Briggs, Margaret --- "Criminal attempts: how close is too close? Inaugural professorial lecture" [2018] OtaLawRw 4; (2018) 15 Otago LR 241

Last Updated: 11 May 2021

241

Inaugural Professorial Lecture Criminal Attempts: How Close is too Close?

Margaret Briggs*

I Introduction

The criminal law is engaged in many line drawing exercises. One that presents real challenges is locating the point at which a person’s conduct passes from mere non-criminal preparation to a fully-fledged attempt to commit an offence. As a general rule, the law does not criminalise a person’s initial acts of preparing or planning to commit an offence.¹ However, acts that come close to the full offence are said to constitute an attempt and are regarded as justifying criminal sanction.
Whereas completed offences tend to provide evidence of a tangible or objectively verifiable outcome,² the same cannot be said of attempts which, by their very nature, fall short of the full offence. Accordingly, it is necessary to identify the “minimum conduct”³ required for an attempt since that marks the first point at which a person can be held responsible for his or her actions.⁴ The problem in any given case is knowing where to draw the line along the spectrum between the actor germinating the first kernel of intention to commit an offence and the final act necessary to convert the attempt into the full offence.
In many cases, identifying an attempt presents few problems, particularly where the actor has done everything he or she can to complete the offence. For example, the actor who fires a gun with murderous intent, but who misses or merely injures the victim is guilty of attempted murder. The cases that raise the more conceptually challenging issues about proximity are those where the actor’s conduct stops short, for whatever reason, at an earlier point on the spectrum. The following discussion investigates New Zealand’s legislative and judicial responses

* Professor, Faculty of Law, University of Otago. This paper is a modified version of my Inaugural Professorial Lecture delivered at the University of Otago on 5 October 2017.

1. New Zealand law recognises some specific preparatory offences: see

below n 66-69 and accompanying text.

2. That is the case with result crimes, which require proof of a prohibited result or consequence, compared with conduct crimes, where the actus reus itself is the prohibited conduct.
3. A Ashworth and J Horder Principles of Criminal Law (7th ed, Oxford

University Press, Oxford, 2013) at 461-462.

4. The inchoate crime of conspiracy may occur earlier in time than an attempt. However, a conspiracy requires the participation of two or more people who agree to commit an offence, whereas an attempt may be committed by one person acting alone.

to identifying when an actor’s conduct becomes a proximate attempt, and concludes by questioning whether there might be an alternative means of reaching a satisfactory answer.

II The Proximity Test

The statutory formulation of criminal attempt is located in s 72 of the
Crimes Act 1961:⁵

Section 72 Attempts

(1) Every one who, having an intent to commit an offence, does or omits an act for the purpose of accomplishing his or her object, is guilty of an attempt to commit the offence intended, whether in the circumstances it was possible to commit the offence or not.

(2) The question whether an act done or omitted with intent to commit an offence is or is not only preparation for the commission of that offence, and too remote to constitute an attempt to commit it, is a question of law.

(3) An act done or omitted with intent to commit an offence may constitute an attempt if it is immediately or proximately connected with the intended offence, whether or not there was any act unequivocally showing the intent to commit that offence.

It is apparent from ss (2) that not all acts will meet the conduct standard required for an attempt: an act that is “only preparation” is “too remote to constitute an attempt”. Subsection (3) offers further context by providing that an act that is “immediately or proximately connected with the intended offence” may constitute an attempt. The words “immediately or proximately” suggest the act or acts must come close to the full offence, although whether in terms of time or physical location or both, is not expressed. Reading subsections (2) and (3) together, it is clear that a line is to be drawn between non-criminal preparation and criminal attempt, but the words do not, and indeed cannot, identify where, exactly, to draw that line.⁶
The statutory formula for proximity has remained unchanged in New Zealand for more than half a century, during which time the courts have applied various interpretations to s 72 in an effort to better identify the

5. Earlier statutory formulations of attempt are contained in the Criminal Code Act 1893, s 76 and the Crimes Act 1908, s 93.
6. E Meehan and JH Currie The Law of Criminal Attempt (2nd ed, Carswell,

Ontario, 2000) at 105.

point at which a proximate attempt begins.⁷ For all that, no single gloss or test has coped well in practice with the complex matrix of different offences and varying factual scenarios that have come before the courts.⁸ Finding the dividing line can depend very much on the type of offence involved and the steps the defendant is alleged to have taken. An attempted fraud, for example, comprises a very different set of steps to an attempted murder. What has become more noticeable in recent times is the retreat from a defendant-friendly approach to proximity, whereby the police must wait until the actor poses a real and present threat to potential victims before being able to step in to prevent a putative attempt. The shift in judicial thinking in New Zealand about the right place to draw the line between non-criminal preparation and conduct that constitutes a proximate attempt is documented in the following discussion.

III Judicial Responses to the Proximity Test

(a) R v Wilcox – the relevance of physical proximity

The Court of Appeal decision in R v Wilcox⁹ played an influential role in judicial thinking on proximity for almost three decades. The case concerned a charge of attempted aggravated robbery of a post office. The defendant and his accomplices had performed a number of acts in pursuance of their plan, including the purchase of balaclavas to conceal their identities and air rifles to present at the post office staff. They loaded the air rifles and set off in a car towards their destination some three miles distant. However, the police had been informed of the plan and stopped the car when it was one mile from the post office.
Justice Woodhouse, who delivered the judgment of the Court, made two noteworthy observations about attempts. First, he stressed the importance of giving independent and careful attention to the two ingredients of mens rea and actus reus.¹⁰ Thus, even if there is proof of a person’s criminal intention to commit an offence, he or she may not yet have performed enough physical actions to qualify as an attempt.

7. Tests developed both in New Zealand and elsewhere, have included the unequivocality test, the last act test, the real and practical step test, and the dangerous proximity test. See generally, S France (ed) Adams on Criminal Law – Offences and Defences (online looseleaf ed, Thomson Reuters) at [CA72.13]–[CA72.15]; AP Simester and WJ Brookbanks Principles of Criminal Law (4th ed, Thomson Reuters, Wellington, 2012) at 243-265; E Meehan and JH Currie, above n 6; A Duff, Criminal Attempts (Clarendon Press, Oxford, 1996) at ch 2; AP Simester and GR Sullivan Criminal Law: Theory and Doctrine (3rd ed, Hart, Oxford, 2007) at 305-309.
8. The proximity problem has not been confined to New Zealand. Canadian

authors Meehan and Currie, above n 6, at 107, have aptly observed that “[g]leaning, from the morass of reported cases, a formula of universal application to all factual situations is like trying to remove chewing gum from an angora sweater.”

9. R v Wilcox [1982] 1 NZLR 191 (CA).

10 At 193.

Secondly, Woodhouse J considered that “independent acts of mere preparation” such as the purchase of the balaclavas or the rifles “[could not] take on a different quality simply by adding them together”.¹¹ Therefore, a basic cumulative or quantitative assessment of the discrete acts done by the actor would not necessarily attain the qualitative measure required to make a finding of proximity.
The Court held that the appellant’s actions had not gone beyond
preparation, Woodhouse J stating:¹²

... if the requirement of proximity or immediacy is to be given any realistic and practical meaning for the purposes of s 72 we do not think it possible to hold that any of the passengers in the car by the time it was stopped by the police were doing more than getting themselves into a physical location close to the post office from which an attempt at robbery could be directly launched.

Justice Woodhouse considered Henderson v The King,¹³ a decision of the Supreme Court of Canada, with facts similar to those in Wilcox. There, three men had driven to within a short distance of the bank they planned to rob, but as they approached they noticed a police car nearby and so drove off. By a majority the Supreme Court held the men were guilty of an attempt to rob on the basis that they “had proceeded so far that within sight of the bank they were frustrated by the presence of the police.”¹⁴ Although Woodhouse J distinguished Henderson because the facts of that case were “rather stronger” than those in Wilcox,¹⁵ he noted with approval the approach taken in the dissenting judgment of Taschereau J:¹⁶

Here, the trio were seen in an automobile in the direction of the bank; but the plot was frustrated by the presence of the police. There was nothing done by the trio, no overt act immediately connected with the offence of hold-up and robbing. Although it may be said that no one could doubt the express purpose of the bandits, I do not believe that it can be held that the mere fact of going to the place where the contemplated crime is to be committed, constitutes an attempt. There must be a closer relation between the victim and the author of the crime; there must be an act done which displays not only a preparation for an attempt, but a commencement of execution, a step in the commission of the actual crime itself.

Whatever should be said about criminals who have arrived at the very scene of a projected crime, Woodhouse J accepted Taschereau J’s general analysis as applicable in the context of the facts in Wilcox.¹⁷

11 At 194.

12 At 195.

13. Henderson v The King (1948) 91 CCC 97 (SCC).
14. At 116 per Estey J.
15. R v Wilcox, above n 9, at 195.
16. Henderson v The King, above n 13, at 105 per Taschereau J (emphasis in original).
17. R v Wilcox, above n 9, at 195.

(b) Wilcox in decline

Wilcox was cited in a number of subsequent cases.¹⁸ Nevertheless, it did not find universal support, with one commentator criticising the decision for failing to cite earlier New Zealand authorities, “which seems to give the law of attempt a narrower scope than that previously allowed.”¹⁹ For example, Wilcox had made no reference to the earlier Court of Appeal decision of R v Bateman,²⁰ where the Court had identified the defendant’s acts as sufficiently proximate in circumstances where the defendant was arrested without the opportunity to travel to the location where the offending was to take place.
Lower courts also cast doubt on Wilcox on two occasions. In Drewery v Police,²¹ a case about attempting to defraud an insurer, Williamson J was concerned that Wilcox may have misled itself as to the application of s 72(3). Rather: ²²

It would seem to be a matter of degree so that if the evidence of intent is strong and clear the proximity or immediacy may not have to be as great as in cases where evidence of intent is reliant upon inferences to be drawn from the nature of the act itself.

R v Burrett, Payne and Philpott (No 2) echoed similar concerns several years later.^.23 Burrett involved the attempted kidnapping of a wealthy businessman from his home in Wellington. The defendants had gone to considerable efforts to conceal the victim, including digging an underground bunker in which to hide the victim until his family paid the ransom the defendants planned to demand for his safe return. However, the police had been watching the defendants for some time and arrested them on the day of the attempt, a few hundred metres from the victim’s home. The question was whether the defendants were already in the throes of the attempted kidnapping when the police stopped them. Applying Wilcox, Hammond J reluctantly concluded the defendants’ conduct was merely preparatory, but noted:²⁴

If it was open to this Court to ‘start afresh’, I have to say that I would prefer a view which did not make proximity as dispositive as it currently is in our

18. See R v Hoenderdos HC Christchurch T31/92, 23 October 1992; R v Bobos [1993] DCR 1105; R v Drummond [1993] NZHC 2651; (1993) 9 CRNZ 228 (HC); R v Gray CA374/98, 18 February 1999; R v Mullaney HC Hamilton T000172, 3 May 2000.
19. G Orchard, Crimes Update (New Zealand Law Society Seminar, October 1990) 63.
20. R v Bateman [1958] NZPoliceLawRp 34; [1959] NZLR 487 (CA), where the defendant had suggested

to a 17-year old youth that they go to the defendant’s house for indecent purposes.

21. Drewery v Police [1988] NZHC 351; (1988) 3 CRNZ 499 (HC). Drewery was a case of attempted

insurance fraud. For a discussion of the decision see Orchard, above n 19, at 63.

22 At 503.

23 R v Burrett, Payne and Phillpott (No 2) HC T3347/02, 13 February 2003. 24 At [17].

law. It is true to say that an attempt has to begin at some point. And that being so, a consideration of ‘proximity’ could hardly be entirely eliminated from consideration in this crime, if we are to maintain a general law of attempt. But for myself I would have preferred a view that proximity is a relative matter. It might vary in its importance having regard to the strength of such other evidence as there is, as to the purposes of the accused. But as defence counsel urged upon me – and Mr Stone appeared to accept – this does not represent the present law in New Zealand.

(c) A fresh start

In 2010 the Court of Appeal in R v Harpur²⁵ took the opportunity to make the “fresh start” referred to in Burrett. Harpur concerned allegations of attempted sexual violation of two children. The defendant sent an adult woman, with whom he was in a text messaging relationship, an explicit video and texts indicating he was involved in the abuse of young children. Under police guidance the woman arranged to meet the defendant, telling him she would bring her 10-year-old sister and four-year-old niece with her. Unbeknown to the defendant, neither child existed in fact.²⁶ The defendant was waiting when a car driven by a female police officer arrived at the arranged meeting place.
The trial Judge discharged the defendant on the basis that, although his intentions were clear, his actions had not gone beyond preparation.²⁷ The Court of Appeal overturned the trial ruling, finding that Harpur had: ²⁸

... performed a number of acts which, taken together, constituted an attempt to commit sexual violation. He had moved beyond mere preparation and, at the time of his arrest, was lying in wait for his victim.

The Court considered that Parliament had “painted a very broad canvas” when drafting s 72 and consequently, “[i]f Parliament for good reason chose not to be prescriptive, this Court should not be either.”²⁹ In the Court’s opinion s 72 permitted the “defendant’s conduct to be considered in its entirety”,³⁰ with effect that “strong evidence of intent,

25 R v Harpur [2010] NZCA 319; (2010) 24 CRNZ 909 (CA).

26. The non-existence of the children was irrelevant for the purposes of attempt. Section 72(1) of the Crimes Act 1961 provides that if a person intends to commit the offence and has committed [proximate acts] he or she is guilty of an attempt “whether in the circumstances it [is] possible to commit the offence or not”.
27. R v Harpur DC Palmerston North CRI-2008-054-223, 18 February 2009.
28. R v Harpur, above n 25, at [44]. By the time the case came before the Court of Appeal, the Crown accepted that the trial Judge had been correct to discharge Harpur with respect to the attempted rape of the 10-year-old, and the hearing proceeded only on the count of attempted sexual violation of the four-year-old by unlawful sexual connection: at [10].

29 At [16].

30 At [36].

as there is in this case, can assist in assessing the significance of acts done

towards the commission of the intended offence.³¹

Harpur’s explicit text messages that revealed his intentions towards the victims were, therefore, a central factor in deciding the proximity of his actions. The Court found support for its approach in Canadian authority, including R v Boudreau where the Nova Scotia Court of Appeal held that any analysis of the actus reus must be viewed in conjunction with the mens rea.³² It also cited with approval the opinion of Canadian academic Kent Roach that “[i]n practice, a more remote actus reus will be accepted if the intent is clear.”³³
The Court of Appeal stated that the direct purpose of its decision was to reassess Wilcox which it thought had been decided wrongly on the facts, rather than try to invent a new test for proximity.³⁴ It is open to argument, however, that the legacy of Harpur goes further than a simple finding that Wilcox was decided incorrectly. Rather, Harpur advances a different way of examining proximity by assessing intention and conduct together. Recall that under the previous Wilcox formulation, a defendant’s intention to commit the offence was assessed independently from the question of whether the defendant’s conduct had progressed far enough, with effect that a defendant’s intentions did not colour the assessment of the steps taken towards the full offence. Under the new Harpur formulation, these two inquiries are dealt with together. The question is whether the change in approach has any effect on where the dividing line between preparation and proximity may be drawn. In other words, does the Harpur method of assessing liability for attempt provide the opportunity for courts to shift the dividing line between preparatory and immediately or proximately connected conduct back further in cases where there is strong evidence of intention?

31. At [38] (emphasis added).
32. R v Boudreau 2005 NSCA 40; (2005) 193 CCC (3d) 449 at [32]. Boudreau concerned a charge of attempted murder where the defendant had pointed a rifle at his estranged partner: see generally, the discussion on the actus reus and mens rea of attempt at [27]-[34]. On the distinction between preparation and attempt in Canadian decisions, see: R v Cline (1956) 115 CCC 18, 4 DLR (2d) 480 (ONCA); R v Sorrell (1978) 41 CCC (2d) 9 (ONCA); Deutsch v R [1986] 2 SCR 2; R v Mantley (2013) NSCA 16.
33. R v Harpur, above n 25, at [25], citing K Roach, Criminal Law (3rd ed, Irwin Law, Toronto, 2000) at 102.
34. At [19] and [37]. Note also the discussion at [45]-[48], where the Court rejected the Crown’s suggestion that it endorse “the real or substantial step” test, which had been adopted in numerous cases including Police v Wylie [1976] 2 NZLR 167 (CA); Drewery v Police, above n 21; R v B (No 5) HC Christchurch T19/01, 7 September 2001; R v Yen [2007] NZCA 203 at [22]- [23]; and R v Towgood [2007] NZCA 359 at [18].

The final case in this overview, Johnston v R,³⁵ confirms Harpur. Moreover, Johnston signals a readiness to give broad licence to what amounts to the ‘strong evidence of intent’. Like Harpur, Johnston involved an alleged attempted sexual violation. The father of the 16-year-old female complainant found the defendant prowling in the backyard of the family home a few metres from a detached sleep-out where the girl slept at night. It was early evening, Johnston was wearing dark clothing and was carrying a torch, and there was evidence that he had been watching the house for some weeks prior to the incident.
The Supreme Court upheld the Court of Appeal’s finding that Johnston’s conduct amounted to attempted sexual violation.³⁶ It agreed with Harpur that evidence of intention is relevant to determining whether a defendant’s actions are proximate. It said:³⁷

Where there is clear intent to commit the completed offence, the maker of the ‘more than preparation’ decision has available to him or her information about what the defendant’s ultimate plan was, which enables him or her to assess more accurately whether the defendant’s acts amount to an attempt to commit the planned offence. Without that information, the acts may be seen as equivocal, and the decision-maker could not be confident that they amount to an attempt to commit a particular offence. This does not turn mere preparation into an attempt. Rather, it is recognising that where clear intent is shown, the decision-maker has a basis to determine whether the conduct is more than mere preparation.

The Court considered that Johnston’s presence on the property and his movement towards the sleepout could not be assessed in terms of whether it was preparatory or proximate without considering the evidence of his intent to sexually violate a teenage girl.³⁸ Yet, unlike Harpur where the text messages provided documentary evidence of the defendant’s intentions, there was no direct evidence of Johnston’s intentions because Johnston maintained that he was on the property to commit burglary.³⁹ Notwithstanding, the Crown sought, and was permitted, to adduce propensity evidence to help establish that Johnston was on the property with the intention of committing sexual violation,

36. The Court of Appeal (Johnston v R [2012] NZCA 559; [2013] 2 NLZR

19) ordered a retrial because the Judge’s summing up had introduced the possibility that Johnston may not have intended the rape to occur that night. The Court held the Judge should have clearly concluded that Johnston’s presence on the property without intent to commit rape that night would not have been proximate. The error amounted to a miscarriage of justice. The appellant was subsequently retried and found guilty for a second time. See R v Johnston [2013] NZHC 3359 (sentencing judgment).

37. Johnston v R, above n 35, at [57].

38 At [59].

39 At [23].

not burglary.⁴⁰ The evidence admitted included two prior convictions for sexual violation in the 1990s, and the evidence of a fellow prison inmate and a half-way house flatmate regarding their conversations with Johnston about his obsession with teenage girls.⁴¹ The Court of Appeal explained that it was: ⁴²

... important to draw a distinction between the clarity of the intent alleged by the Crown and the strength of the evidence to support the intent alleged by the Crown. It is clear from Harpur that in approving the principle that a less proximate actus reus will be sufficient where the intent is clear, this Court meant the intent to be assessed on the basis that the jury accepts the essential attributes of the Crown case.

The Court identified two reasons why Johnston’s presence on the property with the intent to commit sexual violation was sufficiently proximate to constitute an attempt.⁴³ First, as in Harpur and Bateman, Johnston “was effectively lying in wait for the victim”.⁴⁴ Secondly, the Court considered ‘practical’ or ‘policy’ considerations. Had the police known Johnston was on the property (as in Harpur where the police were privy to the defendant’s plans) it would have been impractical for them to wait until he entered the sleep-out before acting because by then there would have been “considerable risk to the complainant”.⁴⁵
In summary, the decisions of Harpur and Johnston have developed a method of assessing proximity that focuses on evidence of the defendant’s intention to assist with the line drawing exercise. This is in contrast to the more compartmentalised test previously applied in Wilcox where the defendant’s intentions and actions were assessed separately. Arguably, the Harpur/Johnston analysis now enables the courts to reposition the

40. Johnston v R [2012] NZCA 559; [2013] 2 NLZR 19 [9]-[10]; Evidence Act 2006, s 40(1). The prosecution may offer propensity evidence about a defendant only if the evidence has a probative value which outweighs the risk that the evidence may have an unfairly prejudicial effect on the defendant: s 43(1). See J Rogers “The Codification of Attempts and the Case for ‘Preparation’” [2008] Crim LR 937 at 943-944, 949, who argues that if a defendant’s earlier (“preparatory”) actions are to be considered as relevant, then the prosecution should not be allowed to use evidence of bad character to show a relevant propensity of the defendant to offend under s 101(1)(d) of the Criminal Justice Act 2003 (UK).
41. Johnston v R, above n 40, at [9]-[10]. The defence unsuccessfully sought

to have all of Johnston’s previous convictions for burglary and other property-related offences ruled admissible. Only those convictions specifically relating to residential burglaries were admitted: see the discussion at [88]-[96].

42. Johnston v R, above n 40, at [29]. 43 At [32].
    44. At [33]. The American Model Penal Code, art 5.01(2)(a) lists “lying in wait” as an indicium of conduct that may be held to constitute a substantial step in a course of conduct planned to culminate in the commission of the intended crime.
    45. Johnston v R, above n 40, at [34].

preparation/proximity dividing line in some cases. Evidence of intention (including propensity evidence the court considers admissible) may be used to reinforce the conduct component in those cases where a close line call is required.

IV Interpreting the Changes

Many aspects of the criminal justice system including policing, procedure and sentencing policy have undergone transformation in the last two decades. Among the more publicised and consciously targeted changes over that time have been the political efforts to ‘get tough on crime’. This has resulted in enhancing police powers to respond to crime and threats of crime, tightening bail, reducing parole, increasing sentences and giving victims of crime a greater voice with the enactment of the Victims’ Rights Act 2002.⁴⁶
It is possible to discern a similar policy-motivated approach filtering through to some parts of substantive criminal law, including attempt. The interpretation of s 72 in the two recent attempt cases, Harpur and Johnston, highlights the less defendant-friendly stance that has been emerging in recent years.⁴⁷ The individuals involved in such cases are unsympathetic characters whom we rightly regard as deserving of censure. For example, the appellant in Johnston had a disturbing criminal record stretching back many years,⁴⁸ and charging him with attempted sexual violation had the practical advantage that, if convicted, he could be sentenced to preventive detention.⁴⁹
There is, then, a persuasive argument that practical or policy considerations directed at the public good rightly outweigh matters of strict doctrinal principle in these difficult attempt cases. The ‘tough on crime approach’ is an understandable response directed at community well-being. Yet it comes with the potential cost of undermining and diluting our general rights. The recent developments in the law of attempt demonstrate the important gatekeeper role played by the courts. Meehan and Currie have explained that “[b]y expanding or contracting the actus reus of attempt, the courts are able to use attempt as a method of controlling the area of conduct, and the nature of conduct, which is legally proscribed.”⁵⁰

46. See M Rowe, “Policing and ‘cracking down on crime”: tough questions

and tough answers” (paper presented to IPS Forum, February 2009).

<igps.victoria.ac.nz/events/downloads/2009/Michael%20Rowe.doc>.

47. Note that in R v Harpur, above n 25, at [19], counsel for the prosecution argued that although only one of a number of appellate authorities on attempts, Wilcox was the most frequently cited case on attempts, possibly because it was the most defendant-friendly.
48. He had a record of more than 120 convictions stretching back almost 30

years: R v Johnston, above n 36 (sentencing judgment).

49. Sentencing Act 2002, s 87.
50. Meehan and Currie, above n 6, at 110.

Whether the correct balance has been achieved between protecting the rights of the individual and enabling the police and courts to carry out effective law enforcement is open to debate. To assess whether the right balance has been achieved, it is worth returning to first principles to consider why it is that we treat an attempt to commit a crime as a crime in its own right. The most obvious answer is that it helps prevent harm, or at least more serious harm, from happening.⁵¹ Conduct leading to the commission of the full crime is a legitimate target because that conduct risks completion of the full crime.⁵² The police require the legal mandate to step in and prevent an actor who is on the way to committing a crime from going any further, without having to wait until the full offence is committed. Similarly, the courts must be able to acknowledge the criminality of the behaviour and be able to punish the attempter accordingly. As the recent case law demonstrates, there is a fine balance between ensuring public safety on the one hand and, on the other, preserving individual freedom unconstrained by police and judicial interference.
There are two theories about the law of attempt which help explain why the courts sometimes lean in favour of the prosecution and at other times, the defendant. The first is the subjective theory, which places the emphasis on the intention element of the offending. Criminal liability is primarily assessed on what the individual intended to do rather than on the actual consequences of their conduct.⁵³ This ‘nip it in the bud’ approach is designed to ensure public safety as far as possible and therefore justifies imposing liability at an early and more remote stage from the full offence.
By contrast, the objective theory focuses on the defendant’s conduct, and imposes liability at a later stage, once the defendant is physically close to committing the crime. Objectivists tend to require a manifest threat to the security interests of others⁵⁴ and place importance on the preservation of individual freedom unconstrained by police and judicial interference. Wilcox is an example of that more defendant-friendly objectivist approach. There, the Court placed emphasis on the physical steps that the defendants had not yet performed when they were stopped

51. See Simester and Brookbanks, above n 7, at 240-241 for a general discussion of the rationales for attempt.
52. See the discussion in Law Reform Commission (Ireland) Inchoate Offences,

LRC 99-2010 (2010) at [1.19].

53. A Ashworth and J Horder Principles of Criminal Law (7th ed, Oxford University Press, Oxford, 2013) at 456. For a discussion of subjective and objective theories see RA Duff Criminal Attempts (Clarendon Press, Oxford, 1996); CMV Clarkson “Attempt: The Conduct Requirement” (2009) 29(1) OJLS 25 at 36-37.
54. Clarkson, above n 53, at 37.

one mile from the Post Office.⁵⁵ The express language of s 72(3), that acts that are immediately or proximately connected with the full offence may constitute an attempt, suggest a closer connection with the principles of objectivism than subjectivism.
The problem is that both sides of the argument are susceptible to extremes. An overly rigid objective approach leads to only the last step or steps counting as an attempt. Delayed intervention risks unjustified harm to individuals or property and leaves the police and the judiciary vulnerable to criticism from the general community. Equally, if the subjective approach is pressed too far, there is a valid concern that overly zealous intervention may risk punishing people who have done little more than form criminal intention.
In New Zealand, the pendulum has recently swung in favour of subjectivism, with Harpur and Johnston placing added emphasis on the defendants’ intentions. Looking to the future then, care must be taken to avoid pushing that approach too far. In persuasive adversarial hands, almost any conduct could be made to appear guilty in the presence of strong enough evidence of the defendant’s intentions.⁵⁶ Thus, the point at which liability is imposed can “depend as much on judicial discretion as on legal definition”, bearing out “legal realist claims that facts decide cases, not law.”⁵⁷ For instance, a possible side effect could be that actors who have voluntarily abandoned their plans of their own accord could be deprived of such a claim if the subjectivist emphasis on intention is pushed too far.⁵⁸

55. Notably, UK courts have also required a close causal connection between the wrongdoer and the victim. In R v Geddes [1996] Crim LR 894, the appellant was apprehended in a school toilet block in possession of materials capable of being used to restrain a child. The Court of Appeal found that the conduct did not amount to attempted false imprisonment. While there was not much room for doubt about the appellant’s intention, the evidence showed he had never had any contact with, nor had confronted, any pupil at the school. See further, R v Toothill [1998] Crim LR 876, where the female occupier of an isolated property saw the appellant in the garden not far from the rear of the house. A knife and a glove were found in the garden, and a condom was found in his pocket. He admitted that he had knocked at the door for directions. The Court of Appeal upheld a conviction for attempted burglary with intent to rape, finding that the crucial step which established that the appellant had gone beyond the preparatory to the executory stage of his plan was that he knocked at the proposed victim’s door. Johnston, above n 40, considered both Toothill and Geddes but did not think the reasoning should be relied on since it preceded Harpur. Commenting on Toothill, the Court of Appeal observed at [27], “an act less proximate than knocking on the door or attempting to enter the sleep out was sufficient to constitute an attempt in [Johnston].”
56. H Stewart “The Centrality of the Act Requirement for Criminal Attempts” (2001) 51 U Toronto LJ 399 at 423.
57. Law Reform Commission (Ireland), above n 52, at [2.44].
58. For a discussion of voluntary abandonment see Duff, above n 53, at 66ff.

The preceding discussion is not to detract from the fact that both Harpur and Johnston involved allegations of disturbing intentions on the part of the actors, and it is hardly surprising that the courts gave added weight to the need to step in quickly to prevent harm to the putative victims. Nor is it difficult to identify other contexts, such as in cases of suspected terrorist activity, where there are equally cogent arguments for nipping nascent criminal activity in the bud as soon as practicable.⁵⁹ It is also worth noting that both Harpur and Johnston concerned charges of attempted sexual offending, and the full effect of the new formulation of proximity in contexts other than sexual offending is yet to be determined.⁶⁰

V Where Next?

Precisely how far back into the history of an actor’s conduct the law should probe in order to find liability is a question with no obviously right or wrong answer. That said, there is growing academic debate in the criminal law around questions of over-criminalisation and preventive justice,⁶¹ and on the limitations that should be placed on the pursuit of prevention through the criminal law.⁶²
Both Harpur and Johnston regard prevention of harm as necessary in the wider interests of justice. If we accept that there are cogent policy

59. See generally, J McCulloch and S Pickering “Future Threat: Pre-crime, State Terror, and Dystopia in the 21st Century” (2010) 81 CLM 32; T Tulich “A View Inside the Preventive State: Reflections on a Decade of Anti-Terror Law” (2012) 21 GLR 209; C Walker “The Reshaping of Control Orders in the United Kingdom: Time for a Fairer Go, Australia!” [2013] MelbULawRw 11; (2013) 37 MULR 143.
60. Dicta in two Court of Appeal decisions provide some insight about the possible future direction of attempt in New Zealand. Shadrock v R [2011] NZCA 388; [2011] 3 NZLR 573 emphasised the importance of crime prevention in the law of attempt. The Court observed at [73]:

... there is a concern that, if the proximity test is interpreted too narrowly, police will not intervene to prevent crimes until it is too late, for fear that no conviction will result. ... R v Harpur went to some effort to emphasise the flexible nature of the proximity test and that it is fact dependent.

New Zealand Police v Radhi [2014] NZCA 327; [2014] NZAR 1019 at [33], citing Harpur at [13], took a similar view endorsing the comment in Harpur that:

The breadth of s 72(1) and the looseness of the language employed suggest that Parliament intended the Courts to apply the provision flexibly and in accordance with the justice of the case, without undue constraint from the Legislature.

61. See for example, D Husak “Reservations About Overcriminalization” (2011) 14 New Crim L Rev 97; RA Duff, L Farmer, SE Marshall, M Renzo and V Tadros (eds) The Boundaries of the Criminal Law (Oxford University Press, Oxford, 2010).
62. AAshworth and L Zedner “Prevention and Criminalization: Justifications

and Limits” (2012) 15 New Crim L Rev 542.

reasons for allowing the police to step in at an early stage to prevent people such as the defendants in Harpur and Johnston from getting any further, the issue is whether ‘attempt’ continues to be the right offence and an appropriate label for the sorts of conduct the wider community deems worthy of punishment. Arguably, classifying the defendant’s actions as an ‘attempt’ is to mislabel what has thus far taken place. Critics might therefore suggest that we should be looking for a more doctrinally satisfying answer.⁶³
While time constraints this evening prevent a comprehensive examination of potential solutions, one option would be to acknowledge that the law is justified in criminalising certain preparatory activity that occurs prior to a full-blown attempt, and re-label it accordingly. The actor’s conduct would still be recognised as a threat requiring preventive intervention, but not to the same degree that it would have been had the conduct reached the stage of immediate or proximate connection needed to qualify as an attempt. A penalty reduction would recognise that the actor had taken fewer steps towards the commission of the full offence and that preparatory conduct is not equivalent to an attempt.
An offence of preparation would assuage subjectivist concerns by recognising the social benefits of stopping the actor at an early stage in order to prevent harm being done, while avoiding policy-driven extensions of the ‘attempt’ label.⁶⁴ It would also satisfy the requirement of giving individuals due notice of what not to do in order to avoid criminal sanction for a particular offence, thereby answering concerns about fair labelling and the current expansion of what qualifies as attempt.⁶⁵
The idea of offences of preparation is not without precedent. New Zealand criminal law currently recognises a number of specific preparatory offences that target certain forms of possession-based or preparatory activity, many of which contemplate the future commission of substantive offences. For instance, the Summary Offences Act 1981 makes it an offence to be found in a public place preparing to commit an offence.⁶⁶ The Crimes Act 1961 also contains several preparatory offences, including having in one’s possession an “offensive weapon or

63. See, for example, I Leader-Elliott “Framing Preparatory Inchoate Offences in the Criminal Code: The Identity Crime Debacle” (2011) 35 Crim LJ 80 at 82, who argues that the extension of criminal liability to a point isolated from the full offence involves “parasitic extensions of liability to ever more remote preparations for crime.”
64. Attempts would apply to activity of a genuinely immediate or proximate nature, although would not be confined to the last step or steps, or to ‘complete attempts’ where the defendant has done all the acts he or she set out to do but has failed to produce the result intended. See A Ashworth “Criminal Attempts and the Role of Resulting Harm Under the Code, and in the Common Law” (1988) 19 Rutgers LJ 725 at 734.
65. Sudden shifts in the judicial approach to attempts risk depriving the individual of fair notice. See Ashworth, above n 64, at 727.
66. Summary Offences Act 1981, s 28.

disabling substance in circumstances that prima facie show an intention to use it to commit an offence involving bodily injury or the threat of fear or violence.”⁶⁷ The Misuse of Drugs Act 1975 prohibits the supply, production, or manufacture of any “precursor substance”,⁶⁸ and the Arms Act 1983 prohibits carrying a firearm with criminal intent.⁶⁹
But it would be wrong to ignore the fact that a general offence of preparation could raise objections, the most obvious being that current concerns about locating the borderline between preparation and proximity would simply be replaced by the same problem once removed. There is a risk that a general offence of preparation could be interpreted as an invitation to extend liability further and still further back.⁷⁰ Critics would argue that this solution would risk normalising criminalisation at an ever earlier point.⁷¹
In conclusion, there is, then, no perfect solution to the line drawing conundrum in the law of attempt. The essence of the problem is, perhaps, best summed up by the master of logic and illogic, Lewis Carroll, who shows us that looking for answers by living backwards in time with the aid of a looking glass tends to produce perplexing outcomes.⁷² To that end, I leave you this evening to contemplate the White Queen’s musings on her preferred method of crime prevention:
When asked by Alice what things she remembered best, the White Queen replied “those that happened the week after next”. The Queen told Alice about the King’s Messenger who was in prison, being punished though “the trial doesn’t even begin till next Wednesday: and of course the crime comes last of all.”
But, “suppose he never commits the crime?” said Alice. “That would be all the better, wouldn’t it?” the Queen said.
Alice felt there was no denying that, “but it wouldn’t be all the better his being punished” she said.

67. Crimes Act 1961, s 202A(4)(b). The Crimes Act also targets certain “sexual grooming” behaviours, such as travelling to meet a young person under the age of 16 having earlier communicated with the young person: s 131B. Other Crimes Act examples are: s 227 (possession of an instrument capable of being used for conversion); s 233 (being disguised or in possession of an instrument for burglary); s 264 (possession of anything capable of being used to forge any document).
68. Misuse of Drugs Act 1975, s 12A.
69. Arms Act 1983, s 55.
70. See Clarkson, above n 53, at 34.
71. In 2007 the Law Commission (UK) published a consultation paper (Conspiracy and Attempts CP 183 (2007)) suggesting an offence of preparation, distinct from proximate attempt, to recognise the culpability of those individuals caught in the earlier stages of getting ready to commit a crime. The Commission later abandoned its proposal: see Conspiracy and Attempts EWLC 318 (2009).
72. L Carroll, Through the Looking Glass (MF Mansfield and A Wessells, New

York, 1899) at 64-65.

“You’re wrong there, at any rate ... were you ever punished?” asked the Queen.
“Only for faults,” said Alice.
“And you were all the better for it, I know!” the Queen said
triumphantly.
“Yes, but then I had done the things I was punished for,” said Alice, “that makes all the difference.”
“But if you hadn’t done them ... that would have been better still; better,
and better, and better!” said the Queen.
Alice was just beginning to say “there’s a mistake somewhere –,” when the Queen began screaming, so loud that she had to leave the sentence unfinished. “What is the matter?’ Alice said. “Have you pricked your finger?”
To which the Queen replied, “I haven’t pricked it yet ... but I soon shall ...”.