Holdouts in the South Pacific: Explaining Death Penalty Retention in Papua New Guinea and Tonga

The South Pacific forms a cohesive region with broadly similar cultural attributes, legal systems and colonial histories. A comparative analysis starts from the assumption that these countries should also have similar criminal justice policies. However, until 2022, both Papua New Guinea and Tonga were retentionist death penalty outliers in the South Pacific, a region home to seven other fully abolitionist members of the United Nations. In this article, we use the comparative method to explain why Papua New Guinea and Tonga have pursued a different death penalty trajectory than their regional neighbours. Eschewing the traditional social science explanations for death penalty retention, we suggest two novel explanations for ongoing retention in Papua New Guinea and Tonga: the law and order crisis in the former and the traditionally powerful monarchy in the latter.

The South Pacific forms a cohesive region with broadly similar cultural attributes, legal systems and colonial histories-a comparative analysis starts from the assumption that these countries should also possess similar criminal justice policies (Johnson and Zimring 2009). In this article, we use the comparative method to explain why PNG and Tonga long pursued a different death penalty trajectory to their regional neighbours, bearing in mind that this cross-sectional analysis has only nine 'data points' and, at any one specific point in time, some nations' death penalty policies (such as PNG's) are still evolving. As we make clear in Parts II and III below, none of the traditional social science explanations for death penalty retention seem to identify PNG and Tonga clearly as regional outliers. Drawing extensively from legislative debates and other political statements, in Part III, we suggest two novel explanations for PNG and Tonga's more durable retention: PNG's ongoing law and order crisis and the status of the Tongan monarchy. Finally, in Part IV, we explore the implications of our findings for future efforts towards promoting and maintaining abolition in the South Pacific and further afield.

II. The Comparative Method and the Death Penalty in the South Pacific
The death penalty in the South Pacific is worthy of study for two reasons. First, despite no regional executions having taken place since a triple hanging in Tonga in 1982, the question is not purely an academic one. PNG sentenced prisoners to death as recently as 2018, and its political leaders routinely called for a resumption of executions over a 67-year hiatus after 1954 (Cornell Center on the Death Penalty Worldwide 2021b; Hands Off Cain 2021a). Similarly, in 2021, the Tongan legislature debated the expansion of the death penalty to drug trafficking offences, although that proposal was ultimately rejected (Fennell 2021). Proposals for reinstatement have also recently arisen in Solomon Islands and Kiribati (Amnesty International 2014; Asia Pacific Report 2021), each of which has failed to ratify the International Covenant on Civil and Political Rights (ICCPR) and its Second Optional Protocol, meaning that a resumption of death sentences faces no formal legal obstacles.
Second, the last 40 years have seen global momentum towards total abolition of the death penalty, especially in countries that have recently experienced a democratic transition or that have not carried out executions in a long time. In 1948, only 15 countries had abolished the death penalty in law, mostly in Latin America. That number increased to 52 in 1988, 89 in 2001 and 108 in 2020 (Amnesty International 2021; Hood and Hoyle 2017). As more countries abolish the death penalty, the normative case against capital punishment strengthens, reflecting state practice in customary international law and increasing political pressure on the holdout states (Hood and Hoyle 2017). Moreover, as regional hegemons and former administrative powers, Australia and New Zealand play a unique role in promoting death penalty abolition in the South Pacific. The puzzle is why, under these conditions, PNG and Tonga resisted the global and regional trends for so long.

The Abolition/Retention Literature
Comparative study of death penalty abolition has generated a sizable social science literature, especially since the most recent 'wave' of abolition following the Cold War and the period of democratic transition in the 1990s. This research has sought to identify the conditions and processes that led to abolition in different political jurisdictions, including in the Asia-Pacific region (e.g., Anckar 2004Anckar , 2014Bae 2007;Boulanger and Sarat 2005;Futamura 2014;Garland 2014;Greenberg and West 2003;Hobson 2014;Hood 1989Hood , 2002Hood and Hoyle 2008Johnson and Zimring 2009;McGann and Sandholtz 2012;Miethe, Lu and Deibert 2005;Neumayer 2008aNeumayer , 2008bRuddell and Urbina 2004;Zimring 2013;). Although these studies have generally portrayed death penalty practice as a dichotomy (i.e., retention versus abolition), the decline of the death penalty around the world is also visible in the decreasing number of capital offences, death sentences passed and number of executions carried out.
These social science studies have identified different variables that explain abolition and retention. The first of these is political structure. Generally, unitary systems of government promote abolition more than do federal systems, democratic governments are more likely abolitionist than authoritarian systems, and left-leaning political parties tend to abolish before right-leaning parties. Second, economic development appears to be correlated with abolition. With several important exceptions (United States, Japan, Qatar and Singapore among them), retentionist nations tend to be at a lower stage of economic development. Third, specific colonial legacies correlate with death penalty retention-former colonies are more likely to be retentionist, as are countries with common law legal systems and a British colonial legacy. Finally, religion and ethnic makeup also play a role. Countries with majority Christian (especially Roman Catholic) populations are more likely to abolish the death penalty than countries with Muslim or Buddhist majorities. Similarly, ethnic homogeneity is correlated with death penalty abolition more than ethnically diverse countries or countries with a legacy of slavery.
Another common theme from the death penalty abolition literature is that domestic and local processes contribute to abolition. The transition of the death penalty from a domestic criminal justice issue to an international human rights concern preceded abolition in many countries. In addition, abolition exhibits regional contagion effects. Abolition of the death penalty in a country's near neighbours increases incentives to abolish, as compared to entirely retentionist regions. Like in Western Europe (Ancel 1962;Dudai 2021), South Pacific nations have tended to abolish gradually, first reducing the number of capital offences and abolishing the mandatory death penalty and then restricting the scope of death penalty practices and the frequency of death sentences and executions. States that transform from enthusiastic executioners to total abolitionists within a short period of time likely underwent a sudden authoritarian-democratic political transition. Oft-cited examples include South Africa (last execution 1989(last execution , abolished 1995(last execution ), Ukraine (last execution 1997(last execution , abolished 2000, Cambodia (last execution and abolition 1989) and Timor-Leste (independence and abolition in 2002). Political transitions are correlated with death penalty abolition because they break down existing networks and institutions and realign countries' international priorities. However, South Pacific nations achieved independence and democratic government peacefully; therefore, most have not experienced a dramatic political transition along these lines.

Explaining Death Penalty Abolition in the South Pacific
Most South Pacific states discontinued or abolished capital punishment for all crimes even before the emergence of death penalty abolition as a central concern of the international human rights movement in the late 1970s. Kiribati and Tuvalu abolished as the same British colony in 1965, Solomon Islands abolished in 1966, and Vanuatu abolished in 1973, while de facto abolition occurred in Fiji (1965), Nauru (before 1932) and PNG (1964) well before capital punishment became an important international human rights issue.
Disuse of the death penalty in these countries is best explained not by the emergence of the international human rights movement, by domestic political leaders or by regional contagion, but rather as a form of colonial legal transplantation. In the United Kingdom, the death penalty was suspended for murder in 1965 and abolished in 1969, 4 and death sentences in Hong Kong, Seychelles and other non-self-governing British colonies were usually commuted after that point. In Kiribati, Tuvalu, Solomon Islands and Vanuatu, the death penalty was abolished via revision of their respective penal codes prior to independence. De facto abolition in both PNG and Nauru, where Australian administrative influence was strongest, fit within the temporal ranges of abolition in Australian jurisdictions (1973 at the federal level, with the last state executions carried out between Queensland in 1913 andVictoria in 1967). Decolonisation of the South Pacific temporally coincided with death penalty reductionism in the English-speaking world and particularly the suspension or abolition of the death penalty in the United Kingdom and Australia.
While de jure abolition of the death penalty in the South Pacific generally followed the pattern in the relevant colonial or trusteeship powers, exceptions remain. Samoa did not abolish until 2004, Nauru until 2016, and Fiji for military offences until 2015. Yet, in each of these three cases, death penalty reductionism was also influenced strongly by the United Kingdom, Australia or New Zealand.
In Fiji, following the lead of the United Kingdom, executions were suspended by the Legislative Council in 1965, before abolition for murder in 1979 (Mitchell 2019 Little over a year after the decision to adopt the Griffith Code in Nauru, Queensland amended its own code to abolish the death penalty altogether (Finnane, this volume). Rather than amending the capital provisions to follow the position in Queensland, the trusteeship authority and later independent governments of Nauru seemingly waited to replace the entire criminal code, which by 2016 had become badly outdated (Hands Off Cain 2021b). 5 Like Nauru, Samoa's death penalty retention lasted significantly longer than that of its trusteeship power, New Zealand, which abolished for a second time in 1961. Although Samoa achieved independence in 1962, it did not follow New Zealand's lead and instead retained the death penalty for murder and treason. Death sentences were imposed until the 2000s, alongside periodic commutations (Death Penalty Information Center 2004; Hands Off Cain 2021c). We observe that in both Nauru and Samoa, de jure death penalty abolition did not automatically follow a political transition, although both nations remained steadfastly execution-free after independence in the manner of their erstwhile administrators (Australia and New Zealand, respectively).

Discerning Intent to Retain or Abolish
Here, we add to our preceding comparative analysis of the death penalty in the South Pacific by consulting the available legislative debates to discern countries' intent as to why they retained or abolished the death penalty. We supplement these legislative debates with the statements of political leaders and state reports to the Universal Periodic Review process at the UN Human Rights Council, as these are also government-prepared statements that make normative claims.
Legislative debates can be useful to determine legislators' intentions as to why they agreed to certain policies and can assist with interpretations of statutory law (Todd 2006). In the present case, the legislative debates reveal the specific concerns of legislators at the time they voted on death penalty retention or abolition. These debates often present multiple and conflicting views but can be analysed for overarching themes-for instance, regarding whether human rights-based arguments or crime control arguments prevailed in the legislative discussion. The legislative debates we analysed for this article are summarised in Table 2. The Draft Hansard of the National Parliament of PNG contains the first reading of the Criminal Code (Amendment) Bill 1990, which reinstated the death penalty for premeditated murder following its abolition in 1974. According to the legislative debates, legislators were keen to resist foreign advocacy to abolish and sought to emphasise the deterrence value of the death penalty for crime control purposes. Legislators also referenced 'payback' killings and tribal fights-long recognised as crime concerns in PNG. Several even made cultural appeals, explaining that tribal communities have different ways of dealing with murder, and identified the country's ethnic diversity as one cause of homicide and retaliation (National Parliament of PNG, 17 July 1991). Supporters of death penalty reinstatement even used the Bible to justify executions. Opposition to the Bill emphasised global trends away from capital punishment and appealed to civilised values and Christianity (National Parliament of PNG, 28 August 1991). Nevertheless, overall, the debates tended to emphasise crime control and penal populism in support of reinstatement. For instance, prime minister Rabbie Namaliu referenced support for crime victims and high crime rates in his opening speech (National Parliament of PNG, 22 May 1991).
In the Fiji legislative debates of 2002, the primary concern of legislators was the 19 May 2000 coup event and the president's commutation of the mandatory death sentence for treason for the principal coup plotter, George Speight. During the House of Representatives debates on Fiji's Penal Code (Amendment) Bill 2002, some legislators frequently referred to Christian values and biblical references to capital punishment, while others viewed the death penalty as a colonial holdover. The legislative debate also included discussion as to whether to use life imprisonment with or without parole as the replacement for the death penalty. The Attorney-General referenced international human rights norms, including UN resolutions on cruel and degrading punishment and the right to life, in his speech to advance legislation abolishing the death penalty for treason. However, the bulk of the debate concerned the 2000 coup attempt. While some legislators were sympathetic to the coup, others were victimised personally by the coup plotters-by being taken hostage (Parliament of Fiji, House of Representatives 2002). Eventually, the House of Representatives unanimously supported the legislation, aiming to ensure that no future acts of treason could result in death sentences.
In the debates that followed in the Senate of Fiji, several senators criticised the timing of the Bill so shortly after the commutation of George Speight's death sentence. However, the senators repeated many of the same themes as in the lower house. Several referenced international human rights concerns, pressure from other Commonwealth countries and the problem of wrongful convictions. Several senators referenced the failure of the death penalty for treason to deter the coup attempt, while others made biblical or other religious references to encourage reconciliation and forgiveness. Ratu George Cakobau Jr., an appointed chiefly member of the Senate, referenced 'our custom' that 'taught us to embrace the principle of forgiveness and reconciliation through the traditional presentation of matanigasu [formal apology]' (Parliament of Fiji, The Senate 2002: 491). However, in the debates, references to Christianity were more common than references to traditional customs.
In the February 2015 debate over the Republic of Fiji Military Forces (Amendment) Bill 2015 in the country's new unicameral legislature, Attorney-General Aiyaz Sayed-Khaiyum explained that the origins of the death penalty for military crimes in Fijian law was the United Kingdom Army Act 1955, which had since been repealed in the United Kingdom itself. He also explained that at the country's UN Universal Periodic Review process, the government had made an undertaking to remove any reference to the death penalty in Fijian law in favour of life imprisonment. He pleaded that 'we do not want to be short on our undertakings' when reporting to Geneva on progress following the review process. The Leader of the Opposition responded to the Attorney-General, noting that 'you do not have to comply with whatever those people say at the United Nations'. The opposition legislator explained that the United Kingdom only repealed the death penalty for military crimes when it stopped having coups-when it no longer needed the deterrence value of the punishment. The vote to proceed on the Bill passed with a 31 to 14 majority (Parliament of the Republic of Fiji 2015: 680-682).
In the January 2004 legislative debates in Samoa, prime minister Tuilaepa Sailele made references to international human rights standards and the position of the death penalty in other countries. He also made a Christian case against the death penalty, as did several other legislators. One legislator, Leao Talalelei, explained that the death penalty 'was brought by people who were in charge of Samoa at the time', but the country has 'matured enough now to make its own laws that are in line with our own culture and traditions' (Parliament of the Independent State of Samoa, 15 January 2004: 786). Another common theme was the role of the head of state in deciding questions of life and death through executive clemency. One legislator wanted to abolish the death penalty 'to assist His Highness, O le Ao o le Malo, by taking away this burden from him' (Parliament of the Independent State of Samoa, 15 January 2004: 794). In contrast to PNG, legislators in Samoa rejected the idea that the death penalty deterred crime because having the death penalty on the books for 52 years had not reduced the homicide rate. Further, the death penalty was 'New Zealand's law', but 'New Zealand repealed the death penalty in its laws while we still live with this law' (Parliament of the Independent State of Samoa 15 January 2004: 797). According to Amnesty International (2004), the Samoan legislative debate was preceded by a notorious rape and murder of a five-year-old girl in April 2003, which led to public demands for executions. In these circumstances, the tone of the debate provides a marked contrast to the 1991 debate in PNG, where penal populism figured much more strongly in the legislature.
In Nauru, the Crimes Bill 2016 was a comprehensive rewrite of Nauru's century-old penal code, which included the repeal of archaic sex offences, the criminalisation of slavery and child labour and the decriminalisation of abortion and suicide (Amnesty International 2016; Hands Off Cain 2022a). In a speech on 12 May 2016, Member of Parliament David Adeang referenced 'international human rights standards' and the country's 'international legal obligations under core human rights treaties' in his introduction to the Bill (Government of the Republic of Nauru 2016: 1). Neither Adeang's speech nor the Bill's explanatory memorandum referenced the death penalty explicitly, reflecting the comprehensive nature of the penal code revision and the fact that capital punishment had long before fallen into disuse in Nauru. A year earlier, Nauru submitted a report to the UN Human Rights Council as part of the Universal Periodic Review process. In this report, the government of Nauru expressed interest in amending the country's 1968 Constitution to remove references to the death penalty. The government also described the proposed new Crimes Act 2016 as a victory for human rights, emphasising the comprehensive nature of the new law, such as new provisions criminalising sexual offences and violence against women (Government of Nauru 2015).
The Tongan legislative debates from August 2021 contain both new arguments and points in common with the debates of other South Pacific nations. First, legislators made frequent references to the Bible and to Christianity, reflecting the views of the religious majority in the country. Second, the posture of the United States as a retentionist nation was raised several times. Third, legislators invoked the 'deterrence' value of the death penalty, for both drug use and murder. Finally, as with the focus on violent crimes within the 1991 debates in PNG, the debate over the Illicit Drugs Control (Amendment) Bill 2021 emphasised the scourge of drug use and the failure of the government to control what was portrayed as a national crisis. Ultimately, however, the expansion of the death penalty to cover the most serious drug offences was rejected by a majority of Tongan legislators (Fennell 2021).
Most recently, on 20 January 2022, members of the National Parliament of PNG debated the Criminal Code (Amendment) Bill to abolish the death penalty for murder again. Unlike the 1991 debates, the legislators cited Christian opposition to the death penalty, the lack of deterrence value and the risk of wrongful convictions. The existence of the death penalty in the United States also played an ambiguous role in the death penalty debates as both a justification for continued capital punishment and, owing to wrongful convictions and racial disparities, an argument in favour of abolition. When the Minister for Justice referenced the country's 'development partners' who were prepared to give 'additional support' upon abolition of the death penalty, another member contested this view because PNG was a 'sovereign state' that 'should not be influenced by outsiders like development partners' (National Parliament of PNG, 20 January 2022: 49-50). The 2022 PNG legislative debates have similarities to the 2002 and 2015 debates in Fiji and the 2004 Samoan debate: each emphasised the forgiveness values of Christianity rather than the retributive values, expressed concern about wrongful convictions in the United States as opposed to seeing the United States as a model and inverted the 'imperialist' argument by pointing out that the death penalty itself was a colonial imposition, rather than seeing the abolitionist movement as foreign or imposed. Overall, PNG's 2022 legislative debate provided the impression that the death penalty had not helped to solve the country's law and order crisis. By the end of the parliamentary session, PNG's legislators had abolished the death penalty for all crimes, replacing it with sentences of life without parole or life with the possibility of parole after 30 years (Asia Pacific Report 2022).

III. Explanations for Papua New Guinea's and Tonga's Retention
As Table 3 illustrates, none of the established explanations for death penalty abolition, either individually or collectively, explain why PNG and Tonga retained the death penalty for longer than other South Pacific nations. Each of the following types of characteristics-political institutions and processes, democratic instability and transition, economic development, the legal system and colonial history-failed to distinguish PNG and Tonga as outliers from their abolitionist neighbours. Neither did PNG and Tonga appear to be affected by regional contagion from countries that had similar cultural attributes or common membership in international human rights treaties and organisations.

Table 3. South Pacific nations and the abolition/retention literature
In this section, we suggest two novel explanations for delayed movement towards death penalty abolition in the two South Pacific holdout states. The first is Tonga's strong monarchical governance tradition, and the second is PNG's long-running law and order crisis, neither of which are covered in detail in the existing social science literature.

The Tongan Monarchy
Tonga is the only hereditary monarchy of the nine South Pacific countries under study, excluding Commonwealth nations that maintain the British monarch as a ceremonial head of state (i.e., PNG, Solomon Islands and Tuvalu). The monarchy is central to daily life in Tonga, and we argue that the retention of the death penalty plays a symbolic role in reasserting the monarch's sovereign power over life and death. The institution of royal pardon, in which the Tongan monarch must receive advice from an advisory body known as the Tongan Privy Council, is emblematic of this power (Criminal Offences Act 1926, art 33(2)-(3)). For most of its history, this provided a weak check on prerogative power. Since 1875 and until a period of reform in the mid-2000s, the Tongan Privy Council was appointed by the monarch and comprised the Cabinet ministers, usually nobles who served at his pleasure and held appointed seats in the legislature-their role was purely advisory (Powles 2014). Before Tonga became de facto abolitionist in the 1990s, the king routinely considered death penalty cases for commutation to life imprisonment (Amnesty International 1979).
Significantly, royal pardons in death penalty cases have long been regarded as a means for hereditary rulers to display authority, benevolence and religiosity, even in constitutionally constrained monarchies (Pascoe 2019). We observe that the death penalty, as one of the trappings of sovereignty, plays a similar symbolic role in Tonga as in other traditional monarchies, such as those in Thailand, Lesotho, Morocco and Brunei. Tonga's system of government even approached that of an 'absolute monarchy' until 2006, when King George Tupou V took power. The new king then withdrew from active decision-making and devolved power to his ministers. When he died in 2012, his successor played no role in Cabinet decisions. In 2010, the Legislative Assembly passed a series of constitutional amendments that codified this power shift and constrained the king's power (Bogle 2019;Powles 2014). While scoring poorly in non-government organisation Freedom House's ratings on political representation, Tonga was nonetheless categorised as 'partly free' within the ranking on democratic governance since its inception 1973, alongside several abolitionist Pacific Island nations during different points in their history (Fiji, Solomon Islands and Vanuatu; Freedom House 2021). Since 2012, Freedom House has categorised Tonga as a 'free' jurisdiction (see Table 3). It is evidently not authoritarianism, so much as Tonga's precise system of government, that is the primary explanation for the country's ongoing death penalty retention.
Political comparison between Tonga and Samoa is instructive. Although they share a common Polynesian culture, Protestant Christian faith and history of European colonialism, Samoa possesses a ceremonial head of state (the O le Ao o le Malo) appointed by the Legislative Assembly. In practice, the Samoan head of state rotates among four chiefly dynasties, although this is not a constitutional requirement. Therefore, the country can be formally classified as a republic rather than a constitutional monarchy. The Samoan legislative debates preceding the country's death penalty abolition in January 2004 reflect some of these differences. Member of Parliament Aeau Peniamina of Falealupo constituency referenced Christian religious values in his speech, before adding, 'I wish to congratulate the Government for this Bill to avoid having to present His Highness O le Ao o le Malo and lawmakers with the difficult task of discussing the laws of this nature, but to give it to God, who alone has the authority' (Parliament of the Independent State of Samoa, 15 January 2004: 785). This view that God, rather than the monarch, is the highest authority and that the elected branch of government should avoid leaving difficult choices like death sentence commutation to the ceremonial head of state resembles republican values more than monarchical ones.
Unlike Samoa, Tonga has the unique distinction in the Pacific of retaining its indigenous, hereditary form of government despite becoming a British Protectorate in 1900. Tongans are intensely proud that their country has always remained self-governing, unlike its Pacific neighbours (Campbell 2011). Tongans believe the monarch is a descendant of the sky god Tangaloa, with the current lineage going back at least 25 generations. The monarchy is supplemented with 33 hereditary noble titles, which confer significant privileges and responsibilities in Tongan society. Following the most recent execution in 1982, Tonga's nobility and appointed member-dominated Legislative Assembly thoroughly debated abolition but decided to keep the current system in place (Robinson 2015). Nevertheless, the democratic opening in Tonga after 2006 has provided an opportunity to challenge the death penalty status quo. In a statement at the UN Human Rights Council during its 2008 Universal Periodic Review, 6 the government of Tonga stated regarding the death penalty: The current determination for constitutional and political reform and the consideration of other international human rights instruments may offer further opportunities for discussion and debate on this issue. (Parliamentarians for Global Action 2022) In August 2021, the Tongan government reintroduced in the Legislative Assembly a proposal to punish drug trafficking with the death penalty. However, although the legislative debates contained penal populist references to the scourge of the drug crisis, a majority of legislators in September 2021 voted down the death penalty in favour of a life imprisonment penalty (Fennell 2021). We believe that this recent rejection of the death penalty for drug trafficking in the Tongan legislature is another indication of a thaw in the political establishment's support for death penalty retention, commensurate with Tonga's shift away from a powerful monarchy towards a more constitutionally constrained governance model.

Papua New Guinea's Law and Order Crisis
By contrast, PNG's death penalty appears to be related to the country's notoriously high levels of violent crime. Based on comparative figures from the UN Office on Drugs and Crime (2018), PNG's number of intentional homicide victims exceeds 500 per year, more than twice that of Australia and 20 times more than any other Pacific Island jurisdiction. The high rate of homicide is partly explicable by the tribal custom of 'payback', although robbery and rape are also common (Browne et al. 2006; Hands Off Cain 2021a).
As the legislative debates summarised above suggest, the high rate of violent crime was the primary legislative justification for reintroducing the death penalty for murder in 1991. Previously, PNG had abolished the death penalty for murder in 1974, during the period of self-government that preceded independence, contemporaneously with abolition at the federal level in Australia. However, the country's law and order situation has since deteriorated. In 2013, the new capital offences of aggravated rape, robbery and sorcery-based killings were added by parliament. These changes followed several high-profile crimes, including the rape and murder of foreign residents and visitors and the gruesome killing of village women for suspected sorcery (Cornell Center on the Death Penalty Worldwide 2021b). The UN has estimated that approximately 200 sorcery-based murders occur annually (Harson 2022).
In a fractured and megadiverse country of remote communities, 800 different languages, limited enforcement capacity and customary practices that counter weak state authority, reimposing the death penalty for murder and then introducing it for robbery and rape were symbolic steps that PNG's elected government could take to show the public that it takes the law and order crisis seriously (Amnesty International 2004;Weisbrot 1985). In this sense, PNG's government is far from alone. During recent decades, other 'weak states' have reintroduced or expanded capital punishment to shore up domestic support in the face of seemingly insurmountable violent crime problems. Examples include the Philippines (1993), Iraq (2004), Liberia (2008) and Chad (2015).
As with Fiji, extraordinary crimes (i.e., treason, piracy and attempted piracy) remained on the books as unused capital offences after PNG's independence. This may have helped to normalise capital punishment in public and political opinion, smoothing the path for reimposition of the death penalty for 'ordinary' crimes. Although no execution has been carried out since 1954, PNG has a robust tradition of executive clemency in death penalty cases (Amnesty International 2004), which may prolong the existence of the death penalty on paper. A sentence that is judicially authorised but not carried out may have a presumed deterrence value while avoiding the political costs of actual executions (Burton-Bradley 1990). In PNG, the National Executive Council, composed of the prime minister and other Cabinet ministers, makes clemency decisions on the advice of an advisory mercy committee and then instructs the governor-general, who cannot refuse the Council's advice (Novak 2016). This is different to Tonga, where the monarch receives advice from his ministers but retains the ultimate authority.
Bruce Ottley, a former Port Moresby magistrate, recently reflected on the crime and law enforcement problems in PNG, noting the 'limited personnel, financial and technical resources of the country's police' and the 'understaffed and underfunded' public prosecutor's office, public solicitor and court system (Ottley 2016: 83). The 1991 PNG legislative debates accentuated some of these concerns. As noted above, penal populism strongly characterised the debates, with the country's law and order crisis at the forefront of the discussion. While the legislative debates in other South Pacific countries emphasised the colonial origins of the death penalty, the PNG legislature in the 1991 debate that led to reinstatement of the death penalty completely inverted the neo-colonialist characterisation, instead portraying the death penalty abolitionist movement as a foreign imposition, divorced from the country's policy needs and cultural values. However, when the National Parliament abolished the death penalty in January 2022, the script was reversed-legislators instead cited the death penalty's colonial origins.

IV. Implications for Abolition
These unexpected explanations for death penalty retention in PNG and Tonga have implications for abolition in the South Pacific and for the abolitionist movement more generally. In PNG, if the primary reason for retention was the high violent crime rate, government policies and outside assistance to control crime without resorting to executions become essential prerequisites to prevent future reimposition (Sakai 2022). As the legislative debates from 1991 also suggest, in PNG's political environment, human rights-based or religious arguments against reimposition may not be as effective as tangible assistance for policing, prisons and courts or, more controversially, customary justice initiatives (Larcom 2015;Ottley 2016). The failure of state-based responses to violent crime since independence, including several states of emergency imposed since 1979 due to gang-based violence, has generated public and political support for capital punishment (Dinnen 1997;May 2004).
In Tonga, the monarchy's status as a central justification for retaining capital punishment suggests two potential avenues for reform. First, with the gradual loosening of royal power following political reforms in the 2000s and 2010s, Tonga is moving towards representative democracy, an essential ingredient for abolition in other parts of the world. Tonga's elected political leaders may feel increasingly empowered to take steps towards de jure abolition without the status of the monarchy interfering. Second, campaigners for abolition may find it helpful to point to other monarchies that do not retain capital punishment. For Tonga, among the most important are the Vatican (Tonga is approximately 15 per cent Catholic) and the abolitionist monarchies in Asia (Bhutan and Cambodia). Samoa, a neighbouring country with a ceremonial head of state who performs similar functions to a constitutional monarch, is also a helpful model. Counterparts in these jurisdictions may be better placed to encourage reform in Tonga compared to abolitionists in the United Kingdom, Australia and New Zealand, where perceived neocolonialism could generate local opposition.
We now turn to the broader practical and academic implications of this research. In this article, we have demonstrated that death penalty abolition in the South Pacific, either de facto or de jure, was historically contingent on British colonialism and its legacies. This is an interesting finding considering the existing academic literature's suggestion that nations with a history of colonial domination and a common law legal system are more likely to retain the death penalty than those without (Anckar 2014(Anckar , 2004Neumayer 2008a). For the abolition movement, the process of decolonisation and the end of UN-sponsored trusteeship fortuitously followed recent death penalty abolition in the United Kingdom, Australia and New Zealand. Accordingly, we propose that peaceful decolonisation can be considered another 'historical turning point' placing retentionist states on a path to abolition, analogous to newly democratised states emerging from a history of conflict (Bae 2007;Futamura 2014;Neumayer 2008a).
The notable counterexample here is that of the Commonwealth Caribbean, where some former British colonies, such as Belize (1981), Antigua andBarbuda (1981), andSt Kitts andNevis (1983), became independent relatively late and through a peaceful process. Each of these jurisdictions still retains capital punishment. One difference with the Caribbean is that the former British colonies in that region were self-governing long before independence, which means they may not have benefited directly from abolition in the United Kingdom itself. In addition, Commonwealth Caribbean countries have high crime rates and populist politics that are more reminiscent of PNG than other South Pacific states. The legislative debates in Samoa and Fiji both recognised the death penalty as a colonial import, originally enacted in penal codes that were drafted by others. Viewing the death penalty as a colonial imposition is an important riposte to the idea, prevalent in the Commonwealth Caribbean, that death penalty abolition is a neo-colonial strategy (Harrington 2004). Respectfully employed, this insight may prove valuable to abolitionists from Australia, New Zealand and the United Kingdom who are advocating in postcolonial settings elsewhere in the Asia-Pacific region.
One final observation from the South Pacific (relevant to PNG and, to some extent, Fiji) is the underestimated danger of retaining capital punishment solely for 'extraordinary' crimes such as treason, piracy and military offences. Some 'extraordinary' crimes, such as wartime offences or genocide, are relatively rare, whereas prosecutions for treason and serious military crimes tend to be more common. Yet, even symbolic retention for crimes that are very rarely prosecuted may normalise capital punishment in the legal system and popular imagination, making reinstatement of the death penalty for 'ordinary' crimes more likely by future governments experiencing law and order crises (Dudai 2021). PNG and Fiji's recent histories remind activists not to overlook 'abolitionist for ordinary crimes only' states when advocating against capital punishment. The historical evidence shows that once a nation becomes abolitionist for all crimes in law, it is very difficult to resume executions in the future (Hood and Hoyle 2015). 7 Most South Pacific states have already reached this threshold, leaving the region on the remarkable cusp of becoming a death penalty free zone. TOSC 31, Webster CJ stated that the death penalty could only be imposed in 'one of the rarest of rare cases where the alternative option of life imprisonment is unquestionably foreclosed'. Most recently, the death penalty was 'seriously considered' but not imposed by Whitten LCJ following the murder of a prominent human rights activist in May 2021 (Radio New Zealand 2021). 3 Having not passed a single death sentence for a decade (Chisholm 2020), PNG first abolished the death penalty for murder by passing a new penal code in 1974, which was drafted by the PNG Department of Justice's Law Reform Committee during the brief period of selfgovernment before independence (1973)(1974)(1975) (Chalmers, Weisbrot and Andrew 1985). However, after independence, the death penalty was still retained on paper for treason, piracy and attempted piracy. There were unsuccessful legislative attempts to restore the death penalty for murder in 1979 and for gang rape in 1984 (Weisbrot 1985). In 1991, parliamentarians succeeded in restoring it for murder and expanded it to rape, robbery and sorcery killings in 2013. One reason that no executions were carried out after 1991 is the lack of a functioning execution mechanism (Hands Off Cain 2021a; The National 2020). 4 Except in Northern Ireland, which abolished for murder in 1973. 5 See Finnane (2022, this volume) on the apparent confusion by Australian-based administrators in 1939 over whether Nauru's laws still authorised the death penalty. The outbreak of World War II then thwarted the administration's plans for abolition in Nauru. 6 There are other signs of abolitionist progress on the diplomatic front. In 2014, 2016 and 2018, Tonga abstained in the UN General

Correspondence
Assembly's biennial anti-death penalty resolution. Tonga had previously voted 'no' and signed the associated note verbale condemning the resolution in 2007, 2008, 2010 and 2012. 7 A further means of guarding against executions under future 'law and order' governments is to become a party to the ICCPR and its Second Optional Protocol, thereby creating binding international law obligations not to reinstate capital punishment or to execute (Schabas 2002).