GCO v Public Prosecutor [2019] SGHC 31

The appellant appealed against his sentences in respect of two charges: the first for outrage of modesty (the “OM offence”) under s 354(1) of the Penal Code (Cap 224, 2008 Rev Ed) (the “Penal Code”); and the second for insulting the modesty of a woman under s 509 of the Penal Code (the “s 509 offence”). A third charge under s 447 of the Penal Code for committing criminal trespass by entering a female toilet to commit the s 509 offence was also taken into consideration in sentencing.

The appellant pleaded guilty to both proceeded charges and was convicted. He was sentenced by the District Judge to serve a term of eight months’ imprisonment and three strokes of the cane in respect of the OM offence, and one month’s imprisonment in respect of the s 509 offence. The sentences were ordered to run consecutively, for an aggregate sentence of nine months’ imprisonment and three strokes of the cane.

The appellant appeals on the grounds that the District Judge failed to appreciate the materials placed before her, and that his sentence is manifestly excessive.

The facts

The facts are drawn from the Statement of Facts to which the appellant pleaded guilty without qualification.

The s 509 offence was committed at 5.00am on 25 November 2015. The appellant was a resident at a Hall of Residence (the “Hall”) at a university in Singapore (the “University”). On that day, the appellant was outside one of the female toilets at the Hall when he heard someone showering. He decided to enter the toilet to peep at the person who was showering. The first victim was showering at the last shower cubicle. The appellant went into the shower cubicle next to hers, locked the cubicle door, climbed the cubicle partition, and intruded into her privacy by peeping into her cubicle. The appellant saw the first victim fully naked. She noticed that someone had peeped into her cubicle, and quickly left the toilet to seek help. The appellant stayed in his locked cubicle. The first victim and her friends obtained the assistance of a campus security officer, who took a photo of the appellant inside the cubicle. The appellant then surrendered to the campus security officer. The first victim was a student at the University at the material time.

The OM offence was committed on 20 April 2017. The appellant, the second victim and her boyfriend were working on a project at a computer lab at the University through the early hours of the morning. The second victim’s boyfriend fell asleep at about 2.00am, and the second victim, at about 4.00am. At about 6.00am, the appellant, who had also fallen asleep, woke up and wanted to use the washroom. As he was walking towards the exit of the computer lab, he noticed the second victim sleeping. She was wearing a pair of denim shorts. He walked towards her, and proceeded to place his hand through the opening of her shorts. Upon feeling someone touch “her vagina area” from underneath her shorts, the second victim woke up, whereupon the appellant quickly walked away. The second victim saw the appellant walking away from her, and informed her boyfriend about the incident. She and her boyfriend confronted the appellant, who apologised to both of them. Subsequently, she informed the University authorities, and also lodged a police report concerning the incident. The second victim was the appellant’s classmate at the University at the material time.

Decision below

The District Judge’s Grounds of Decision can be found in Public Prosecutor v GCO [2018] SGMC 54 (the “GD”). The appellant pleaded guilty to both charges, and consented to have the s 447 charge taken into consideration in sentencing.

The District Judge considered that probation would not be appropriate. The appellant was 26 years old, and well above the age of 21, below which the presumptive primary sentencing consideration is rehabilitation: GD at [22]. Although adult offenders could be placed on probation, the District Judge was not persuaded that there were exceptional circumstances to warrant calling for a pre-sentence probation report. The District Judge referred to a memo prepared by an Institute of Mental Health (“IMH”) psychiatrist dated 5 July 2017, and an IMH report dated 19 February 2018 (collectively, the “IMH reports”). Although the appellant had been diagnosed by an IMH psychiatrist to be suffering from voyeurism and fetishism, the District Judge observed that these mental conditions did not remove the appellant’s mental ability or capacity to control his actions and refrain from committing criminal acts, citing Public Prosecutor v Chong Hou En [2015] 3 SLR 222: GD at [23]. Instead, these labels were merely “clinical description[s]” of a “perverse behavioural option”: GD at [23]. Indeed, the IMH reports did not find that the appellant suffered from any psychiatric condition that was causally related or had substantially contributed to the appellant’s commission of the offences: GD at [24]. In the circumstances, the dominant sentencing considerations remained general and specific deterrence. Probation was not justified, and rehabilitation, if it was necessary, could take place in the prison setting. The needs of deterrence were best served by an imprisonment term.

In respect of the OM offence, the District Judge applied the sentencing framework set out by the High Court in Kunasekaran s/o Kalimuthu Somasundara v Public Prosecutor [2018] 4 SLR 580 (“Kunasekaran”). The District Judge determined that the offence fell within Band 2 of the sentencing bands, because the appellant had intruded upon the private parts of the second victim, namely her vagina, and did so while the second victim was vulnerable because she was sleeping. There were therefore two offence-specific factors which applied: GD at [27]. The District Judge determined, however, that the offence was not at the higher end of Band 2: GD at [28].

Turning to the offender-specific factors, the District Judge noted the fact that the appellant had committed the OM offence even though he had been served with a 12-month conditional warning for having peeped over the shower cubicle wall in 2015. The warning was administered only on 16 February 2017; the appellant had offended a mere two months after receiving the warning. The District Judge found that this was aggravating: GD at [29]. In addition, she found it aggravating that the offending behaviour had escalated from an offence under s 509 of the Penal Code, to one under s 354(1) of the Penal Code: GD at [29].

The District Judge found as mitigating the fact that the appellant had pleaded guilty at the first available opportunity, and that he had no previous convictions: GD at [30].

Taking into account all these factors, the District Judge held that eight months’ imprisonment was appropriate for the OM offence. She noted that the Defence had also submitted for eight months’ imprisonment (but with no caning) if imprisonment was to be ordered: GD at [30].

The District Judge also ordered caning, following the guidance set out in Kunasekaran at [50] that caning ought to be imposed where the victim’s private parts are intruded upon. The District Judge opined that there was no reason not to impose caning. The appellant had no choice but to stop his actions when the second victim woke up; and the appellant’s actions were particularly intrusive because he had gone beneath the second victim’s shorts and over her underwear. Further, the appellant’s actions were “particularly brazen and bold” given that he had practically molested the second victim under her boyfriend’s nose: GD at [28]. There was also some suggestion of abuse of trust: the District Judge noted that the second victim was someone known to the appellant because she was his classmate, and thus would have felt safe and secure to sleep in the computer lab with him at the same place.

As for the s 509 offence, the District Judge held that it was an aggravating factor that the appellant had peeped at the first victim while she was fully naked in the shower, citing Chong Hou En: GD at [31]. It was also aggravating that the offence was committed in the early hours of the morning when there would have been hardly anyone in the vicinity to render assistance to her. The appellant had also clearly premeditated the offence as he decided to enter the female toilet. The District Judge referred to the Prosecution’s table of precedents, and noted that those who had committed s 509 offences in a similar manner received sentences in the range of four to six weeks’ imprisonment: GD at [32]. She therefore held that a sentence of one month’s imprisonment was appropriate in this case.

The District Judge held that because these were unrelated offences, the sentences should run consecutively to ensure that the appellant would be punished for each offence: GD at [33]. Thus, the total sentence ordered was nine months’ imprisonment and three strokes of the cane.

The parties’ cases

The appellant’s case

The appellant’s core contention in this appeal is that his sentence is manifestly excessive. In addition to his submissions filed for the present appeal, the appellant also adopted the submissions made in the mitigation plea in the Magistrate’s Court below.

The appellant submitted that this court should consider the option of probation by ordering a pre-sentence probation report, in light of the appellant’s mental conditions. The appellant also suggested in his written submissions that a Mandatory Treatment Order (“MTO”) might be possible.

If this court is not minded to adopt either option, however, then the appellant further submitted that the District Judge had erred in her application of the Kunasekaran framework. First, the District Judge erred in finding that the second victim’s private parts were intruded upon when the facts only mentioned that the “vagina area” was touched, and not the “vagina” itself. Second, the District Judge erred in failing to recognise that the act of molest was in the form of a mere fleeting touch. Third, the District Judge erred in taking into account as aggravating the fact that the appellant had offended while serving out the period of his 12-month conditional warning. Fourth, the sentence ordered is manifestly excessive when examined against comparable sentencing precedents post-Kunasekaran. Fifth, caning should not have been imposed; the precedents show that caning is not always imposed even where private parts are intruded upon. Sixth, credit should be given to the fact that the appellant apologised to the second victim upon being confronted by her; more outrage would have been caused to the second victim had the appellant not apologised.

If the above errors are rectified in this appeal, this court should find that because the second victim’s private parts were not intruded upon, the offence fell within Band 1 of the Kunasekaran sentencing bands. Further, once the factors wrongly found to be aggravating are removed from the sentencing analysis, this court should hold that a short custodial sentence or a fine should suffice.

As for the s 509 offence, the appellant submitted that the District Judge erred in having overly relied on the Prosecution’s table of precedents, most of which were unreasoned decisions of little precedential value. The only reasoned decision cited was also one which was out of accord with the sentencing trend, as the judgment for that decision itself explained. The sentencing trend instead shows that the typical sentence for an offence of this nature is a fine. Further, the District Judge erred in finding that the offence was premeditated; rather, it was committed on the spur of the moment.

The Prosecution’s case

The Prosecution’s case in this appeal is essentially an affirmation of the District Judge’s decision.

The Prosecution submitted that an MTO must be rejected out of hand because the Criminal Procedure Code (Cap 68, 2012 Rev Ed) (“CPC”) does not allow an MTO to be ordered in respect of an offence under s 354(1) of the Penal Code. As for probation, the Prosecution submitted that the District Judge rightly found that no “exceptional circumstances” applied in this case for the court to find that the sentencing principle of rehabilitation outweighs that of deterrence, both general and specific.

The District Judge also did not err in her application of the Kunasekaran framework.

The Prosecution further cited multiple aggravating factors here that reinforce the District Judge’s decision. These include: (1) the offence having been committed at an educational institution; (2) the offence having been surreptitiously committed under the cloak of darkness; (3) the appellant having preyed on his own classmate; and (4) the appellant’s boldness and brazenness in his conduct because he committed the offence while the second victim’s boyfriend was nearby.

As far as offender-specific factors are concerned, the District Judge also rightly took into account the fact that the appellant had reoffended a mere two months after being given a conditional stern warning, and the fact that the appellant had escalated in his offending from peeping over the shower cubicle, to committing molest.

Taking into account the offence-specific and offender-specific factors, the sentence of eight months’ imprisonment was appropriate.

Caning, too, was appropriate. There was intrusion upon a sleeping victim’s private parts. Although there was no skin-to-skin contact in this case, caning was justified because of the particularly intrusive manner in which the appellant placed his hand beneath the second victim’s shorts to touch her vagina area, albeit over her underwear. Caning reflects the sentencing principles of retribution and deterrence in this case.

The Prosecution did not make detailed submissions on the s 509 offence. It relied on a table of precedents and submitted that the present sentence is in line with those precedents.

Issues to be determined

There are three issues before this court. For analytical clarity, they will be examined as follows:

(a) Should probation or an MTO have been ordered?

(b) If not, did the District Judge err in her application of the Kunasekaran framework?

(c) Did the District Judge err in ordering imprisonment for the s 509 offence?

Issue 1: Should probation or an MTO have been ordered?

The first issue to be examined is whether either probation or an MTO should have been ordered, because if either option is taken then it is unnecessary to consider further whether the Kunasekaran framework was rightly applied, or whether imprisonment should have been ordered in respect of the s 509 offence.

In oral arguments before me, the appellant withdrew his submission that an MTO could be ordered. That concession was rightly made. The Prosecution is correct that there is no legal basis for this court to order an MTO. Section 337(1)(c) of the CPC provides that a community order, including an MTO, cannot be ordered in respect of an offence specified in the Third Schedule to the Registration of Criminals Act (Cap 268, 1985 Rev Ed). The offence of outrage of modesty under s 354(1) of the Penal Code is such a specified offence. There is therefore no legal basis for this court to consider an MTO.

I turn then to examine whether probation can be ordered. I recently touched on the relevant principles concerning when probation should be ordered in my decision in Public Prosecutor v Lim Chee Yin Jordon [2018] 4 SLR 1294 (“Jordon Lim”). In brief, probation responds to the principle of rehabilitation and can only be justified where rehabilitation is the dominant sentencing principle: Jordon Lim at [29]. Rehabilitation as a sentencing principle will presumptively take precedence where young offenders are involved, seeing as they are in their formative years and have better prospects of being reformed: Jordon Lim at [30]–[31].

That being said, adult offenders above 21 years of age can still be sentenced to probation, although this would be the exception rather than the norm. Indeed, in Sim Wen Yi Ernest v Public Prosecutor [2016] 5 SLR 207, I noted at [28] that older offenders might in fact “be more receptive to probation as they are generally more mature and better able to understand their responsibilities, the consequences of breaching probation, and the significance of being afforded a chance for reform”.

What is required, however, is that the offender demonstrate an extremely strong propensity for reform, or that there be exceptional circumstances warranting the grant of probation: Jordon Lim at [33], citing Goh Lee Yin v Public Prosecutor [2006] 1 SLR(R) 530 (“Goh Lee Yin”). Moreover, even if rehabilitation is found to apply, it can be displaced by the need for deterrence. Deterrence would become relatively more prominent, and rehabilitation correspondingly less so, if the offence is serious or the harm caused is severe, amongst other considerations: Jordon Lim at [35]. Rehabilitation would come to the fore, and deterrence recede in significance, if the offender suffers from a mental condition causally linked to the commission of the offence: Jordon Lim at [37].

I turn now to apply the above principles to the facts of the present case. Here, the appellant was approximately 23 years old when he committed the s 509 offence in 2015, and 25 years old when he committed the OM offence in 2017. He was well above the threshold of 21 years of age at and below which the presumptive dominant sentencing principle is rehabilitation. The appellant would therefore have to demonstrate an “extremely strong propensity for reform”, or show “exceptional circumstances” justifying the imposition of probation.

The appellant’s submissions in respect of rehabilitation are primarily founded on the IMH reports. The appellant has been diagnosed with voyeurism and fetishism. The IMH memo of 5 July 2017 does not set out anything more than this diagnosis, and is perfunctory in nature. The IMH report of 19 February 2018 is more detailed. This report was requested by the appellant’s previous counsel, M/s Rajah & Tann (“R&T”). Of greatest relevance is R&T’s specific request that the IMH psychiatrist give information as to the “Diagnosis of the patient and its implications on his general behavior/tendencies”. In response, the IMH psychiatrist indicated that the appellant was diagnosed with voyeurism and fetishism, and has recurrent sexual thoughts and anxiety associated with those thoughts. The IMH psychiatrist did not opine as to whether there was a causal or even substantial contributory link between the appellant’s mental conditions and his commission of the offences.

The existence of a mental condition does not ipso facto displace the need for deterrence and bring rehabilitation to the fore; it is necessary to go further and ask whether the mental condition was causally linked or had substantially contributed to the commission of the offences. The IMH reports do not shed any light on that question. To be fair to the psychiatrist, he was not asked to do so – he was asked only to opine as to the appellant’s diagnosis and its implications on his “general behaviour” and “tendencies”.

I considered the option of obtaining clarification from the IMH psychiatrist as to whether there was such a causal or contributory link in respect of these two offences, but concluded it was not necessary to do so. I say this for three reasons.

First, so far as the diagnosis of voyeurism is concerned, I consider that the Prosecution is right in citing Chong Hou En for the proposition that voyeurism is a clinical description for what is essentially a perverse behavioural option that does not deprive a person of his self-control: Chong Hou En at [55]. This was the subject of intense cross-examination and extensive expert submissions in Chong Hou En itself, and it is difficult to see how the conclusion reached as to voyeurism there would be different in this case of voyeurism here. As Chan J made clear in Chong Hou En at [48], the inquiry in that case was of a more general nature and not confined only to the specific case at hand: “[the] experts were engaged in order to assist the court in shedding light on the nature of voyeurism – specifically whether it deprives a voyeur of his exercise of self-control at the various stages of preparation to the stage when he acts out his fantasies”. If voyeurism does not cause the appellant to lose his self-control, it would be difficult to say that rehabilitation has come to the fore. This therefore deals with the diagnosis of voyeurism, although I accept it is no answer to the diagnosis of fetishism.

Second, so far as fetishism is concerned, it appears that the appellant’s fetish is for cross-dressing. The appellant is described as enjoying the thrill of being able to pass off as a female while almost getting seen, and deriving excitement from the prospect of not being caught while clothed in female attire. So described, it seems highly unlikely that the appellant’s ability to control himself would have been affected by this particular mental condition; indeed, such a submission was not made even in the mitigation plea below. That said, I am mindful that we have not had the benefit of medical advice on this particular issue. I note that at no time has the Defence said that there was any kind of causal or contributory link between the psychiatric conditions and the offences; nor does the description of the appellant’s behaviour even by the Defence suggest that there was one.

Third, and in any event, it seems to me that even if the appellant were found to possess some potential for rehabilitation, it would be eclipsed or significantly outweighed by deterrence in the present case because the offence is a serious one, following the High Court’s guidance in Public Prosecutor v Koh Wen Jie Boaz [2016] 1 SLR 334. The fact that outrage of modesty under s 354(1) of the Penal Code is serious is clearly indicated by the fact that an MTO cannot be ordered in respect of such an offence. The seriousness of s 354(1) offences which by their very nature are already serious is then compounded by the specific aggravating factors in this case such as the exploitation of the vulnerability of the sleeping victim in the case of the OM offence. Therefore, on the facts of this case, rehabilitation would recede in significance as compared to deterrence. It is thus unnecessary to further consider probation as a sentencing option. The appellant has not shown that exceptional circumstances apply in his case for probation to be considered.

I note also the appellant’s submission that he has an extremely strong propensity for reform. I accept that the appellant’s compliance with his counselling and psychiatric treatment schedules, the strong family support he receives from his family and his girlfriend, and his clean record up to the point of these offences suggest there is some truth to this. The appellant might therefore be said to fall within the other condition specified in Goh Lee Yin that he has an “extremely strong propensity for reform”. But it is unnecessary to go further into this point because, as I have just pointed out, deterrence outweighs rehabilitation in this case.

For completeness, I note that the appellant also places some reliance on the unreported decisions in Public Prosecutor v Jee Guang You, and Public Prosecutor v Tan Jian Yong, where an MTO and probation respectively were imposed for adult offenders even though they had been convicted of multiple molestation offences. These authorities do not assist the appellant as no reasons have been furnished for the courts’ decisions.

Further, it would not be appropriate to rely on newspaper reports of the courts’ decisions. The newspaper report for Jee Guang You itself indicates that the MTO was ordered in respect of the offender insulting the modesty of a woman, and not outraging the modesty of a woman. This is quite different from the present case.

Similarly, it appears from the newspaper report and other documents tendered that the offender in Tan Jian Yong was only slightly over the threshold of 21 years of age at and below which rehabilitation would be the presumptive dominant sentencing principle, seeing as he is reported to be 22 years old on 19 December 2013, and the offences were said to have been committed in May and October the previous year. In such circumstances, it is perhaps understandable why probation was ordered. The same cannot be said of the appellant here.

Issue 2: Did the District Judge err in her application of the Kunasekaran framework?

Having determined that neither an MTO nor probation is appropriate in this case, it is necessary to examine whether the District Judge erred in her application of the Kunasekaran framework.

The framework is well-established. At the first step of the Kunasekaran framework, the court must first determine which of the three sentencing bands the offending act falls within by considering the offence-specific factors, namely the degree of sexual exploitation, the circumstances of the offence, and the physical or psychological harm caused to the victim: Kunasekaran at [45], [48] and [49]. Next, at the second step of the framework, the court should consider offender-specific factors that are aggravating or mitigating: Kunasekaran at [45], [48] and [49].

I now examine the offence-specific factors raised by both parties which the parties argue change the approach taken by the District Judge.

Offence-specific factors

(1) Intrusion upon the second victim’s private parts

First, the appellant disputes that the second victim’s private parts were intruded upon. The basis for this argument is that the Statement of Facts only records that the second victim’s “vagina area” was touched. The appellant says it is crucial that the Statement of Facts was not more specific; it failed to say that the vagina itself was touched. The appellant derives support for this argument from the observations of Chan J in Kunasekaran, where the victim complained that her “groin area” had been touched. Chan J said the following at [55]:

It should also be noted for completeness that the second victim in her first information report to the police reported that someone had touched her “groin area” while she was sleeping in the computer lab.

The Prosecution says in response that the reference to “vagina area” in this case is clear enough. It essentially submits that this court should not split hairs.

I am minded to agree with the Prosecution on this point. While I do consider that the Prosecution could have framed the charge with greater specificity, it seems to me that the degree of ambiguity in this case is not as great as that in Kunasekaran. The groin area, as Chan J pointed out, covers a much wider region of the body than a person’s private parts. Conversely, the “vagina area” is a much narrower area, and I consider that even if the vagina itself had not been touched, the touching was sufficiently proximate to the vagina that it ought to be considered an intrusion of the second victim’s private parts in any event.

(2) The touch was more than fleeting

The appellant also submits that the act of molest was merely in the form of a fleeting touch.

That submission cannot be accepted. There are two conceivable senses to the word “fleeting”. In the first sense, there is the suggestion that the touch was merely momentary; in other words, it is fleeting in terms of time. In the second sense, there is the suggestion that the touch was lightly made and that it was not intrusive. The appellant’s actions were not fleeting in either sense of the word. Here, the appellant had to insert his hand through the opening of the second victim’s shorts, and manoeuvre his hand into position to touch her vagina area from underneath her shorts, but over her underwear. This could not be described as a brief or quick touch. It was, as the District Judge correctly found, highly intrusive.

(3) The Prosecution’s proposed aggravating factors

The Prosecution has raised a number of offence-specific aggravating factors in support of the District Judge’s decision. It bears noting that not all of these were raised below.

The key aggravating factor is that the second victim was vulnerable because she was asleep, and therefore unable to protect herself. The appellant concedes that he did take advantage of the second victim while she was sleeping and in a vulnerable position.

In addition to this, the Prosecution in its written submissions argues that four other aggravating factors were present here. I shall dispose of them summarily, because they too were raised only summarily and not developed in any detail.

I first note that the Prosecution no longer relies on one of these factors. The Prosecution originally suggested that a possible aggravating factor was that the offence was “surreptitiously committed under the cloak of darkness”. The Prosecution eventually withdrew this submission in oral argument. This was rightly conceded as the Statement of Facts says nothing about whether the lights were switched on or switched off in the computer lab at the time of the offence.

I turn then to the three remaining factors. The first proposed aggravating factor is that the offence was committed at an educational institution. This carries overtones of the exploitation of vulnerable students. But such an argument is hardly persuasive at this level where the educational institution is a university, and the appellant and the second victim are both fairly close in age. There is hardly anything to distinguish this offence from those committed in some other venue.

The second proposed aggravating factor is that the appellant preyed on his own classmate, when she was entitled to feel safe sleeping in the laboratory where the appellant was also present. The short answer to this is that the appellant and the second victim were peers, being classmates, and there is nothing to suggest that the appellant was placed in some kind of position of trust vis-à-vis the second victim which he abused or exploited.

The third proposed aggravating factor is that the appellant was brazen and bold in his conduct because he molested the second victim with full knowledge that the second victim’s boyfriend was sleeping close to her. In the District Judge’s words, the appellant molested the second victim “practically under the boyfriend’s nose”: GD at [28]. I fail to see how this is an aggravating factor. The second victim and her boyfriend were both asleep at the time. It did not take the appellant much brazenness or boldness to act while safe in the knowledge that neither of them was awake to stop him. It might have been brazen and bold if the appellant had instead committed his acts while they were awake to witness him and stop him from carrying them out.

Conclusion on offence-specific factors

In sum, neither the appellant’s nor the Prosecution’s arguments on the offence-specific factors really alters the District Judge’s analysis. There was indeed a high degree of sexual exploitation on these facts, involving the intrusion into the second victim’s private parts while she was asleep and vulnerable. The presence of these aggravating factors suggests that the acts fell within Band 2 of the Kunasekaran framework. The absence of skin-to-skin contact with the second victim’s private parts and the absence of any further aggravating factors also suggest that the acts did not fall within the upper end of Band 2.

Band 2 has a sentencing range of five to 15 months’ imprisonment: Kunasekaran at [49]. Having regard to the facts of this case, I agree with the District Judge that eight months’ imprisonment would be appropriate as an indicative starting point.

Offender-specific factors

I turn then to consider the offender-specific aggravating and mitigating factors.

(1) The appellant’s conditional stern warning is irrelevant to sentencing

The District Judge held that the appellant reoffending just two months after having been served with a 12-month conditional warning was an aggravating factor. The Prosecution supports this on appeal. The appellant submits that this was wrongly taken into account.

The relevant principles concerning the legal weight to be given to a warning by law enforcement agencies were set out in the High Court decision of Wham Kwok Han Jolovan v Attorney-General [2016] 1 SLR 1370 (“Jolovan Wham”). There, the High Court held that a warning is not binding on its recipient such that it affects his legal rights, interests or liabilities: at [33]. Instead, a warning is “no more than an expression of the opinion of the relevant authority that the recipient has committed an offence”: at [34]. A warning “does not and cannot amount to a legally binding pronouncement of guilt or finding of fact”, because only a court of law has the power to make such a pronouncement or finding: at [34]. It followed from this that a court “is not entitled to treat a warning as an antecedent or as an aggravating factor since it has no legal effect and is not binding on the recipient”: at [44].

These propositions of law were recently cited with approval by the High Court in Public Prosecutor v Raveen Balakrishnan [2018] 5 SLR 799 (“Raveen Balakrishnan”), which also elaborated the point that a stern or conditional warning issued by the relevant authorities in the exercise of prosecutorial discretion is incomparable with a judicial determination of guilt or a judicially determined sentence: at [113]–[115].

In oral submissions before me, the Prosecution suggested that the holding in Jolovan Wham was applicable only to stern warnings, but not to a conditional stern warning, which was what was given to the appellant here. The Prosecution recognised that the Attorney-General had submitted before the court in Jolovan Wham that “it would be wrong for a court to take into account a prior warning, whether as an antecedent or not, for the purpose of sentencing”, and moreover, that the Attorney-General had “stressed that the Prosecution would not in future mention a prior warning to a court for the purpose of enhancing a sentence”: Jolovan Wham at [43]. But the Prosecution suggested that those comments were to be confined only to stern warnings simpliciter, where the alleged offences which formed the basis of the stern warnings would not be raised before the court upon the offender being brought to court for a separate, later offence. Conversely, in the case of conditional stern warnings, upon the condition being breached, the offender would expect to be charged for the offence which was the subject of the warning.

I note that no reference was made to Jolovan Wham or Raveen Balakrishnan in the proceedings below with respect to the legal effect to be given to a warning in sentencing. In my view, it is not correct to take into account the fact that the offender has either (a) received a conditional stern warning; or (b) breached the condition of that conditional stern warning, as discrete aggravating factors in sentencing. I say this for two primary reasons.

First, it appears to me that what the High Court held in Jolovan Wham applies equally to stern warnings and conditional stern warnings. A stern warning is nothing more than an expression of the relevant authority’s opinion that the offender has committed an offence, and that if he were to subsequently engage in criminal conduct, leniency may not be shown to him and he may be prosecuted for the subsequent conduct: Jolovan Wham at [33]–[34]. The conditional stern warning is essentially that same warning appended with a condition that the offender must not reoffend, and if the offender breaches the condition by reoffending, it would be the authority’s intention to proceed to charge the offender for the offence he was warned about (“the warned offence”), in addition to the fresh offence he has just committed. One is the expression of an opinion; and the other, an expression of opinion coupled with a statement of intent. It is not apparent to me how the inclusion of this additional statement of intent creates any meaningful distinction between the two types of warning. After all, a statement of intent merely serves to inform of a course of action that might be taken in the future.

Further, the statement of intent to prosecute for the warned offence does not bind the offender in any way. It was made clear in Jolovan Wham at [34] that a warning does not bind the recipient or affect his legal rights. Instead, a warning plays only the informational function of communicating to the recipient that if he subsequently engages in criminal conduct, he might be prosecuted for it: at [33]. The communication in the case of a conditional stern warning is a little more specific, in that it goes further to warn him of the authority’s intent to prosecute him for the warned offence if he should reoffend. But this, too, is only informational in nature. It cannot bind the alleged offender because what the authority intends to do is the authority’s prerogative. Moreover, it may not even bind the authority because, as the court in Jolovan Wham noted at [37], the Attorney-General is not bound to consider whether a prior warning has been given before deciding whether to prosecute.

Because a conditional stern warning, like a stern warning, is only informational in nature, there is no reason to find that a person having received this information and reoffending in spite of this information should be considered ipso facto to have aggravated his offence. The offender did not legally bind himself not to do certain things, such that his doing them ought to be considered aggravating and warranting increased punishment.

Second, it is relevant to consider how the deterrent effect of the conditional stern warning is to be achieved. It is sufficiently clear from the Prosecution’s submissions that in general the conditional stern warning requires the alleged offender not to reoffend within a certain span of time, and the warning is that the Prosecution would charge the offender for the warned offence if it turns out that this offender offends again in that given span of time.

Framed in those terms, it appears to me that upon the condition being breached by the appellant reoffending – in this case, committing the OM offence after having been given a conditional stern warning for the s 509 offence – the promised action was duly taken by the Prosecution in charging the appellant for the s 509 offence as well as the OM offence. Whatever leniency that was extended to him through the conditional stern warning for the s 509 offence was extinguished and thus it appears to me that the warning was spent. There can be no deterrent value to be derived from a hidden consideration not spelt out to the alleged offender: that if he reoffends, the very fact of reoffending in breach of the conditional stern warning will also be held against him as a separate aggravating factor in sentencing for the fresh offence.

It is well-settled that the prosecutorial discretion must be judiciously exercised and whether the Prosecution ultimately chooses to prosecute for the warned offence is something entirely within its power. Indeed, the recipient of the conditional stern warning cannot be heard to complain of being taken by surprise should he end up being prosecuted for the warned offence. After all, he should expect that to occur if he reoffends. But the question whether something is an aggravating factor or not must be a legal question that only the courts can determine. It has not been established that reoffending in breach of a warning is an aggravating factor. Instead, the contrary was plainly set out and accepted by the Attorney-General’s Chambers in Jolovan Wham at [43]–[44]. I consider that there is no reason to adopt a different position in this case.

In the present case, there is nothing to suggest that the conditional stern warning given to the appellant specifically said that reoffending in breach of the condition to the warning would be considered an aggravating factor over and above the threatened action of preferring a charge, as was ultimately done. Thus, even if I am wrong in my views on the above principles, it would not be correct in this specific case to find it an aggravating factor that the appellant committed the OM offence in breach of the condition.

In addition, the Prosecution suggests that the appellant ought to have cherished the opportunity not to be charged for the s 509 offence by not committing another offence. He is said to have “squandered” this opportunity when he committed the OM offence and thus this ought to be considered an aggravating factor. This argument was made only very briefly, and only at the stage of oral arguments before me. In the face of the established position in Jolovan Wham, with which I am in broad agreement, I do not presently consider it necessary to delve into whether or not merely being warned (whether conditionally or otherwise) ought to be considered an “opportunity” that can be squandered, with possible legal implications in terms of aggravation of sentence.

For the reasons outlined above, I am of the view that the District Judge erred in law in finding as a discrete aggravating factor the fact that the appellant had committed the OM offence during the 12-month conditional stern warning period. It remains the case that a warning – whether a stern warning, or a conditional stern warning – has no legal effect in sentencing.

All this, however, is not to say that the conditional stern warning serves no real purpose. It carries force because the authority is in a position to carry out its threatened course of action, ie, to prosecute for the warned offence. The means by which the conditional stern warning seeks to have an effect on deterring criminal behaviour is through the exercise of prosecutorial discretion to prefer a charge for the warned offence; it is not because reoffending in breach of a warning will be considered to be an aggravating factor in the sentencing analysis.

(2) Escalation of the offences

The Prosecution also submits that a separate aggravating factor present here is the escalation in the appellant’s offending from insulting the modesty of a woman by peeping over the wall of the shower cubicle in a female toilet (the subject of the s 509 charge) in November 2015, to committing outrage of modesty one and a half years later in April 2017.

This submission poses an interesting conceptual question: can conduct constituting criminal behaviour that takes place prior in time to a separate charge be considered an aggravating factor for that charge if the offender is only convicted on both charges at the same time? This question arises from our facts. Although the appellant committed the actions which comprise the subject of the s 509 charge in November 2015, he was never charged for that offence then. He was only charged for that offence in 2017 when the charge for the OM offence was also brought against him, and he pleaded guilty and was convicted and sentenced on both charges together.

It is trite law that it is an aggravating factor for an offender to have committed an offence of a similar nature to the one for which he is presently being charged, because it may reflect a pattern or tendency for repeat offending: Tan Kay Beng v Public Prosecutor [2006] 4 SLR 10 (“Tan Kay Beng”) at [14]. Even dissimilar antecedents that clearly manifest a marked and progressive proclivity towards criminal activity or a cavalier disregard for the law may be relevant: Tan Kay Beng at [14]. But the appellant here cannot be said to have an antecedent in respect of his actions committed in 2015, because he was convicted and sentenced on both the OM and s 509 offences at the same time.

An antecedent refers to the appellant’s prior conviction(s), and not his prior criminal behaviour. This is evident from the case law and relevant academic commentary. The High Court in Tan Kay Beng (at [15]) cited an extract from Thomas, Principles of Sentencing: The Sentencing Policy of the Court of Appeal Criminal Division (Heinemann, 2nd ed, 1979) with approval. I reproduce the extract here:

This extract makes it clear that an antecedent refers to an offender’s previous convictions, because the references to “offences” in “earlier offences” and “previous offences” must mean a legal determination of guilt. Singapore academic commentary concurs. Kow Keng Siong has this to say at para 07.072 of Sentencing Principles in Singapore (Academy Publishing, 2009):

It is clear, therefore, that in the absence of a conviction for an accused’s prior behaviour, it is not to be considered an antecedent. It is evidence of his predisposition or character at best.

Specific deterrence, as a sentencing principle, addresses an offender’s propensity to offend, which usually manifests itself in the form of repeat offending, in other words, the presence of antecedents. The escalation in criminal offending is met with a corresponding escalation in criminal punishment. But simply because this principle might be implicated does not mean that one should stretch the meaning of an “antecedent” and take it out of its established meaning, ie, a previous legal determination of guilt in relation to a criminal offence. The criminal behaviour might have occurred in the past, but the legal determination of it came at the same time as the latter offence for which the former acts are argued to be aggravating. The legal determination was not made in the past. The appellant must therefore be considered a first offender in respect of both charges here, as he was convicted on both charges at the same time. Indeed, the Prosecution has not attempted to argue that the appellant is a repeat offender in this case.

The question then is whether it is still correct to say that the appellant has escalated his criminal behaviour, when he was never convicted for the s 509 offence before being convicted on the OM offence. In my view, it is still a relevant aggravating factor that the appellant has carried out criminal acts on two occasions. Specific deterrence is directed at deterring the individual offender from reoffending, and whether it applies depends on whether the offender has a propensity to reoffend. This propensity is discernible from the frequency of his criminal behaviour, ie, the fact that two similar sexual offences were ultimately committed. The principle of specific deterrence is therefore raised on these facts.

That said, the present situation should be considered less aggravating than the situation where the offender was previously convicted for the first set of criminal behaviour. In that scenario, it is well-established that having a relevant criminal antecedent is an aggravating factor justifying more severe punishment, because the punishment the offender previously received was not sufficient to deter him from offending again. The previous conviction and punishment would have served as a signal to that offender not to reoffend. The appellant here, however, has not received such a signal.

If the frequency of offending is relevant, the next question that follows is whether it is relevant to take into account the fact that the criminal behaviour forming the subject of the first offence was less serious than that forming the subject of the later offence. In my view, this is also relevant. The risk of reoffending is not only to be discerned from the offender committing two offences. It is also discernible from the offender committing more serious criminal acts. This is the “escalation” in criminal behaviour that has occurred in this case. The Prosecution is correct that the appellant’s criminal actions have increased in severity, from insulting the modesty of a woman, to outraging the modesty of a woman.

Thus, I accept that the escalation in criminal behaviour is an aggravating factor in this case, because the appellant’s criminal behaviour has given rise to two separate sexual offences, the latter being more serious than the first. This clearly demonstrates his propensity to reoffend. Specific deterrence is rightly engaged in this case.

(3) Mitigating factors

The mitigating factor in this case is that the appellant pleaded guilty at the first available opportunity, thereby demonstrating remorse and also saving judicial and prosecutorial resources. I consider that the appellant’s apology to the second victim can also be given some weight as a demonstration of remorse. But this will be taken in the round with the early plea of guilt as evidencing the appellant’s remorse.

Conclusion on imprisonment under the Kunasekaran framework

Having regard to the aggravating and mitigating factors in this case, I see no reason to disturb the indicative starting point of eight months’ imprisonment. The mitigating factor of the appellant’s early plea of guilt and apology effectively negates the aggravating factor of the escalation in criminal behaviour.

Should caning have been imposed

The next question, however, is whether caning should have been imposed. The District Judge applied the guidance from Kunasekaran at [50] that caning ought to be imposed as a “starting point” where the outrage of modesty involves the intrusion upon the victim’s private parts or sexual organs. The District Judge acknowledged that there was no skin-to-skin contact, but considered the molestation here to be “particularly intrusive” as the appellant went beneath the second victim’s shorts and over her underwear.

In my view, caning ought not to be imposed in this case. Caning has typically been imposed on offenders in cases applying the Kunasekaran framework where there was skin-to-skin contact, or the contact was prolonged, or there was an element of restraint applied to the victim, although these aggravating factors should not be taken to be exhaustive. None of these three factors is present here. As for the District Judge’s point about the sexual exploitation being particularly intrusive, that has already been adequately accounted for in the substantial term of imprisonment imposed. There is nothing to suggest that the appellant sought deliberately to prolong the contact, or that he persisted after the second victim woke up. Nor is there any suggestion of abuse of trust, which featured in the Prosecution’s precedent of Liew Hoo Ling v Public Prosecutor (Magistrate’s Appeal No 9155 of 2016), which is the most factually analogous to the present case. It is therefore appropriate that the sentence of caning be set aside.

Issue 3: the s 509 offence

The District Judge held that the appropriate sentence for the s 509 charge was an imprisonment term of one month. The District Judge considered that it was aggravating that the first victim was fully naked, citing Chong Hou En. It was also aggravating that the offence was clearly premeditated because the appellant decided to enter the female toilet to carry out his actions: GD at [31].

The appellant submits that the District Judge erred in finding that the offence was premeditated, and erred also in placing too much reliance on the Prosecution’s table of sentencing precedents, which precedents were mostly unreported. The appellant submits that the usual sentence for peeping tom cases like the present is a fine.

The Prosecution affirms the decision of the District Judge, and has adduced one new precedent, Public Prosecutor v Ge Xiang (Magistrate’s Arrest Case No 912768 of 2017).

There is authority for the appellant’s submission that the normal sentence is a fine. In Public Prosecutor v Mohammed Liton Mohammed Syeed Mallik [2008] 1 SLR(R) 601, an appeal was brought by an offender who had been convicted under s 509 for taking four photographs of the victim in the nude. A large part of the judgment is concerned with the aggravating factor of the use of technology in that case. But the Court of Appeal did opine as follows at [89]:

Chan J in Chong Hou En also took the view that he “would be chary in concluding that a custodial sentence should be the starting point the moment a recording device is used”: at [77]. Instead, he found that the factors of planning and premeditation were more crucial in the sentencing analysis. Since the position is that a custodial sentence is not warranted as a starting point even where the aggravating factor of the use of technology is present, it can be said that the threshold for a custodial term ought to be even higher where no such use of technology is present, as here.

The precedents cited by the Prosecution are mostly unreported decisions, as the appellant has noted. Moreover, they do appear to conflict with the reported decision in Tan Pin Seng v Public Prosecutor [1997] 3 SLR(R) 494 (“Tan Pin Seng”). The commentary in Sentencing Practice in the Subordinate Courts (LexisNexis, 3rd ed, 2013) takes the view (at p 606) that in general, a fine of $1,000 to $2,000 is the norm for an s 509 offence, citing Tan Pin Seng. Resort to the sentencing precedents therefore does not appear to be of much assistance in this case.

The key point then is whether the offence was premeditated. It is certainly true that the appellant should not have entered the female toilet. But that is the subject of the s 447 charge that has been taken into consideration for purposes of sentencing. Otherwise, there is nothing to suggest that the appellant had applied a great deal of foresight or planning prior to committing the offence. All he had to do was to walk into the toilet, and find the relevant cubicle. The facts as described in the Statement of Facts do not suggest that he had given a great deal of thought to planning this entry; instead, the facts appear to suggest that he acted on a whim “when he heard someone showering inside the toilet”.

There is some suggestion in the GD that the District Judge thought the appellant came to be outside the toilet because he was harbouring some sinister intent. However, that is not disclosed in the Statement of Facts. All that is said is that at about 5.00am, the appellant was “outside the female toilet”. And although it is true that the appellant in his own mitigation plea admits to having an urge to “have with him pieces of female clothing while taking showers in the hall’s female toilets as part of his desire to act as a female”, this general pattern of behaviour does not permit a clear inference that the appellant had gone to the female toilet with the specific intent to insult the modesty of a woman. It amounts at most to an admission that he habitually trespassed into female toilets. After all, the mitigation plea itself also records that the appellant was “suddenly overcome by an inexplicable urge to peep” at the first victim, which appears to be consistent with the Statement of Facts. In this instance, I consider that the benefit of the doubt should be given to the appellant. I therefore find that the appellant did not premeditate the s 509 offence, and the custodial threshold has not been crossed.

In my view, it is therefore appropriate that the sentence of one month’s imprisonment be substituted with a fine of $2,000, in default two weeks’ imprisonment.

Conclusion

For the foregoing reasons, I allow the appellant’s appeal in part. In respect of the OM offence, a sentence of eight months’ imprisonment is appropriate, and the sentence of caning is set aside. In respect of the s 509 offence, a fine will suffice. The aggregate sentence is therefore eight months’ imprisonment, and a fine of $2,000, in default two weeks’ imprisonment.