1 The appellant, aged 23 at the time of the offence, pleaded guilty to the following four drug-related charges in the District Court:

(a) One charge of possessing not less than 9.98g of diamorphine for the purpose of trafficking, an offence under s 5(1)(a) read with s 5(2) of the Misuse of Drugs Act (Cap 185, 2008 Rev Ed) (“MDA”) punishable with enhanced punishment under s 33(4A)(i) of the MDA (“the trafficking charge”);

(b) One charge of consuming methamphetamine, an offence under s 8(b)(ii) of the MDA punishable under s 33(4) of the MDA (“the enhanced consumption charge”);

(c) One charge of possessing not less than 1.12g of MDMA, an offence under s 8(a) of the MDA punishable with enhanced punishment under s 33(1) of the MDA (“the possession charge”); and

(d) One charge of possessing utensils for the intended consumption of drugs under s 9 of the MDA punishable under s 33(1) of the MDA (“the utensils charge”).

2 Five other drug-related charges were taken into consideration for the purpose of sentencing (“the TIC charges”): three for the repeat offence of trafficking of various drugs, one for joint possession of methamphetamine punishable under the enhanced regime, and one for joint possession of utensils for drug taking.

3 The appellant pleaded guilty at the court below and the District Judge (“the Judge”) convicted her and sentenced her to 21 years and three months’ imprisonment, with the sentences for the trafficking charge and the utensils charge to run consecutively: see Public Prosecutor v Katty Soh Qiu Xia [2018] SGDC 50 (“the GD”) at [3]. The appellant now appeals against her sentence. Given that the bulk of her imprisonment term stems from the trafficking charge, she understandably focuses her appeal on that sentence.

4 What is notable about this appeal is that between the time the Judge delivered his decision on 8 March 2018 and the time I heard this appeal on 28 September 2018, a new High Court authority emerged which is on all fours with the present case. On 29 June 2018, Sundaresh Menon CJ delivered his decision in Public Prosecutor v Lai Teck Guan [2018] SGHC 151 (“Lai Teck Guan”) which arose out of a Magistrate’s Appeal to the High Court. Lai Teck Guan laid down the sentencing framework for a repeat offender trafficking in up to 15g of diamorphine. Prior to Lai Teck Guan, the sentencing benchmarks for trafficking in diamorphine were set out in Vasentha d/o Joseph v Public Prosecutor [2015] 5 SLR 122 (“Vasentha”) (for less than 10g of diamorphine) and Public Prosecutor v Tan Lye Heng [2017] 5 SLR 564 (“Tan Lye Heng”) (for 10g to less than 15g of diamorphine). However, both High Court authorities dealt solely with the sentencing benchmarks for first-time offenders. Under the MDA, repeat offenders face a different sentencing regime from first-time offenders.

5 In determining the sentence for the trafficking charge, the Judge chose to apply Vasentha by mathematically extrapolating the degree of uplift from the sentencing benchmarks in that case (which are meant to apply to a sentencing range of five to 20 years) to fit the sentencing range for a repeat offender (ie, ten to 30 years): GD at [41]. However, this approach was expressly rejected in Lai Teck Guan (at [30]). Menon CJ stated that when sentencing a repeat offender, the court is concerned with not just the quantity of drugs trafficked, but also the circumstances in which the repeat offence came about.

6 I note that as a general rule, judicial pronouncements are presumed to be retroactive in effect until and unless expressly stated otherwise. Further, the onus of establishing that there are grounds for the court to exercise its discretion to limit the retroactive effect of a judgment is on whoever seeks the court’s exercise of that discretion (see Adri Anton Kalangie v Public Prosecutor [2018] 2 SLR 557 at [70]). Before me, both the Prosecution and the appellant accept that Lai Teck Guan is applicable to the present case. ¹ Therefore, I do not consider it necessary to examine whether or not the doctrine of prospective overruling applies. Accordingly, given Menon CJ’s express rejection of the pure mathematical extrapolation approach, the Judge’s decision on sentence for the trafficking charge is wrong in principle and on that basis alone is liable to be set aside.

7 Be that as it may, I heard parties’ arguments and reserved judgment to scrutinise the precedents and sentencing benchmarks that the parties had relied on. For the reasons that follow, I allow the appellant’s appeal and set aside the sentence imposed by the Judge. In its place, I impose an aggregate sentence of 15 years and nine months’ imprisonment, with the trafficking charge and the utensils charge continuing to run consecutively and the remaining charges to run concurrently.

8 The detailed statement of facts which the appellant admitted to can be found at [4] to [16] of the GD. In any event, the facts are uncomplicated and can be summarised as follows. On 4 July 2016, the appellant’s accomplice was arrested and subsequently implicated the appellant as her supplier. Thereafter, the appellant was arrested and a search was conducted on her residence. The drugs and utensils that formed the basis of the charges against her were discovered. At the police station, the appellant’s urine was analysed and found to contain methamphetamine.

9 The appellant admitted that the diamorphine in her possession was for sale and she would earn a commission for each transaction. The appellant stated that she was selling the drugs to supplement her income. The appellant had previously been convicted on 11 July 2013 for trafficking in a controlled drug for which she was sentenced to reformative training. This therefore rendered her liable for enhanced punishment for the trafficking charge under s 33(4A)(i) of the MDA.

10 As alluded to earlier, the Judge convicted the appellant, who pleaded guilty, and sentenced her to the following:

(a) 21 years’ imprisonment for the trafficking charge;

(b) the mandatory minimum of three years’ imprisonment for the enhanced consumption charge;

(c) nine months’ imprisonment for the possession charge; and

(d) three months’ imprisonment for the utensils charge.

The sentences for the trafficking charge and the utensils charge were ordered to run consecutively with the rest of the sentences to run concurrently. The aggregate sentence was therefore 21 years and three months’ imprisonment. The appellant’s sentence was ordered to be backdated to commence from 5 July 2016, the date on which the appellant was remanded (at [3] and [59] of the GD).

11 In relation to the trafficking charge, the Judge adapted the sentencing benchmark in Vasentha to formulate a similar sentencing benchmark for a repeat offender trafficking in less than 10g of diamorphine (ie, 10 to 30 years) (at [41] of the GD). Applying the sentencing benchmark that he had formulated, the Judge found that the indicative starting point for the sentence to be imposed on the appellant was 22 years (GD at [54]). This sentence corresponded to the highest end of the sentencing range, given that the appellant had trafficked close to the highest end of the weight range. The Judge found that a clear aggravating fact was the five TIC charges, three of which involved trafficking. However, he considered as a mitigating fact the relative youth of the offender and that apart from her stint in reformative training, she had not yet been imprisoned for a significantly long period. Accordingly, he moderated the starting point sentence down by one year to arrive at a sentence of 21 years for the trafficking charge (GD at [55] and [56]).

12 On the enhanced consumption charge, the Judge gave the mandatory minimum of three years without further elaboration. On the utensils charge, the Judge also did not elaborate on his decision given that the sentence was not challenged. On the possession charge, the Judge found that the Prosecution’s proposal of nine months was well within the bounds of sentencing precedent and would be more realistic given the appellant’s antecedents and the number of charges that she was facing (GD at [57]).

13  Finally, the Judge ordered the sentence for the trafficking charge and the sentence for the utensils charge (which was the shortest sentence) to run consecutively, again taking into account the appellant’s youth and that she had hitherto not served such a long imprisonment sentence (GD at [58]).

14 The parties’ cases on appeal are heavily reliant on the sentencing framework set out in Lai Teck Guan. Therefore, I consider it helpful to briefly set out the framework here before moving on to summarise the parties’ cases.

15 In Lai Teck Guan, Menon CJ set out the approach to be taken when sentencing repeat-offenders who traffic in less than 15g of diamorphine as follows (at [38]):

16 Following from this, Menon CJ set out a table from which a sentencing judge could determine the starting sentence and the indicative uplift to be imposed based on the weight of the diamorphine (Lai Teck Guan at [42]):

17 For completeness, I also consider it helpful to set out the statutorily imposed sentencing ranges under the MDA for the various diamorphine trafficking offences as follows:

18 On appeal, the appellant seeks to have her sentence for the trafficking charge reduced to between 12 to 17 years’ imprisonment. ⁶ The appellant’s main contention is that, based on the sentencing benchmark in Lai Teck Guan, the indicative starting point for her sentence should be between 16 to 21 years’ imprisonment. Therefore, by calibrating the existing sentence based on a starting point of 22 years’ imprisonment, which exceeds the indicative starting point obtained when applying Lai Teck Guan, the sentence imposed by the Judge is manifestly excessive. ⁷

19 Additionally, the appellant emphasises that rehabilitation should be the dominant sentencing consideration, given her young age and the importance of re-integrating her back into society. The rehabilitative efforts of the Singapore Prison Service may be undermined if she is given an overly long prison sentence and consequently released at an age when she will be unable to find sustained employment. ⁸

20 The Prosecution agrees with the appellant that Lai Teck Guan is the applicable law, and therefore that the Judge was wrong to have used a sentencing framework which was based solely on a mathematical uplift from Vasentha. ⁹ The Prosecution submits that applying the sentencing framework in Lai Teck Guan to the present case, the sentence should be calibrated as follows:

(a) If the appellant were a first time trafficker, the starting point should be 15 years’ imprisonment since 9.98g is extremely close to the upper limit of the 9 to 9.99g weight range. ¹⁰

(b) The indicative uplift should be five years’ imprisonment, given that the appellant had re-offended a mere ten months after being released from the Reformative Training Centre (“RTC”). ¹¹ This would result in an uplifted sentence of 20 years’ imprisonment.

(c) A slight upward adjustment of one years’ imprisonment would be warranted to take into account the appellant’s overall culpability given the TIC charges. ¹²

21 This would result in a final sentence of 21 years’ imprisonment, which is essentially the same as what was imposed by the Judge. Therefore, the sentence cannot be said to be manifestly excessive.

22 The first issue to be determined is whether or not the sentencing framework in Lai Teck Guan ought to even apply to the present case. If I should find that Lai Teck Guan ought to apply, I will then determine what the correct sentence should be based on that framework.

23 If, however, I should find that Lai Teck Guan ought not to apply, I will then have to consider what the appropriate sentencing framework should be and what sentence the application of such a framework would yield.

24 With respect, I am of the view that the sentencing framework in Lai Teck Guan ought not to apply in its entirety because there are errors in certain parts of the framework. I will first analyse the errors arising from the part of the framework relating to the terms of imprisonment, given that this would be directly relevant to the present case. For completeness, I will then make some observations on the part of the framework dealing with the number of strokes of the cane. I demonstrate these errors in the form of a graph as shown below:

25 As a preliminary point, I have, for convenience, indicated certain weight points as ending in “.99g”. In doing so, I mean to refer to the weight point which is just slightly less than the next whole number, as opposed to the exact value of “.99g”. For example, “9.99g” is actually meant to refer to the weight point which is infinitesimally less than 10g.

26 I derived this graph by plotting out the data points as set out in the table at [42] of Lai Teck Guan (reproduced in this judgment at [16] above) on two axes. The key features of this graph are as follows:

(a) The weight of the diamorphine (in grams) is represented on the horizontal axis (x-axis), and the corresponding sentence (in years of imprisonment) is represented on the vertical axis (y-axis).

(b) The bottom-most curve indicates the imprisonment terms which would be imposed on a first-time offender (the “First Time Curve” or the “FT Curve” in short). These figures are derived from Vasentha at [47] (for less than 10g of diamorphine) and Tan Lye Heng at [125] (for 10g to 14.99g of diamorphine), which were subsequently consolidated in Lai Teck Guan.

(c) The middle curve indicates the imprisonment terms which would be imposed on a repeat offender if the minimum indicative uplift were to be given (the “Minimum Indicative Uplift Curve” or the “Min Curve” in short).

(d) The upper-most curve indicates the imprisonment terms which would be imposed on a repeat offender if the maximum indicative uplift were to be given (the “Maximum Indicative Uplift Curve” or the “Max Curve” in short).

(e) One assumption that I had to make is that the relationship between each of the data points (as represented by the dots in the graph) is linear. For example, moving from the first data point at the bottom left corner of the graph (at slightly more than 0g and five years’ imprisonment) to the next data point (at 3g and six years’ imprisonment), I assumed that the indicative starting sentence increases at a constant rate as the weight of the diamorphine increases. This is represented graphically by a straight line connecting the two data points. In my view, this is a logical assumption to make, given that the range of sentences in terms of years of imprisonment necessarily lie on a continuum which corresponds to the weight of the drugs trafficked. Therefore, each weight point of diamorphine (based on the amount as stated in the trafficking charge) should, in theory, have a unique indicative starting point sentence of imprisonment which corresponds with it.

27 On a visual inspection of the graph above, three errors immediately become apparent which will be discussed below.

28 There are “gaps” in the range of indicative starting sentences present at several weight points, as indicated by the dotted ovals in the graph at [24] above. To illustrate why this is problematic, let us assume that the weight of the diamorphine in a hypothetical first-time trafficking offence is exactly 8g. The graph above shows that at this particular weight point of 8g, there are two possible starting point sentences ie, nine or ten years’ imprisonment. There are three sub-issues which arise from the gap problem.

(a) First, a sentencing judge confronted with a weight point of 8g will be unable to derive a single starting point sentence for a notional first-time offender when applying this framework. Therefore, there may be discrepancies between the starting point sentences adopted by different sentencing judges even though the weight of the diamorphine may be exactly the same. This should not be the case given that at this first stage of the inquiry, the sentence for a notional first-time offender should be determined solely based on the weight of the diamorphine and no further discretion is exercised: see Vasentha at [48]. This is elaborated on at [49] below.

(b) Second, there is a sudden jump of about one years’ imprisonment, from slightly less than nine years’ imprisonment at the 7.99g weight point to slightly more than ten years’ imprisonment at the 8.01g weight point, even though the severity of the criminal conduct has only increased very slightly (ie, 0.02g from 7.99g to 8.01g of diamorphine). I made this same point in my recent decision of Nurun Novi Saydur Rahman v Public Prosecutor and another appeal [2018] SGHC 236 at [105], where I stated that:

(c) Third, it is inexplicable why as one moves from 6.99g to 7.01g (a higher weight of diamorphine) that the Min Curve and Max Curve should indicate a sentence that is more lenient by one year, when one would logically expect the sentence to increase as the weight of diamorphine trafficked increases by 0.02g from 6.99g to 7.01g. The same problem occurs when one moves from the 8.99g to 9.01g where again the Min Curve and Max Curve indicates a sentence that is more lenient by one year, when one would logically expect the sentence to increase and not decrease when the weight of diamorphine trafficked increases by 0.02g from 8.99g to 9.01g.

29 As can be seen at the top-right corner of the graph (indicated by the dotted rectangle), where the weight of diamorphine is 14.99g (which results in a starting point sentence of 29 years’ imprisonment) and the maximum indicative uplift of two years’ imprisonment is imposed, the resultant sentence of 31 years’ imprisonment actually exceeds the statutory maximum of 30 years’ imprisonment (see [17] above).

30 The gradients of the curves do not increase consistently as the weight of the diamorphine increases. The gradient of the curves (which represents the rate of change of the sentence for each unit change in the weight of the diamorphine) can generally be seen to be gradually increasing as the weight of the diamorphine increases by every one gram. To illustrate:

(a) Between 0g to 2.99g: the sentence increases at a rate of 1/3 year per gram (ie, gradient of 1/3 year per gram).

(b) Between 3g to 4.99g, and 5g to 6.99g: the sentence increases at a rate 1/2 year per gram.

(c) Between 7g to 7.99g: the sentence increases at a rate of one year per gram.

(d) Between 8g to 8.99g: the sentence increases at a rate of three years per gram.

31 However, between the 9g to 9.99g weight points, the gradient inexplicably decreases to a rate of two years per gram. This is wrong in principle because it means that past a certain weight point (ie, 9g), the increase in the sentence of an offender suddenly becomes more lenient for the same increase in the weight of diamorphine.

32 To be clear, I would, in principle, have been able to accept a completely linear gradient (ie, a constant rate of change) throughout the whole weight range from 0g to 9.99g or an increasing gradient (ie, an increasing rate of change) for increments in the weight of diamorphine throughout the whole weight range from 0g to 9.99g. However, what I cannot accept is a gradient which is increasing as the weight of the diamorphine increases but inexplicably decreases after a certain point (ie, past the 9g weight point up to the 9.99g weight point). This is especially so when the indicative sentences for this weight range of between 9g and 9.99g are still far below both (a) the prescribed maximum sentence of 20 years’ imprisonment for the weight range of less than 10g for a first-time offender; and (b) more importantly, the mandatory minimum sentence of 20 years’ imprisonment at the 10g weight point for a first-time offender. The fact that the full spectrum of sentences up to the prescribed maximum sentence of 20 years’ imprisonment is still far from being fully utilised at this high weight range of between 9g and 9.99g, and the presence of the high mandatory minimum sentence of 20 years’ imprisonment at the 10g weight point both have the effect of pulling the sentences for a notional first-time offender within the higher end of the weight range (ie, between 7g and 9.99g) faster upwards as the weight increases and approaches the 10g weight point. This is demonstrated by the steeper gradients as one passes the 7g weight point.

33 In other words, the duty of the court to consider the full spectrum of sentences (see Suventher Shanmugam v Public Prosecutor [2017] 2 SLR 124 (“Suventher”) at [26]), which in this case is up to a maximum of 20 years’ imprisonment for less than 10g of diamorphine, and the fact that the mandatory minimum sentence is 20 years’ imprisonment at the 10g weight point necessarily forces the gradient between the 9g to 9.99g weight range to become steeper as one approaches the 10g weight point. This is to allow more of the sentencing range to be utilised before the weight of the diamorphine crosses into the 10g and beyond category, where the sentence would have to start at 20 years’ imprisonment. Hence it is incongruous that in this higher weight range for a first-time offender between 9g and 9.99g, the gradient for the sentence should suddenly become less steep than the gradient for the sentence in the lower weight range between 8g and 8.99g when the indicative sentence is still far below both (a) the prescribed maximum sentence of 20 years’ imprisonment for this weight range below 10g; and (b) the mandatory minimum sentence of 20 years’ imprisonment at the 10g weight point itself.

34 Therefore, given these errors which I have identified above, I must respectfully decline to adopt a wholesale application of the sentencing framework in Lai Teck Guan to the present case before first correcting these errors.

35 Having said that, I agree with the general principles laid down by Menon CJ in Lai Teck Guan, and will endeavour to adhere to these principles even as I modify the sentencing framework to correct the errors. In particular, I agree that for a repeat trafficking offence, the quantity of the drugs should not be the sole determinant of the sentence to be imposed. The multi-factorial inquiry into the severity of a repeat offence is accounted for in Lai Teck Guan, by taking into consideration the circumstances of the re-offending when determining the indicative uplift to be imposed.

36 I also agree with Menon CJ that a “relatively long sentence is likely to result in compounded severity because it induces a sense of hopelessness that would negate rehabilitative prospects” (at [36]). This principle, coupled with the narrowing sentencing range at the higher ends of the spectrum for an uplift in sentence for a repeat offender, explains why the minimum indicative uplift decreases as the weight of diamorphine increases.

37 With these principles in mind, I proceed to consider what the Lai Teck Guan sentencing framework would look like if these errors were to be corrected.

38 After correcting the errors identified above, while adhering as closely as possible to the principles set out by Menon CJ in Lai Teck Guan, I derived a sentencing framework as depicted by the graph below:

The solid curves represent the original Lai Teck Guan framework, while the dotted curves represent my modified framework. I will address and explain each of the modifications I have made in turn.

39 I begin with the FT Curve for the weight range of under 10g of diamorphine, located at the bottom left corner of the graph. There were two errors with this curve, namely the gap problem and the inconsistency problem. To resolve the gap problem, I first shifted the portion of the FT Curve between 8g and 9.99g down by one year. As a consequence, the weight point of 8g yields only one unique sentence ie, 9 years’ imprisonment, thus resolving the gap problem at the 8g weight point. As for the inconsistency problem, I had to ensure that the gradients along the FT Curve increase (or at least remain constant) as the weight of diamorphine increases. Therefore, I swapped the gradient of the line between 8g to 8.99g (three years per gram) (see [30] above) with the gradient of the line between 9g to 9.99g (two years per gram) (see [31] above), which solves the problem of the gradient of the line between 9g to 9.99g becoming less steep than the gradient between 8g to 8.99g, when it should have been the opposite of that. The dotted line reflecting the revised framework now shows the gradient of the line between 9g to 9.99g being steeper than the gradient of the line between 8g to 8.99g, which is conceptually more logical. Overall, the resultant FT Curve as modified by the dotted lines is more consistent and logical when one analyses the relative increase in sentence at the various weight points within each weight range for each similar increase in the weight of diamorphine trafficked as one moves along the different weight ranges from 0g to 9.99g.

40 I turn next to the Max Curve and the Min Curve for the weight range of under 10g of diamorphine. Similarly, the errors here lie with the gap problem and the inconsistency problem. To resolve the gap problem, I shifted: (a) the portion of both the Min Curve and the Max Curve between 7g and 7.99g up by one year, (b) the sentence of 20 years’ imprisonment for the Max Curve at the 9g weight point down by two years to 18 years’ imprisonment, (c) the sentence of 17 years’ imprisonment for the Min Curve at the 9g weight point down by two years to 15 years’ imprisonment, and (d) the portion of both the Min Curve and the Max Curve between 9g and 9.99g down by one year. In doing so, the inconsistency issue is also naturally resolved given that the gradients along both curves now increase as the weight of the diamorphine increases. This modification also ensures that (a) the difference between the minimum and maximum indicative uplifts at every weight point is kept constant at three years throughout the entire weight range from 0g to 9.99g; (b) the minimum and maximum indicative uplifts at the 9.99g weight point are maintained at three and six years respectively (per Lai Teck Guan at [42]).

41 Finally, I turn to the FT Curve, Min Curve and Max Curve for the weight range of between 10g and 14.99g of diamorphine, located at the upper right side of the graph. I begin with the Max Curve. The main error here relates to the statutory maximum problem. I resolved this by reducing the sentence for a repeat offender with the maximum indicative uplift at the 14.99g weight point (ie, 31 years’ imprisonment) down by one year, which results in a sentence of 30 years’ imprisonment (which is exactly at the statutory maximum). Thereafter, I resolved the gap problem in the Max Curve by reducing the sentences of 28 years and 29 years’ imprisonment at the 11.5g and 13g weight points respectively down by one year each. The inconsistency problem does not arise because the resultant gradients are increasing from the rate of 2/3 year per gram between the 10g and 12.99g weight range, to the rate of one year per gram between the 13g and 14.99g weight range.

42 In order to calibrate the Min Curve, I ensured that the original differences between the Max Curve and Min Curve are maintained where possible, which correspond to the differences between the maximum and minimum indicative uplifts as set out in Lai Teck Guan:

(a) At the 14.99g weight point along the Min Curve: The original difference between the maximum and minimum indicative uplifts was one year. In order to maintain that difference, I reduced the sentence of 30 years’ imprisonment at the 14.99g weight point along the Min Curve down by one year to 29 years’ imprisonment.

(b) At the 12.99g weight point along the Min Curve: The original difference between the maximum and minimum indicative uplifts was two years. In order to maintain that difference, I reduced the sentence of 27 years’ imprisonment at the 12.99g weight point along the Min Curve down by one year to 26 years’ imprisonment.

(c) At the 11.49g weight point along the Min Curve: The original difference between the maximum and minimum indicative uplifts was three years. In order to maintain that difference, I reduced the sentence of 25 years’ imprisonment at the 11.49g weight point along the Min Curve down by one year to 24 years’ imprisonment.

(d) At the 10g weight point along the Min Curve: The original difference between the maximum and minimum indicative uplifts was three years. There was no need to alter the sentence at the 10g weight point along the Min Curve given that the difference is maintained.

There is no need to further manipulate the Min Curve given that the inconsistency problem does not arise.

43 Finally, in calibrating the FT Curve for the weight range of between 10g and 14.99g of diamorphine, I adopted a similar approach and ensured that the original differences between the FT Curve and the Min Curve where possible are maintained to reflect the minimum indicative uplifts as stated in Lai Teck Guan:

(a) At the 14.99g weight point along the FT Curve: The original minimum indicative uplift was one year. In order to maintain that original minimum indicative uplift of one year, I reduced the sentence of 29 years’ imprisonment at the 14.99g weight point along the FT Curve down by one year to 28 years’ imprisonment.

(b) At the 12.99g weight point along the FT Curve: The original minimum indicative uplift was two years. In order to maintain that original minimum indicative uplift of two years, I reduced the sentence of 25 years’ imprisonment at the 12.99g weight point along the FT Curve down by one year to 24 years’ imprisonment. Whereas at the 13.01g weight point along the FT Curve, the sentence of 26 years’ imprisonment is reduced by two years to 24 years’ imprisonment to resolve the gap problem.

(c)  At the 11.49g weight point along the FT Curve: The original minimum indicative uplift was four years. In order to maintain that minimum indicative uplift of four years, I reduced the sentence of 22 years’ imprisonment at the 11.49g weight point along the FT Curve down by one year to 21 years’ imprisonment. Whereas at the 11.51g weight point along the FT Curve, the sentence of 23 years’ imprisonment is reduced by two years to 21 years’ imprisonment to resolve the gap problem.

(d) At the 10g weight point along the FT Curve: The original minimum indicative uplift was four years. There is no need to alter the sentence at the 10g weight point along the FT Curve given that the minimum indicative uplift of four years is maintained.

There is no need to further manipulate the FT Curve given that the inconsistency problem does not arise.

44 Based on the modified graph at [38], I set out the revised sentences for a notional first-time offender, a repeat offender with minimum indicative uplift, and repeat offender with maximum indicative uplift at various weight points:

45 I highlight that this table only indicates the sentences at certain specific weight points. Therefore, if a particular offence involves a weight of diamorphine that falls in between these weight points, a sentencing judge applying this framework should interpolate linearly from the figures set out above to obtain the appropriate indicative sentence. Alternatively, the sentencing judge could obtain the appropriate indicative sentence by referring to and simply reading off from the graphs at [38] above as modified by the various dotted lines.

46 It is not strictly necessary for me to consider the issue of caning, given that the appellant is a woman and is therefore ineligible for caning. Be that as it may, I will, for completeness, apply the same graphical analysis I have adopted above to the benchmark sentences for caning set out in Lai Teck Guan. Plotting out the data points for the weight of diamorphine vis-à-vis the number of strokes of the cane as set out in [42] of Lai Teck Guan on two axes, I derived the following graph:

47 The key features of this graph are as follows:

(a) The weight of the diamorphine (in grams) is represented on the horizontal axis (x-axis), and the corresponding sentence (in terms of the number of strokes of the cane) is represented on the vertical axis (y-axis).

(b) For each weight range as predetermined in Lai Teck Guan (ie, 0g to 3g; 3g to 5g; 5g to 7g …), there are two possible starting point caning sentences if an offender were to fall within that weight range (eg, at 1.5g, the court could choose to impose either 5 strokes or 6 strokes of the cane). Accordingly, I have drawn two parallel lines to demonstrate the two possible starting points for the caning sentence at all the various weight ranges for a first-time offender. The bottom-most line (represented by the solid line) indicates the minimum number of strokes of the cane which could be imposed on a notional first-time offender at a particular weight point (the “First Time Minimum Caning Line” or “FT Min Caning Line” in short). The higher line (in long dashed and dotted lines) just above the FT Min Caning Line at each weight point indicates the maximum number of strokes of the cane which could be imposed on a notional first-time offender at a particular weight point (the “First Time Maximum Caning Line” or “FT Max Caning Line” in short).

(c) The middle line (ie, the line immediately above the FT Max Caning Line) indicates the number of strokes of the cane which would be imposed on a repeat offender if the minimum starting point (ie, based on the minimum number of strokes of the cane for a notional first-time offender) and the minimum uplift for a repeat offender were imposed (the “Minimum Indicative Uplift Caning Line” or “Min Caning Line” in short). To illustrate, consider the weight range from 0g to 3g. At this weight range, Lai Teck Guan provides for two possible starting point sentences for caning based on what a notional first-time offender would receive ie, five or six strokes of the cane. The Min Caning Line represents the number of strokes of the cane to be imposed if the minimum starting point ie, five strokes and the minimum uplift ie, five strokes were imposed, for a total of ten strokes.

(d) The upper-most line indicates the number of strokes of the cane which would be imposed on a repeat offender if the maximum starting point (ie, based on the maximum number of strokes of the cane for a notional first-time offender) and the maximum uplift required for a repeat offender were imposed (the “Maximum Indicative Uplift Caning Line” or “Max Caning Line” in short). To illustrate, consider the weight range from 0g to 3g. At this weight range, Lai Teck Guan provides for two possible starting point sentences based on what a notional first-time offender would receive ie, five or six strokes of the cane. The Max Caning Curve represents the number of strokes of the cane to be imposed if the maximum starting point ie, six strokes and the maximum uplift ie, six strokes were imposed, for a total of 12 strokes.

(e) It would be open to a sentencing judge to choose the minimum starting point for a notional first-time offender and impose the maximum uplift and vice versa, which would result in the sentences as represented by the points in between the Max Caning Line and the Min Caning Line.

(f) At the 10g weight point and above, the number of strokes of the cane converge at the mandatory 15 strokes of the cane.

(g) The key difference between this graph and the graph for the imprisonment terms at [24] above is that there can be no assumption of linearity in this case. This is because the number of strokes of the cane are necessarily discrete and must be represented in whole numbers. Therefore, even though Lai Teck Guan states that for the weight range of between 0g to 3g, the number of strokes of the cane for a notional first-time offender is “5 – 6 strokes”, it can only actually mean five or six strokes and nothing in between.

48 Having analysed the graph at [46] above, there are two errors which are apparent to me. First, the gap problem arises at multiple points which makes it difficult to determine the appropriate sentence to impose at certain weight points. For example, at the 5g weight point, if a sentencing judge wanted to impose the minimum starting point with the minimum uplift, he would still be confronted with two possible choices ie, ten or 11 strokes. This could lead to inconsistencies in the application of the framework as I have discussed at [28(a)] above.

49 Second, and somewhat related to the first problem, is that there are two possible indicative starting point sentences for caning available to the sentencing judge at any given weight point. This is problematic because there could potentially be inconsistencies in the indicative starting point sentence adopted by different sentencing judges, even though the weight of the diamorphine may be exactly the same. In Vasentha at [47] and [48], Menon CJ set out the indicative starting points for the sentences to be imposed on first-time offenders trafficking in diamorphine in tabular form and went on to state that:

The passage cited above demonstrates that the indicative starting points as set out in Vasentha (which were subsequently adopted in Lai Teck Guan to determine the sentence for a notional first-time offender) are based solely on the weight of the diamorphine. Therefore, at this stage of the inquiry, the sentencing judge does not exercise any discretion because his choice of an indicative starting point should be determined solely by the weight of the diamorphine. The sentencing judge’s discretion, if any, only comes in at the next stage of the inquiry where he can adjust the indicative starting point sentence to reflect the offender’s culpability and the presence of aggravating or mitigating factors.

50 If the indicative starting point sentence is supposed to be determined solely by the weight of the diamorphine, then it logically follows that the indicative starting point sentence adopted by different sentencing judges should be the same so long as the weight of the diamorphine is the same. However, with the existing framework where there are two possible caning sentences at any given weight point eg, five or six strokes at the 2g weight point, it is possible that different sentencing judges may adopt different indicative starting points even though the weight of the diamorphine is the same. Therefore, it would, in my view, be more useful if the framework were modified such that each weight point only has one possible corresponding starting point sentence for caning. The sentencing judge can then exercise his discretion thereafter in determining the uplift for the caning, and calibrating the sentence further based on any aggravating or mitigating factors.

51 After correcting the two errors which I have identified above, I derived a sentencing framework as depicted by the following graph:

52 To solve the first error identified at [48] above in relation to the gap problem, I adjusted the weight ranges such that each weight point would have a unique corresponding sentence for caning. In the original framework, there were overlaps in the weight ranges, eg, for the weight ranges of 0g to 3g and 3g to 5g, there is an overlap at the 3g weight point. This created the gap problem because there would be one set of sentences for the 0g to 3g weight range and another set of sentences for the 3g to 5g weight range, therefore at the 3g weight point there would be two possible sentences for caning. Therefore, by redefining the weight ranges to remove the overlapping weight points, I ensured that each weight point would have a unique sentence. To illustrate this, I changed the 0g to 3g weight range to 0g to 2.99g instead. Therefore, that weight range no longer overlapped with the next weight range of 3g to 4.99g.

53 To solve the second error as identified at [49] and [50], I removed one of the possible sentencing options such that there would only be one indicative starting point at each weight range ie, by removing the dashed and dotted line of the FT Max Caning Line. Comparing the original graph based on the Lai Teck Guan framework for caning to the modified graph, it is apparent that there is now only one line representing the indicative starting point sentences for caning at each weight range, as opposed to the two parallel lines in the original version.

54 I did not consider it necessary to make any further adjustments to the Max Caning Line and the Min Caning Line. Therefore, the maximum uplifted sentence and the minimum uplifted sentence as set out in Lai Teck Guan remains the same. What I have done is to remove the need for the sentencing judge to have to choose between multiple indicative starting point sentences at the first stage of the inquiry, since he is to determine the indicative starting point based solely on the weight of the diamorphine. The sentencing judge exercises his discretion at the second stage of the inquiry, where he determines the appropriate uplift to impose. In my view, this coheres more with the framework envisioned in Vasentha and Lai Teck Guan, as the sentencing judge only exercises his discretion at appropriate stages in the inquiry.

55 Based on the modified graph above, I set out the revised caning sentences for a notional first-time offender, a repeat offender with minimum indicative uplift, and repeat offender with maximum indicative uplift at the various weight ranges in tabular form:

56 At first blush, it may seem as though the sentencing framework which I have formulated above causes the sentencing exercise to become overly rigid and mechanistic. However, I emphasise that the sentencing judge still retains the final discretion in first determining the amount of uplift to impose, and second in calibrating the final sentence based on other aggravating and mitigating factors. This is because the framework set out here only applies to the identification of an indicative starting sentence. In my view, this strikes a proper balance between having consistency in sentencing on one hand, and allowing for sentences to be tailored according to the unique facts of each case on the other.

57 In Dinesh Singh Bhatia s/o Amerjeet Singh v Public Prosecutor [2005] 3 SLR(R) 1, V K Rajah J (as he then was) stated at [24] that “[s]entencing is neither a science nor an administrative exercise. Sentences cannot be determined with mathematical certainty. Nor should they be arbitrary.” Be that as it may, I am of the view that where the court does come up with a sentencing framework, it should endeavour to ensure that the framework is at the very least coherent and capable of yielding consistent results. Even if the sentencing framework or benchmark is not meant to determine the final sentence to be imposed, coherence and consistency is necessary to ensure that the framework can be logically and usefully applied. This is especially so when sentencing frameworks, such as those in Lai Teck Guan and Vasentha, determine a starting point sentence based on a single metric such as weight of the drug. In such cases, the application of the sentencing framework should give a consistent indicative starting point that corresponds exactly to the weight of the drug. Otherwise, it would defeat the purpose of having sentencing frameworks or benchmarks which are meant to guide sentencing judges in providing an indicative starting point.

58 I emphasise that the modifications which I have made above are not intended to displace the principles underlying the framework in Lai Teck Guan. Rather, I merely hope to resolve some of the errors which may hinder the practical application of the framework.

59 Before I proceed to apply this revised framework to the facts of the present case, I pause momentarily to explain why there is a sudden jump in imprisonment terms from the 9.99g weight point to the 10g weight point (see the graph at [38] above). I stated at [28] above that there should be no unexplainable jumps or gaps in sentence when the severity of the criminal conduct has only increased very slightly. Despite this, there is a sudden increase of 6 years’ imprisonment from 14 years to 20 years’ imprisonment when the weight point increases from 9.99g to 10g of diamorphine for a first-time notional offender (see the dotted part of the FT Curve at the 9.99g and 10g weight points).

60 The sudden increase or jump in this specific instance can be attributed to parliament providing for a mandatory minimum sentence of 20 years’ imprisonment for an offender trafficking in 10g and above of diamorphine. Therefore, even though the offender’s criminal conduct has only increased very slightly, the statutorily imposed mandatory minimum imprisonment term of 20 years for trafficking in not less than 10g causes the sentence imposed to increase substantially at the 10g weight point itself. I term this phenomenon “legal bad luck”, in the sense that it is the offender’s legal “bad luck” if he or she crosses into the 10g weight point and beyond because his sentence at the starting point is bound to jump at least to the mandatory minimum level, which is exactly what has been depicted on the FT Curve at the 10g weight point.

61 I turn now to apply the revised framework to the facts.

62 The appellant was found to possess 9.98g of diamorphine for the purposes of trafficking. I agree with the Prosecution that 9.98g is extremely close to the upper limit of the weight range, and accordingly the appellant should receive a sentence at the uppermost end of the range for indicative starting sentences. Therefore, based on my revised framework, the appellant would receive a sentence of 14 years’ imprisonment if she was a notional first-time offender.

63 The next step is then to determine the appropriate indicative uplift to apply to a repeat offender. Applying the indicative uplift as set out at [44] above, the appellant should receive an indicative starting sentence of between 17 and 20 years’ imprisonment. In order to determine the appropriate uplift to impose, the court must have “due regard to the circumstances of the repeat offence” (Lai Teck Guan at [38(c)]).

64 In Lai Teck Guan, the offender was first convicted of drug-related offences in November 2000, for which he was ordered to undergo seven years of corrective training. In February 2009, he was again convicted of multiple drug-related offences, including one charge of possessing diamorphine for the purpose of trafficking. The offender was sentenced to six years’ imprisonment and six strokes of the cane for that charge. Soon after the offender was released from prison, he was convicted of an offence of consuming morphine and was then placed under drug supervision for 24 months starting October 2014. After his drug supervision ended, he then committed the offences which were the subject of the charges he faced in that case, which included trafficking in 7.75g of diamorphine (Lai Teck Guan at [52]).

65 Menon CJ observed, at [53], that the offender was a recalcitrant offender who had not been rehabilitated despite the community order, nor was he deterred by his stint in prison. Menon CJ further stated that the offender had not taken advantage of the fact that he was sentenced to only slightly more than the mandatory minimum for his previous offence and that he was placed under drug supervision. With these considerations in mind, Menon CJ decided that the appropriate uplift would be six years’ imprisonment and three strokes of the cane. The framework which Menon CJ had applied provided for an indicative uplift of between four to seven years’ imprisonment at the 7g to 8g weight range. Therefore, the uplift of six years’ imprisonment was on the higher end of that range.

66 The circumstances of the appellant’s reoffending in the present case can hardly be said to be as severe as the offender’s in Lai Teck Guan. The appellant was previously convicted of drug-related offences in July 2013, which included having in her possession diamorphine for the purpose of trafficking. Given that her convictions for those offences were just two months shy of her 21st birthday, she was ordered to undergo reformative training, as opposed to being sentenced to the mandatory minimum of five years’ imprisonment for a first-time offence of trafficking in under 10g of diamorphine. The appellant was subsequently released from the RTC on 4 September 2015. She was then arrested for her current offences on 4 July 2016, ten months’ after her release from the RTC.

67 While I agree with the Prosecution that the appellant had re-offended soon after she was released from RTC, I do not find her to be as recalcitrant as the offender in Lai Teck Guan. The offender in Lai Teck Guan had committed drug-related offences over a period of six years, as opposed to just three years in the appellant’s case. Further, he had re-offended despite having served a term of imprisonment. The appellant on the other hand has never been sentenced to imprisonment before, having only served time in the RTC for her prior offence of trafficking in 0.33 grams of diamorphine. Therefore, I am of the view that the indicative uplift for the appellant should be on the lower end of the range.

68 In the circumstances, I would impose the minimum uplift in sentence on the appellant, resulting in an indicative starting sentence of 17 years’ imprisonment. The final step is for me to consider whether there are any other aggravating or mitigating factors which warrant further adjustments to this indicative starting sentence.

69 In terms of mitigating factors, counsel for the appellant submits that the appellant is still relatively young, and an overly long prison sentence would adversely affect her chances of re-integrating back into society upon her release. ¹³ I agree that the appellant is indeed young and has her whole adult life ahead of her. There is a very real concern that she will only be released from prison at an age where it will be difficult for her to find sustained employment. Under such difficult circumstances, she may once again be driven back to a life of crime in order to make ends meet. The appellant contends that rehabilitation should be the dominant sentencing consideration, while the Prosecution contends that it should be deterrence. In my view, regardless of whether it is via rehabilitation or deterrence, the ultimate goal of sentencing is the prevention of crime. Therefore, I am not inclined towards imposing an overly long imprisonment sentence if to do so would have the contrary effect of increasing the risk of the appellant re-offending.

70 Be that as it may, the Prosecution submits that an aggravating factor in the present case is that the appellant had multiple trafficking charges taken into consideration for sentencing. ¹⁴ Indeed, the Court of Appeal in Public Prosecutor v UI [2008] 4 SLR(R) stated at [38] that where the TIC offences and the offences proceeded with are similar in nature, the sentence which the court would otherwise have imposed for the offences proceeded with would be increased. I therefore accept that the appellant’s multiple TIC charges are an aggravating factor and have to be accounted for in the final calibration of the sentence.

71 Balancing the mitigating and aggravating factors that I have considered above, I further reduce the appellant’s sentence by one year and six months’ imprisonment. I therefore set aside the sentence of 21 years’ imprisonment imposed by the Judge for the trafficking charge and impose a sentence of 15 years and six months’ imprisonment in its place.

72 I must emphasise that despite the reduction in sentence, the appellant still faces a significant and substantial term of imprisonment. In my view, this sentence provides enough of a deterrent effect without being overly crushing. To borrow the words of the appellant’s counsel: the appellant has made her bed and is now sleeping on it, but the length of her sentence should allow her enough time to get up and see the dawn of a new day, rather than for her to get up just in time to see the sun set on her. ¹⁵

73 For these reasons, I make the following orders:

(a) On the trafficking charge, I set aside the sentence of 21 years’ imprisonment. In its place, I impose a sentence of 15 years and 6 months’ imprisonment.

(b) On the enhanced consumption charge, the possession charge, and the utensils charge, I affirm the sentences imposed by the Judge.

74 The sentences for the trafficking charge and the utensils charge should run consecutively as the Judge ordered. The remaining sentences are to run concurrently. The resulting aggregate sentence is 15 years’ and nine months’ imprisonment, and it is to be backdated to 5 July 2016, the date on which the appellant was remanded.

75 As a final point, counsel for the appellant had informed the court that if an offender were to be sentenced to more than 20 years’ imprisonment, that offender would not be eligible to enrol in the education programmes available in the prisons. DPP Winston Cheng replied saying that to his knowledge, there is no absolute bar and prisons will decide each application on its merits. I directed the Prosecution to confirm this position and to inform counsel accordingly. Indeed, the appellant will be facing a good portion of her early adulthood in prison, which is often the formative years where people pick up the skills they will need to maintain their livelihood. It is imperative that she should spend her time in prison productively, in the hope that when she is eventually released she will be able to quickly re-integrate herself into society, be gainfully employed and be a good mother to her children.