appellant had failed to perform any independent calculations for all the permanent corbels of the viaduct, even though the permanent corbels were key structural elements of the viaduct. Further, while the appellant did have a system in place to identify and check the key structural elements, he only played a managerial and high-level supervisory role in that system. Instead, his subordinates did most of the labour. As the facts went on to show, this was woefully inadequate and fell far short of what Parliament had intended in relation to the role of an accredited checker. The appellant’s failure to adequately check on the work done by his accredited checking organisation further militates towards a conclusion that the appellant’s culpability was, at the very least, towards the middle-end of the medium band. While it cannot be said that the appellant had failed to even ensure the existence of some checks taking place, it also cannot be said that he was an independent check on the work done by the QP and other individuals in the legislative framework for building control. The appellant’s failure to independently check on his subordinates’ work, and his taking the risk to assume that they had executed their duties in strict compliance with the provisions of the BC Act, points strongly to a finding of at least medium culpability.

56 As for determining the harm created by the appellant’s breach, I am satisfied that the DJ had not erred in concluding that the harm caused was high. As outlined above, it was accepted by the appellant that eight of the 10 permanent corbels were under-designed, with five of those corbels being unable to bear their own intended weight during the construction stage. ²⁴ There was thus a serious risk that the permanent corbels would fail, whether during construction or after the completed viaduct had been opened to traffic. A collapse at either stage would have placed life and limb in clear danger, illustrating that the harm caused, which includes potential harm, was high.

57 Further, the potential harm engendered included not only potential death or injury to persons, but also potential damage to essential public infrastructure. The consequential inconvenience and economic harm to road-users, cost of rectifying the damage, and undermining of public confidence in the structural integrity of Singapore’s public infrastructure are all significant considerations which point towards there having been high harm on the instant facts.

58 I note that it was argued by the Defence that actual harm would probably not have resulted, given the Report adduced by the criminal motion. The position taken by Mr van der Meer is broadly that given that fractures would have occurred as the ductile corbels bent, it was likely that the design errors would have been picked up by other parties like the builders or QP prior to any collapse. Thus, it was argued that the harm caused could not be described as being high.

59 The Prosecution made a number of replies to this argument, notably that (a) such an argument would undermine the rationale for having an accredited checker in the first place, given that the accredited checker is supposed to provide an independent check and not depend on other parties picking up structural failings; (b) the notion that design errors would have been picked up in time before any failure is fundamentally speculative; and (c) such an argument improperly conflates actual and potential harm because potential harm arises even if there is a likelihood that the root cause is detected before the harm eventuates. To those arguments one might add the fact that the Nicoll Highway collapse occurred despite there having been warning signs in advance of the ultimate catastrophic collapse. Those warning signs had not been sufficiently picked up on by all the parties involved. Ultimately, even taking the position advocated by Mr van der Meer at its highest, an important factor in the legislative framework of safety laws is to ensure that each and every step is indeed taken to protect safety, so that a perfect storm of errors is avoided. Arguing that others would have picked up on problems one is supposed to have addressed oneself cannot be a basis for reducing the harm engendered on the facts.

60 Overall, on the first limb of the two-stage framework, the appellant would fall within the medium culpability and high harm sector in the sentencing matrix outlined earlier. The indicative sentence would therefore be between 10 to 15 months’ imprisonment. In that regard, the starting point the DJ identified of 12.5 months’ imprisonment is largely appropriate.

61 The primary aggravating factor here was the abandonment of the very duty imposed by the legislation. I am mindful of the need to avoid double-counting, but underscore that there is no evading the fact that the appellant’s acts fundamentally involve a very significant abdication of the duties imposed on him.

62 A further aggravating factor which the DJ does not appear to have expressly recognised in her Grounds of Decision is the fact that the appellant had initially lied to the BCA. At [9] of the statement of facts, the appellant admitted that he had “initially claimed on 26 July 2017 that at the submissions stage, he had performed original calculations, checked the adequacy of the permanent corbels and found them to be inadequate”. He only admitted after he had been asked for evidence of his original calculations, which he unsurprisingly could not provide, that no calculations had been done at all. This admission only took place on 21 September 2017. It is trite that a sentencing court should take into account the full panoply of the accused person’s behaviour in sentencing, and this unsurprisingly includes the fact that the accused has been dishonest or evasive when under investigation. The effect of the accused person’s dishonesty – whether it be in the form of delays in investigations or wastage of public resources – may be taken to further exacerbate the fact of the accused having lied, but that should not take away from the fact that the accused’s dishonesty in dealing with the subsequent investigations is an aggravating factor which courts should take into consideration, even outside the context of the BC Act.

63 In mitigation, the appellant relies primarily on (a) the “clang of the prison gates” principle; (b) his plea of guilt; and (c) having taken steps to remedy the problem. I consider each of these in turn.

64 Defence counsel does not, I must underline, seek to rely on the conception of the "clang of the prison gates" principle outlined in cases like Siah Ooi Choe v Public Prosecutor [1988] 1 SLR(R) 309 (“Siah Ooi Choe”) to argue that his client deserves different treatment simply because of his position. Rather, the Defence has argued in substance instead that the appellant’s acts are out of character and aberrations in what has otherwise been a distinguished career. For the reasons I outline below at [70], I am of the view that only limited weight should be placed on that argument.

65 Apart from the conception of the “clang of the prison gates” principle relied on by the appellant, there is also a different understanding of what the principle means in the local context. In Tan Sai Tiang v Public Prosecutor [2000] 1 SLR(R) 33 (“Tan Sai Tiang”), the Court identified the underlying premise of the “clang of the prison gates” principle at [40], as being that “the shame of going to prison is sufficient punishment for that particular person convicted” and further stated that, “in order for the principle to be applicable, the convicted person must have been a person of eminence who had previously held an important position or was of high standing in society”. This conception of the “clang of the prison gates” principle appears to have been derived from Siah Ooi Choe (at [6] and [7]).

66 Insofar as Siah Ooi Choe and Tan Sai Tiang considered that a reduction in sentence is merited when the shame of going to prison is punishment enough, because of the eminence or high standing of the accused, these decisions, with respect, cannot represent the law. I note that the Prosecution took strong issue with these cases. I outline two difficulties with such a conception of the principle:

(a) First, relying on a person’s standing or position in society as a justification for leniency in sentencing is clearly at tension with the notion that all are equal in the eyes of the law. As Sundaresh Menon CJ observed at [1] of Public Prosecutor v Siow Kai Yuan Terence [2020] SGHC 82 (“Terence Siow”):

I am unable to reconcile the notion that a person of high standing should be granted additional leniency with the fundamental principle that justice should be applied equally to all.

(b) Second, insofar as the conception of the “clang of the prison gates” principle in question relies on the offender’s eminence and past contributions to society, with respect, this is not at all a sound basis for a more lenient sentence. In Stansilas Fabian Kester v Public Prosecutor [2017] 5 SLR 755 (“Stansilas”) at [84], Menon CJ made clear that it was necessary to “justify the mitigating value of public service and contributions by reference to the four established principles of sentencing: retribution, prevention, deterrence (both specific and general) and rehabilitation”. Without actually elucidating the relationship between an accused person’s eminence and past contributions on the one hand, and particular sentencing objectives which should be met on the facts on the other, there is a real worry that the Court may descend into ‘moral accounting’ or sentencing offenders on the basis of their ‘moral worth’: Stansilas from [88] to [92]. Past contributions may be relevant insofar as they show a capacity for reform, and hence reduce the need for specific deterrence, but are even at the very highest only of modest weight, and are liable to be displaced where other sentencing considerations assume greater importance: Stansilas at [102(c)]. I would further note that the Courts do not play the game of ‘Monopoly’, and that past contributions do not confer a ‘Get out of Jail Free’ card on an accused person.

67 In Siah Ooi Choe, reliance was also placed on the offender’s previous contributions to the country and society. It is, with respect, difficult to see how such contributions can be relevant to a charge involving fraud. But even setting that aside, the Court in Siah Ooi Choe expressly found at [4] that the facts of the case were “highly exceptional” and that “the charge preferred against the appellant is one of the lowest levels in terms of criminality under s 406(a) of the Companies Act”. With respect, it is highly doubtful that Siah Ooi Choe would be followed at all today.

68 Crucially, when one looks at the dicta of Lane CJ in the English case of R v Iorwerth Jones (1980) 2 Cr App (S) 134 (“Iorwerth Jones”), which was cited at length in Siah Ooi Choe ([64] supra), it is not clear that Lane CJ’s view actually supports the conception of the “clang of the prison gates” principle which Siah Ooi Choe appears to reach. In Iorwerth Jones, Lane CJ observed (at 135) that:

69 What is apparent from Iorwerth Jones thus appears to be that the offender was of generally good character, had been a first-offender, and that the offence had only involved relatively small sums of money. The offender in question had pleaded guilty to defrauding the Inland Revenue of £2,740, and I cannot see how Iorwerth Jones can be construed as showing that the “clang of the prison gates” principle operates on the basis of the accused person’s high standing in society or his eminence. At most, the judgment acknowledged that the accused person’s imprisonment may have significant knock-on effects in that the nine employees of his company may be rendered unemployed. I therefore do not see Iorwerth Jones as supporting the conception of the “clang of the prison gates” principle as espoused in Siah Ooi Choe and other similar cases. Rather, that principle should be understood as a question of the mitigatory effect of a long clean record, and of the criminal behaviour thus being out of character, which is what appellant’s counsel is, in substance, arguing here.

70 It has been recognised in cases such as Terence Siow ([66] supra; at [56(e)]) and Public Prosecutor v Teo Chang Heng [2018] 3 SLR 1163 (“Teo Chang Heng”) (at [18]) that a clean record and good conduct may show that offences were committed out of character and thus an aberration. This might then reduce the need for specific deterrence as a sentencing consideration. While that is the case, I am uncertain whether the accused person’s clean record and good conduct should necessarily be taken to reduce the need for retribution as a sentencing consideration, and in any event, I am even more doubtful whether or not considerations such as the offences having been committed out of character and as an aberration should apply where the offence is one of a failure to show due care or to discharge functions responsibly. The entire purpose of the offence created by s 18(3) of the BC Act is the prevention of what will typically be inadvertent dereliction of duty, and that will, it is hoped, be an act which is out of character. Accordingly, it would undermine the purpose of s 18 of the BC Act if the mere fact that the offence is committed out of character and as an aberration would suffice to justify a more lenient sentence. It is precisely aberrations and uncharacteristic oversights which s 18 seeks to protect against.

71 The Defence has also sought to rely on the appellant’s timeous plea of guilt as a mitigating factor. I am mindful in this regard of the discussion of the weight to be placed on a plea of guilt as a mitigating factor at [56(a)(i)] of Terence Siow, and do not propose to add substantively to it at this point. All I note is the observation by the Court of Appeal at [71] of Chang Kar Meng v Public Prosecutor [2017] 2 SLR 68, that whether and if a discount should be accorded to an accused person who pleaded guilty is a fact-sensitive matter that depends on multiple factors. A prescriptive one-third or one-quarter discount should be eschewed in favour of a substantive consideration of the facts and whether or not the plea of guilt genuinely evinces the offender’s effort to own up to his mistakes and to minimise further harm to the victim: Ng Kean Meng Terence v Public Prosecutor [2017] 2 SLR 449 at [69].

72 On the instant facts, while I accept that the appellant had pleaded guilty and therefore saved judicial resources, I note that he had done so (a) only after having initially lied about having carried out his own calculations; (b) after having been asked to produce those calculations but being unable to; and (c) in the face of clear evidence that the QP’s plans, which he had approved, contained deficient structural designs. It is well established that a plea of guilt in circumstances where the Prosecution would have had little difficulty in establishing the offence would, at least prima facie, reduce the weight to be placed on such a plea, though this will once again depend on the entirety of the facts: Wong Kai Chuen Philip v Public Prosecutor [1990] 2 SLR(R) 361 at [13] and [14], and Than Stenly Granida Purwanto v Public Prosecutor [2003] 3 SLR(R) 576 at [23]. Accordingly, some, albeit limited, weight could be placed on the appellant’s plea of guilt.

73 Taking steps to remedy the problem is one of the mitigating factors recognised in GS Engineering ([35] supra; at [77(f)]). In that case at [92], the High Court viewed the fact that the offender had been proactive in ascertaining the cause of the accident and commissioning a safety consultancy firm to investigate the accident as a mitigating consideration. Of course, the offender in GS Engineering had been the main contractor, and it would not be entirely appropriate to foist the same expectations on the instant appellant in his role as the accredited checker. The instant appellant asserts that he has put in place an improved job workflow to remedy the lapse. In particular, the lead engineer now has to prepare an accredited checker “Review Form” with the accredited checker before the accredited checker’s evaluation report is submitted. That improved workflow does little to deal with and prevent similar lapses when it was his own omission to check at all. I cannot see how the workflow improvement pointed to can be deemed to be relevant to the precise nature of the appellant’s breach.

74 On balance, bearing the aggravating and mitigating circumstances in mind, it cannot be said that the sentence imposed by the DJ was manifestly excessive. If anything, six months’ imprisonment may, on the application of the two-stage test outlined above at [51], be on the considerably shorter side, with a higher sentence more appropriate. Given all the circumstances and the fact that the Prosecution has abandoned its appeal against sentence, it would not be appropriate for me to increase the sentence in this case suo motu.

75 As a final and separate point, I address the issue of Short Detention Orders (“SDOs”) because the appellant sought to argue that if not a fine, an SDO should be imposed instead. SDOs would rarely be a substitute for imprisonment where the framework requires substantial rather than shorter periods of imprisonment to be imposed given the relatively short duration of detention under an SDO. Rather, an SDO may, in specific situations, be used to buttress sentences at the lower end of the scale or where exceptional circumstances exist. As noted by the Court in Teo Chang Heng ([70] supra; at [15]), SDOs do “carry a punitive element and [are] inherently also capable of serving to deter”, though I reiterate that this applies primarily to situations warranting only relatively short periods of custody.

76 It is, however, not strictly necessary to decide the issue in this case given my findings above, and I will say no more on this topic.

77 Considering the framework adopted, the culpability of the appellant, and the harm engendered, as well as the relevant offender-specific factors, I am of the view that the sentence of six months’ imprisonment imposed by the DJ cannot be said to be manifestly excessive. I accordingly dismiss the appeal.