CASES

 

Duress by indirect circumstances in english and south african law: a comparison - R v Brandford [2017] 2 All ER 43; [2016] EWCA Crim 1794

 

 

1 Introduction

While duress by means of direct threats can provide a defence in criminal law, the legal question is whether threats conveyed indirectly are capable of providing a valid defence in criminal law. More specifically, can indirect threats then also be used as a means of defending another party's interests that are under attack? There appears to be both academic support and precedent to answer this question in the affirmative (S v Pretorius 1975 2 SA 85 (SWA); Burchell Principles of Criminal Law (2016) 279). In the Pretorius case, the court made it clear that while mens rea is not relevant to the enquiry, the defence must be "confined within the strictest and narrowest limits because of the danger attendant upon allowing a plea of necessity to excuse criminal acts" (289). These limits include that the threat must have been imminent and, more specifically, it must have been necessary for the accused to avert the danger by any reasonable means (285). This is the legal question that arose in R v Brandford ([2017] 2 All ER 43; [2016] EWCA Crim 1794), and which will be examined in the light of a comparison between English law and South African law in relation to the defence of duress and necessity respectively.

 

2 Facts

The accused and her boyfriend were arrested by the police in the course of a police operation that involved the supply and distribution of Class A drugs in the London Borough of Lewisham. A number of drug runners, acting as street dealers, were charged and tried alongside each other (par 6), including the accused's boyfriend, Alford, and one Karemera. The accused was charged with concealing drugs in her vagina. There were 121 packages. Seven consisted of wraps of crack cocaine and 44 wraps contained heroin with an estimated street value of between £1,500 and £2,300 (par 8). It was alleged by the accused that she had only become involved in the conspiracy on the night before her arrest on 26 August 2014. She had agreed to carry drugs for her boyfriend, so assisting the conspiracy (par 9). Her defence was that Alford had approached her for assistance on the basis that he had inherited a debt from a former friend, "Allman", who had been murdered. He alleged that his life would be in danger if the drugs were not distributed (par 10).

On the basis of these facts, a criminal trial resulted in the conviction of each accused on two counts (count 1, concerning cocaine, and count 2, concerning heroin) of supplying controlled drugs in contravention of s 1 of the Criminal Law Act 1977 (par 2). Following the majority verdict of the court on these two counts, Brandford was sentenced to 28 months' imprisonment on count 1 and 28 months' on count 2, which sentences were to run concurrently (par 3).

In the judge's summing up, the judge noted that the appellant had not been physically compelled to secrete the drugs, but had done so at the "urgent request" of her boyfriend (par 20). The judge therefore withdrew the defence of duress from the jury (par 26). The judge was of the view that Brandford's belief in a threat to kill Alford was not reasonable, and nor would a reasonable person of ordinary firmness view it as such, in the absence of immediate conclusive proof that the threat would also be carried out (par 24). This was because she had no first-hand knowledge of the threats (also called "hearsay duress")(par 23). This was so because Alford made use of coded quotes, such as "it would not be nice for me" (par 24). Furthermore, both Alford and Karemera had testified that a loss of drugs would simply result in an extended period of drug dealing (par 24). The appellant had argued unsuccessfully that the judge was incorrect in withdrawing the defence from the jury because the threats had not been conveyed directly to the appellant (par 24).

The grounds of appeal thus related to the judge's treatment of hearsay evidence - that is, the question whether threats always had to be conveyed directly for a successful reliance on a defence of duress (par 23).

 

3 Judgment

In the Court of Appeal, the crux of the appellant's argument concerned the judge's decision to withdraw the defence of duress from the jury (par 1). More specifically, the argument centered on the judge's treatment of "hearsay duress", which, was rejected since there was nothing precluding the use of "hearsay duress" (par 23). The appellant contended that there was no "basic irreconcilability" between the pressure created by a relationship, and fear, which forms the basis of duress (par 23). The court noted that the former term makes use of affection whereas the latter is based on fear. The distinction between these two terms was important since the former term involves pressure bought to bear by one party against another in order to manipulate that person without serious threat of death or injury, and is based on affection shared between the parties.

This distinction between pressure and fear is also noteworthy since pressure will not establish a defence of duress (par 24). The distinction is also important since the English legal system "leans heavily" against use of hearsay evidence, especially where the threat has not been directly conveyed (par 25). The defence counsel argued that the evidence should have been placed before the jury. Defence counsel contended that Brandford was of good character and had in fact only appreciated the severity of the situation the night before the crime was committed and therefore had not voluntarily associated herself with any criminal activity. This, the defence contended, should have been left within the purview of the jury to determine (par 27). However, on analysis of the accepted principles of duress, the court would eventually confirm that that such a defence would not have been accepted by the jury (par 47) and therefore dismissed the ground of appeal and the appeal against Brandford's conviction as a whole (par 51).

The court proceeded to examine whether duress can indeed be regarded as a defence where the duress in question does not include direct threats that lead to the criminal conduct; it came to the conclusion that while it may not necessarily be a fatal bar to a defence, the manner in which the threat is conveyed is but one of the circumstances that the court will take into consideration. Hearsay evidence in cases of duress can be used to demonstrate a defendant's state of mind and that he or she had "good reason to fear death or personal injury" (par 28; see Subramaniam v Public Prosecutor [1956] 1 WLR 965 970).

Another reason that duress should not be available as a defence is based on policy grounds since such a defence would frustrate the legitimate aims of government in controlling the A-class drug trade (par 25). The court proceeded to examine several pertinent decisions in this regard to determine whether defence counsel was correct in its submission that a defence of duress could equally be applicable in the context of threats that were made indirectly. Defence counsel noted there was no authority that precluded the applicability of hearsay duress (par 27).

On the facts of the case, the Court of Appeal noted that while fear (which forms the foundation of duress) and pressure (that emanates from an intimate relationship and which exploits a person's affection) are different, the trial court was incorrect in concluding that they are irreconcilable (par 40). They can operate in a cumulative manner (par 40). In addition, whether or not a defence of duress can be founded rests on whether the pressure based on exploitation of relationship is accompanied by threat of death or serious bodily injury (par 40).

The court confirmed the conviction but allowed the appeal against the sentence, deeming the sentence passed to have been manifestly excessive despite the fact that the accused played a significant role in a category 3 offence as a willing courier of 121 wraps of Class A drugs, with knowledge they would be sold in Portsmouth (par 53). The court substituted her sentence with 21 months' imprisonment (par 56).

As to the question whether the judge should have withdrawn the defence from the jury, the court noted that judges always have a discretion to exercise a "robust and reasoned approach" when it comes to "fanciful" defences such as duress (R v Brandford supra par 44; see also R v Hammond [2013] EWCA Crim 2709 14). Although in this case the trial court judge's "imperfect" reasoning had led to the exclusion of the evidence, it essentially made no difference to the end result: there was no immediacy of threat nor an inability to take evasive action (Laird "Case and Comments:

R v Brandford (Olivia)" 2017 Criminal Law Review 554 556; see also R v Hasan) [2005] 4 All ER 685 28). This was because there was no immediate threat to Alford since the threat had been conveyed the night before they left for London. Furthermore, Brandford had knowingly participated by buying latex gloves and had proceeded to watch him parcel the drugs as well as continued to carry drugs on her person, long after Alford had disposed of his (par 46). This was clearly an example of voluntary association where the jury would have no choice but to convict (par 46).

 

4 The nature of the defence of duress in English law

Brandford raised the question of whether duress can successfully be

invoked as a defence (R v Hasan [2005] UKHL 22, [2005] 2 AC 467) in the case of drug trafficking or drug dealing (Storey "Duress by Indirect Threats" 2017 Journal of Criminal Law 91 94. In this respect, see R v Aikens [2003] EWCA Crim 1573; R v McDonald [2003] EWCA Crim 1170, where the defence was raised, albeit not successfully). The central question is therefore is not whether it can be a defence, but rather whether the defendant can satisfy all the grounds in order for the defence to be successful. For instance it would be beneficial if it could be shown that there is no voluntary association with criminals. (Storey 2017 Journal of Criminal Law 91 94).

The defence of duress by threats can be characterised as necessarily involving a choice of unsavoury alternatives:

"Thus, although the defendant must honestly believe that force is necessary, this belief is not required to be reasonable, as opposed to the requirement that the response be reasonable." (Freer "Driving Force: Self-Defence and Dangerous Driving" 2018 Cambridge Law Journal 9 10)

This choice of alternatives plays a key role since it has the ability to exculpate the defendant in a particular case. However, policy considerations demand that such a defence be narrowly circumscribed Therefore, the defence is only available if the two-prong test is satisfied:

"Would the defendant have been impelled to act as they did because, as a result of what they reasonably believed the threatener had said or done, they had good cause to fear that if they did not so act the threatener would kill or seriously injure them?" (R v Hasan [2005] UKHL 22; [2005] 2 AC 467)

Notably, the above quote suggests that only a limited category of threats could qualify as a defence - that is, if there were threats of death or serious bodily harm (R v Hasan [2005] UKHL 22, [2005] 2 AC 467 par 21) or a threat directed to a member of the defendant's immediate family or a person for whose safety the defendant would reasonably regard themselves as being responsible (R v Brandford supra par 32). In instances where the second stage of the test is reached - that is, where death or serious bodily injury is a likelihood - then the matter was one left for jury determination (see R v Lynness [2002] EWCA Crim 1759 24-25). Thus the question of whether the appellant held a reasonable belief in death or serious bodily harm from Alford becomes a crucial question that needs to be determined (Ashworth and Horder Principles of Criminal Law 7ed (2013) 206), and this is where the primary source of criticism in this case lies. Where no circumstances existed where the defence could be found, then such a defence has to be withdrawn from the jury (R v Bianco [2001] EWCA Crim 2516). Circumstances where such a defence would be withdrawn, for instance, can be found where there was no immediacy of threat, or through the doctrine of prior fault - that is, where a person had voluntarily joined a criminal enterprise (R v Ali [1995] Crim LR 303; R v Heath [2000] Crim LR 109; R v Harmer [2001] EWCA Crim 2930; R v Ali [2008] EWCA Crim 716; R v Hussain [2008] EWCA Crim 1117; see also Percival "Cases in Brief: Brandford [2016] EWCA 1794; December 2, 2016" 2017 1 Archbold Review 2). In Hasan, the court noted:

"[N]othing should turn on foresight of the manner in which, in the event, the dominant party chooses to exploit the defendant's subservience. There need not be foresight of coercion to commit crimes." (R v Hassan supra par 37)

A line of authorities on duress by threats seems to suggest that, while indirect threats can in principle be relied upon (see R v Hudson, R v Taylor [1971] 2 All ER 244, [1971] 2 QB 202), the courts follow a direct approach -that is, the more directly the threat is conveyed, the more likely it will be capable of establishing defence (R v Brandford supra par 39; see also Northern Ireland v Lynch [1975] 1 All ER 913 932, [1975] AC 653 687). However, central to withdrawing the defence from a jury was the view that the crux of the issue is not the manner in which the threat is transmitted, but rather whether the threat was immediate, imminent and whether it was made with sufficient potency to have influenced the accused (Laird 2017 Criminal Law Review 554 556). In other words, the manner in which the threat is relayed is but one fact that the courts take into consideration and does not widen the scope of the defence (Laird 2017 Criminal Law Review 556).

One aspect that was particularly noteworthy about this judgment was the trial court judge's discussion of the context in which indirect threats could be made. The judge was of the view that there was a clear distinction between a person whose free will was overwhelmed as a result of fear as opposed to the case in question where the accused was merely pressurised to act as a result of the romantic relationship that she shared with Alford (R v Brandford supra 23).

The Court of Appeal rejected the trial court's position, noting that the two concepts are different and could in fact operate in a cumulative manner (Laird 2017 Criminal Law Review 557). This is because the pressure based on a relationship exploits infatuation or affection, whereas the second concept (fear, which lies at the heart of duress) is based on fear (R v Brandford supra par 40). Laird points out that this raises two questions: (1) do compulsion and pressure that arise in the context of a certain type of relationship give rise to a new form of defence? (Laird 2017 Criminal Law Review 557); and (2) was there any immediacy of threat to constitute a complete defence? (par 33).

What is noteworthy about this "new" defence is that, like duress, it is also predicated on the principle of compulsion (Laird "Evaluating the Relationship Between Section 45 of the Modern Day Slavery Act 2015 and the Defence of Duress: An Opportunity Missed?" 2016 6 Criminal Law Review 395 398). Furthermore, compulsion must be ascribed to some form of "relevant exploitation" (Laird 2016 Criminal Law Review 395 398). Section 76 of the Serious Crimes Act 2015 provides for the new offence of controlling and coercive behaviour, which is limited to intimate or family relationships and which behaviour is capable of causing harm and special vulnerability to victims in these settings (Edwards "Coercion and Compulsion Re-Imagining Crimes and Defences" 2016 Criminal Law Review 876 877).

The implications of accepting such a defence are radical and problematic at best. First, since compulsion is not defined, it will have to be interpreted broadly. This means that since compulsion is subjectively tested, no evidence of threats or outward action is necessary (Laird 2016 Criminal Law Review 398). Therefore, viewed from the defendant's perspective, she could not have helped but act as she did. Such an approach does not accommodate the restrictive nature of the rest of the elements of the defence (Laird 2016 Criminal Law Review 398). Secondly, what criteria are envisaged in relation to the causation element? It appears that the strict requirement for causation - namely, the "but for" test - is relaxed and a lower criterion would suffice (Laird 2016 Criminal Law Review 398).

Thirdly, the common-law approach to duress, which is expressed in the case of Valderrama-Vega ([1985] Crim. L.R. 220), has raised the problem of the cumulative effect of the pressure under which the defendant operated. In that instance, the Court of Appeal, despite upholding the defendant's conviction, noted that juries should not be directed on the basis of whether the defendant acted solely as a result of threats of death or grievous bodily harm (Laird 2016 Criminal Law Review 397). As Laird notes, this element requires a strict interpretation and if the defendant would have committed the crime irrespective of any "relevant exploitation" or cumulative impact of the relationship, then Brandford would not necessarily have been able to rely on the defence (Laird 2016 Criminal Law Review 397). Not only does such an approach require "mental gymnastics" in determining if defendants would have acted as they did, but, further, it has the potential to greatly narrow the ambit of the defence (Laird 2016 Criminal Law Review 397). For instance, determining the meaning of "compulsion", or more specifically "coercion and controlling" behaviour, is dependent on "case specific factual context" (Edwards 2016 Criminal Law Review 878). Furthermore, practical problems arise in relation to proof concerning what degree of coercion is required and the type of evidence necessary to demonstrate that the defendant was compelled (Edwards 2016 Criminal Law Review 878).

What appears to be clear is that Brandford did not appreciate the true nature of the threat to Alford until the night before the events of 27 August, and any notion of voluntary association could be discounted (R v Brandford supra 27). However, in this instance, the jury was not given an opportunity to canvass this defence and therefore their jurisdiction was usurped in respect of this matter (R v Brandford supra 27). Despite flawed reasoning concerning the basic irreconcilability between fear and pressure based on a relationship, the judge was entitled to withdraw the defence. Therefore, an exploitation of a relationship without a "relevant threat of death or serious injury of sufficient potency, cannot found duress" (R v Brandford supra 40).

In this case, it was clear from the evidence that there was no immediacy of threat. This was demonstrated by the following factors: first, the vagueness of the threats made as well as the absence of the identity of the perpetrator making the threats (R v Brandford supra par 46); secondly, the absence of an immediate threat on the night in question - that is, she was able to purchase latex gloves and other items freely, and was able to contemplate the option of contacting either her father for assistance to pay off those threatening Alford or even the police; thirdly, the existence of the opportunity of escaping from the threat by disposal of the drug (Storey 2017 Journal of Criminal Law 93; see also R v Pommell ([1995] 2 Cr App R 607), despite which, Brandford insisted that Alford continue with the course of action; and finally, Brandford's reaction to the threats in this instance, and threats made on previous occasions such as the January pepper spray incident, the murder of Allman and the June stabbing, which did not correspond with the conduct of an individual who was fearful for Alford's life (R v Brandford supra par 46). These factors demonstrate the point that while indirect threats may form the basis of a defence of duress, in practice, they may provide a reason for withdrawing the defence from the jury (Laird 2017 Criminal Law Review 556).

In relation to the contention that indirect threats can form the basis of new defence as set out in section 45 of the Modern Slavery Act 2015, it has been held that they are insufficient to form the basis of duress on the basis of the above discussion (Laird 2017 Criminal Law Review 557). Nowhere is this more clearly demonstrated than in the traditional test used to assess duress as expounded in R v Graham ([1982] 1 All ER 891, and followed in the subsequent cases of R v Howe [1987] 1 All ER 771; [1987] AC 417); R v Hasan [2005] 2 WLR). In R v Brandford (supra), the court again highlighted the accepted objective test for duress:

"[Would] the sober person of reasonable firmness, sharing the characteristics of the defendant, ... not have responded to whatever he reasonably believed [the threatener] said or did by taking part [in the offence]?" (par 31)

The first leg of the test is subjective in nature - that is, did the defendant entertain an honest belief, as opposed to a reasonable belief, that their life was in danger? (James "Duress: Objective Test" 2007 Journal of Criminal Law 193 194). This means that the reasonable person would share the same characteristics, including psychiatric impairments that would not make them more vulnerable or timid but in fact genuinely more susceptible to threats (James 2007 Journal of Criminal Law 194; see also R v Bowen [1996] 2 Cr App R 157). Does this mean that if the standard for a "reasonable person" is a "reasonable victim of exploitation", does this turn the enquiry into a subjective one? (Laird 2016 Criminal Law Review 400). In addition, the question of what constitutes a relevant characteristic has yet to be determined but would appear to suffer from the same deficiencies as its common-law predecessor (Laird 2016 Criminal Law Review 400).

 

5 South African law

It is instructive to compare the English legal position in relation to the defence of duress as dealt with in Brandford with the position in South African law. In this regard, the defence of necessity in South African law will be juxtaposed with the English law counterpart of duress of circumstances, in order to compare and contrast the nature of the respective defences, and to see whether South African law would take a similar approach to the factual scenario that arose in Brandford.

The legal position pertaining to necessity has been set out in S v Goliath (1972 (3) SA 1 (A)). Necessity can constitute a complete defence, even in cases of murder (S v Bailey 1982 (3) SA 772(A) on the basis that heroism is not expected from ordinary people in life-and-death situations (25B-D). No distinction is made between threats induced by natural causes or by means of human agency (Yeo "Compulsion and Necessity in African Criminal Law" (2009) Journal of African Law 90 93; see also S v Goliath supra 24). For such a defence to be successful, certain conditions must be met. These include that:

a) a legal interest be threatened;

b) the threat have already commenced or be imminent;

c) the threat not be caused by the accused's fault;

d) the threat makes it necessary for the accused to avoid the danger; and

e) reasonable means be used to avert the danger (Burchell South African Criminal Law and Procedure: General Principles of Criminal law (2011) Vol 1 148).

English law requires that the existence of the threat need only be based on an honest belief on the part of the defendant. However, in South African law, because necessity operates as a justification ground, not only must the threat be real, but it must be of such a degree that no reasonable person would be able to withstand it (S v Goliath supra 11D; S v Peterson 1980 (1) SA 938 946 E-F; Burchell "Unravelling Compulsion Draws Provocation and Intoxication Into Focus" 2001 South African Journal of Criminal Justice 363). In other words, the accused's beliefs are not considered a factor regarding the enquiry into unlawfulness (Burchell Criminal Law 4ed (2013) 162). The accused's beliefs only become relevant when his or her conduct is proved to be unlawful - that is, where fault is present on the accused's part (Burchell Principles of Criminal Law 4ed (2013) 166-167). This point is crucial, in light of our new constitutional dispensation, as well as the culture of crime and violence and "blatant" disregard for human life (S v Mandela 2001 (1) SACR 156 (C) 166i-j).

Noting these points, the court in Mandela rejected a defence of necessity where certain factors were absent, such as the immediacy of life-threatening compulsion (168b). After the pronouncement in Mandela, it appears as if necessity has been relegated to the realm of criminal excuse: achieving a compromise between limits of human fortitude and constitutional ideals, such as the right to life (Le Roux "Killing Under Compulsion, Heroism and the Age of Constitutional Democracy" 2002 South African Journal of Criminal Justice 100 104; S v Mandela supra 168c-d). While fault is a requisite condition for necessity (S v Bradbury 1967 (1) SA 387 (A) 404H; S v Lungile 1999 (2) SACR 597 (SCA) 603c-d), case law has demonstrated that convictions will not solely be based on association with an organised crime syndicate that is known for a vengeful disciplinary code of conduct (per Holmes JA in S v Bradbury supra 404H, quoted in S v Mandela supra 164i-j). Rather, fault is merely one of the factors that courts use to determine whether an accused can successfully rely on the defence (Le Roux 2002 South African Journal of Criminal Justice 101). The ruling in Bradbury would be confined to members of a gang who at least know or foresee the violent nature of the gang and its code of vengeance, which they may be compelled to follow (Burchell 2001 South African Journal of Criminal Justice 363).

Although, in theory, the defence could be available in the Brandford scenario if it could be shown that there was no question of voluntary association, it would fail since the issue turns not solely on the question of whether she joined the gang, but rather on the absence of immediacy of life-threatening compulsion (S v Mandela supra 168b) and therefore any action taken was not necessary to avert the danger. Although the appellant in that case had argued that she had not appreciated the true nature of the threat up until the night in question (R v Brandford supra par 27), any reasonable person in the position of the appellant would have appreciated the nature of the threat that Alford faced (par 46). This is because of the length of the appellant's association with Alford and his dealings with the criminal syndicate (R v Brandford supra par 46). These included three previous incidents known to the appellant, which included the January pepper spray incident, the Allman murder,and the June stabbing (R v Brandford supra par 27), which she alleged she only "half believed" (par 27). Further, Brandford's general demeanour and conduct was not indicative of a person who was unduly fearful following this series of linking events (R v Brandford supra par 27).

The effect of the Mandela decision, in a particular context other than that of Brandford, is that the defence of necessity has been relegated to the realm of criminal excuse. Since the accused is exercising a choice to protect his life over that of another person, the defence will only prevail where heroic acts that extend beyond the capacity of a reasonable person would have been necessary to avert any possible harm. In such cases, it would be evident that the accused lacked the requisite culpability needed (S v Mandela supra 167c-e; Le Roux 2002 South African Journal of Criminal Justice 103).

Does this mean that the Mandela case is advocating the normative approach? (Le Roux 2002 South African Journal of Criminal Justice 104). Le Roux notes that in light of the rejection by Davis J of compulsion as a justification ground, a compromise solution was reached: acknowledging the constitutional principle of the right to life but also acknowledging the "limitations of human fortitude: heroism cannot be expected of people who are facing their own death" (Le Roux 2002 South African Journal of Criminal Justice 104). In other words, killing under compulsion would only be a defence where the accused committed the offence without a blameworthy state of mind - that is, with an absence of criminal capacity or knowledge of unlawfulness (Le Roux 2002 South African Journal of Criminal Justice 104). Le Roux noted that the normative evaluation of blameworthiness takes place, not because mens rea is absent, but where and precisely because it is present in terms of the psychological approach (Le Roux 2002 South African Journal of Criminal Justice 104).

However, it is submitted that a move towards a normative approach is problematic for a number of reasons. First, it places traditional necessity and putative necessity on an equal footing as a defence to fault. This could lead to an obfuscation of the two defences since with traditional necessity all requirements for the defence must be met, whereas with putative necessity, the normative concept of fault rests on whether or not the necessity arose from a mistake of law or "unavoidable ignorance" (Van Oosten "The Psychological Fault Concept Versus the Normative Fault Concept: Quo Vadis South African Criminal Law (Continued)" 1995 THRHR 568 574). In contradistinction, relying on putative necessity implies that the requirements for traditional necessity have not been met. The normative concept of fault implies that the defence will only be successful on the basis of circumstances of mistake of law or "unavoidable ignorance" (Van Oosten 1995 THRHR 568 574). However, it seems that if the accused lacked awareness of unlawfulness and there was therefore an absence of intention in terms of psychological fault, the defence would still not be available. Real necessity, on the other hand, affords a full defence despite the presence of intention. It also raises the issue of how an unlawful killing committed in circumstances of necessity with the presence of intention can be harmonised with a conviction on the basis of crime with intention, where awareness of unlawfulness is absent on account of mistake of law or unavoidable ignorance (Van Oosten 1995 THRHR 568 574). Lastly, necessity as a defence to fault, as opposed to unlawfulness, leads to the same result as necessity as a defence to unlawfulness rather than to fault (Van Oosten 1995 THRHR 568 574).

It is submitted that the position in South African law, pre Mandela, is preferable since the traditional justification ground of necessity adequately takes the accused's personal characteristics into account when personalising the objective standard. It does not require, like English law, a standard of heroism (S v Goliath supra 25) that the accused cannot reasonably be expected to meet (Williams "Necessity: Duress of Circumstances or Moral Involuntariness" 2014 Common Law World Review 1 5). Not only is it "morally unfair" to punish those that fall short of the standard but it also breaches the principle of fault (Williams 2014 Common Law World Review 6). South African courts would also not face the dilemma that English courts have faced in deciding which factors should be taken into consideration in the objectiveness enquiry (Virgo "Are the Defence of Provocation, Duress and Self-Defence Consistent?" 2002 Archbold Review 4 6). Although the English approach has been that characteristics are only relevant if sufficiently defined (Virgo 2002 Archbold Review 4 6; see also R v Bowen [1997] 1 W.L.R. 294, 300) and not self-induced (Virgo 2002 Archbold Review 4 6; Cochrane v Her Majesty's Advocate 2001 S.C.C.R. 655 par 21), the problem still rests with the notion that a defence of compulsion may be available in the context of close relationships (Virgo 2002 Archbold Review 6) The problem with this approach is that excuse requires that human frailties be considered as part of the standard of the "reasonable person" whereas "abnormality of mind" denotes a shift of the defence to the realm of excuse. However, as has correctly been pointed out, pressure must always be accompanied by threat of death or serious bodily injury (Laird 2017 Criminal Law Review 557). This was not the case in Brandford, and it therefore would also not be successful in terms of South African law.

In relation to the question whether such a defence would be available under a similar set of facts, the following points are noteworthy. The central focus is whether there was voluntary association. The ruling in Bradbury would be confined to members of a gang who at least know or foresee the violent nature of the gang and its code of vengeance that they may be compelled to follow (Burchell 2001 South African Journal of Criminal Justice 363). Although the appellant argued that she had not appreciated the true nature of the threat up until the night in question (R v Brandford supra par 27), any person in the position of the appellant would have appreciated the nature of the threat that Alford faced (par 46). This is because of the length of the appellant's association with Alford and his dealings with the criminal syndicate (R v Brandford supra par 46). These included three previous incidents known to the appellant, which included the January pepper-spray incident, the Allman murder and the June stabbing (R v Brandford supra par 27), which she alleged she only "half believed" (par 27). Her conduct was also not that of an individual who was unduly fearful following this series of linking events. Her evidence indicated that she only "half believed" what Alford had told her about the modus operandi of the syndicate (for instance, the January pepper spray incident, the Allman murder and the June stabbing (R v Brandford supra par 27)). Even if the defence of necessity were in principle available on the basis of the "relatively low standard" for assessing necessity as set out in S v Goliath (supra), she would not be viewed as having acted reasonably in the circumstances. This is because she had other reasonable alternatives, which clearly demonstrates that she did not act reasonably (S v Goliath supra). These circumstances include a lack of immediacy of life-threatening compulsion (S v Goliath supra) and the fact that she had the opportunity to change her mind, contact the police or ask her father for assistance (S v Mandela 2001 (1) SA 156 (C)).

 

6 Concluding remarks

It is submitted that a bifurcated approach, which is currently followed in the English law of compulsion and duress, is problematic and ought to be avoided at all costs. This approach was given prominence in the case of Mandela, where the court implied that, in cases of compulsion, a normative approach ought to be adopted. It is submitted that a "one-size-fits-all approach" as followed prior to the Mandela decision is to be preferred for the following reasons. First, the dichotomy between justification and excuse plays an indispensable role in extrapolating and expounding the goals that criminal law seeks to achieve. One of these goals is ensuring legitimacy by elucidating the problems in criminal responsibility, which are essentially geared at reflecting community values. The second goal of criminal law is efficiency. This can be achieved by ensuring that the distinction in the law maintains coherence and clarity needed for correctly attributing blame (Mousourakis "Distinguishing Between Justifications and Excuses in the Criminal Law" 1998 Stellenbosch L. Rev. 165 180). These goals can be attained by maintaining the traditional approach to necessity- that is, by treating it as a justification ground. The primary difficulty with adopting a pragmatic approach as has been followed in Mandela and in the English law is that such an approach does not sufficiently maintain key distinctions. The case of Mandela, taken to its logical conclusion, leads to the same primary problems that have been critiqued in the English approach - namely, a failure to maintain critical distinctions. For example, consider the defence of necessity and duress of threats and circumstances. The English courts have gone on to develop duress of circumstances as an excusatory defence, which is problematic since the defence now covers circumstances that are viewed as necessity in other jurisdictions (Williams 2014 Common Law World Review 5). The English approach also appears to incorporate into the liability enquiry aspects that belong in the sentencing enquiry. This is clear when one considers how the courts have grappled with which factors to take into consideration in the test for duress, whether by threats or circumstances. The test now incorporates a subjective criterion: did the defendant in the circumstances as she reasonably believed them to be have good reason to fear that serious bodily harm or death would follow if the offence was not committed (Virgo 2002 Archbold News 4). Excuse requires that human frailties be considered as part of the standard of the "reasonable person", possibly denoting a shift to the realm of excuse (Mousourakis Stellenbosch L. Rev. 178). To prevent this, the courts will have to exercise circumspection when determining which factors should be imputed to the reasonable person standard to ensure that such a standard is not rendered inapplicable. "Abnormality of mind" is a completely different legal defence, one which falls outside the realm of "human frailty" (Mousourakis Stellenbosch L. Rev. 179). The introduction of a novel defence of compulsion, as suggested by Laird, would merely create more confusion, not only conflating two very different defences but also introducing a subjective element into the liability enquiry. In addition, diminished capacity is an issue that is to be dealt with at the sentencing stage of the trial. This novel defence highlighted by Laird was not raised in Brandford. However, it bears reiterating that while courts are final arbiters of the meaning of legal terminology, certainty is essential. If it is correct that, for a defence to qualify as an excuse, it is not essential that the coercive circumstances substantially reduce the decision-making capabilities of the actor involved (Chiesa "Duress, Demanding Heroism, and Proportionality" 2008 Vand. J. Transnat'l L. 741 759), then it creates a misleading understanding of the term "emotional pressure". The implication is that, although the actor may have been faced with a hard choice, it merely had the effect of limiting his capacity, not excluding it. In other words, he retained the capacity to choose (Chiesa 2008 Vand. J. Transnat'l L. 741 759). This distinction plays an important role in South African law. The basic tenets of criminal liability demand that the conduct of the actus reus be proved to be voluntary, or put another way, that it be subject to the conscious will of the accused. Once the accused is shown to be acting voluntarily, there is no need to ask for a second time whether the conduct is voluntary (Louw "The End of the Road for the Defence of Provocation?" (2004) South African Journal of Criminal Justice 200 204). Capacity is not capable of gradation in South African law, but is taken into account at the sentencing phase (S v Bradbury supra 394 F-G; Van Oosten 1995 THRHR 568 579).

In contradistinction, it could be pointed out that in English law, the courts still punish those who substantially lack capacity to control their actions (Chiesa 2008 Vand. J. Transnat'l L. 741 760). Chiesa makes use of the example of the defendant who loses self-control, shoots his wife's rapist, and injures him. In such an instance, the law does not provide the defendant with a partial or full excuse for his conduct should the rapist survive (760). The opposite is also true in South African law. Louw has noted that courts have acquitted the accused despite the presence of capacity (Louw "S v Eadie: Road Rage, Incapacity and Legal Confusion" 2001 South African Journal of Criminal Justice 206 215, discussing S v Moses 1996 (1) SACR 701 (C) 714H-I). Since diminished capacity should rightly be taken into consideration at the sentencing stage of the trial, there must be another important theoretical reason to maintain the distinction between justification and excuse for determining liability (McCauley "Necessity and Duress in Criminal Law: The Confluence of Two Great Tributaries" 1998 Irish Jurist 120 127). This is despite the fact that both justification and excuse could in principle lead to an acquittal.

Another problem that arises concerning the justification/excuse dichotomy is that it presents certain practical problems. For instance, in cases of duress and excuse, a person who assists another in the commission of a crime should be convicted as an accessory, even though the principal offender is excused (Mousourakis 1998 Stellenbosch L. Rev. 166). In contradistinction, an accessory to a crime would escape criminal liability if the principal successfully pleads a justification-based defence (Mousourakis 1998 Stellenbosch L. Rev. 166). The basis for an acquittal would be putative defence. This is the position in South African law, and it is submitted that it would be preferable for a defence of necessity to be assessed on the basis of a test that reflects the proper nature of the inquiry: in general terms, an objective assessment for the actus reus, and a subjective assessment for mens rea (Taitz "Compulsion as a Defence to Murder: South African Perspectives" 1982 Law & Just. - Christian L. Rev. 10 17).

The distinction between justification and excuse is important and should remain a justification ground: the basis on which an acquittal rests serves a "symbolic function in criminal law since if it is raised as legitimate defence, it assumes that the holder of the right can use force against an unlawful attack" (McCauley 1998 Irish Jurist 120 127). Describing a defence as one of justification sends a clear message that the conduct is approved. Furthermore, a person

"cannot turn away from his concrete interests when he is evaluating the [dilemmatic choice with which he is confronted] ... the state acknowledges that, even though from an objective point of view the interests of a person who acts under duress have no more weight than the interests of the actor's innocent victim, it is comprehensible that citizens attach more value to their own ends." (Chiesa 2008 Vand. J. Transnat'l L. 760)

Finally, in accordance with the approach of the Court of Appeal in Brandford, it may be concluded that there is no logical basis for excluding an accused from a defence of necessity on the basis of indirect threats. However, the success of such a defence would be dependent on whether there was "immediacy of life-threatening compulsion".

Samantha Goosen

University of KwaZulu-Natal, Pietermaritzburg