Implementation of the Convention against Torture in Taiwan: Filling the Gap in the International Struggle against Torture?

Abstract

The universal system of protection against torture established by the United Nations (UN) is confronted with daily incidents of torture and other cruelties in all regions of the world. Moreover, despite the ratification of UN treaties, most governments lack a genuine commitment to address these abuses. In contrast, the anti-torture safeguards under the UN Convention against Torture and its Optional Protocol are currently being implemented in Taiwan, even though Taiwan is not part of the UN and cannot participate in the international human rights dialogue. The process of incorporation of UN anti-torture documents by a non-UN member is all but easy. This article shows, however, that commitment to combat torture goes beyond the UN treaty system and might be a welcome contribution towards the current debate on fostering compliance with human rights treaties both in Asia-Pacific and around the globe.

Nearly forty years ago, the United Nations Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (“Convention” or “UNCAT”)¹ established a normative framework based on the prohibition and prevention of torture and other cruel, inhuman or degrading treatment or punishment (“ill-treatment”). In the light of a universal consensus on the need to combat torture,² the prohibition of torture has developed into a fundamental principle of international customary law and a peremptory norm of general international law (jus cogens).³

Given this non-derogable and absolute character and universal acceptance, some may assume that torture and other ill-treatment no longer exist in the contemporary world. However, as Manfred Nowak, the former United Nations Special Rapporteur on Torture, has clarified, the contrary is accurate. Torture continues to be regularly and systematically practised in a vast majority of States.⁴ This reality has been confirmed by Nowak's successors, Juan E. Mendéz⁵ and Nils Melzer,⁶ and nothing indicates that the situation is going to improve in the near future. In a recent report, Melzer has observed that governments worldwide lack a credible commitment to the absolute and universal prohibition of torture and ill-treatment, and that nine of the ten allegations of torture have been entirely ignored by governments in all regions of the world.⁷ In addition, only a minority of States Parties fully comply with its core obligations, such as the criminalization of torture under Article 4 of the Convention and a duty to establish universal jurisdiction based on Articles 5–9 of the Convention.⁸ The discrepancy between the prohibition of torture in the books and its prevalence in practice has been characteristically described by Lutz Oette as one of “the most paradoxical norms of international human rights law.”⁹

A plausible explanation of Oette's paradox derives from the strong international emphasis on ratification of the Convention¹⁰ on one hand, and the reluctance of the States Parties to adhere to its provisions on the other. For many countries, ratification is regarded as gaining legitimacy in the international community¹¹ and avoiding being labelled as “pariah States”.¹² In such “window-dressing” ratification,¹³ governments have a minimal interest to implement treaty provisions at the national level.¹⁴ Some scholars note that “disguised” ratification could result in side effects such as rejection of any allegations of torture, obstruction of investigation,¹⁵ or even advancement of repression.¹⁶

Regrettably, the Committee against Torture (“CAT”), a monitoring body for the implementation of the Convention, is regarded by many as having a minimal impact on treaty compliance, and an increasing number of scholars even consider it defective and toothless.¹⁷ Against this background, it could be assumed that the “paradoxical norm” of torture only illustrates general deficiencies of international enforcement mechanisms, which evidently have a minimal demonstrable impact on national compliance with the treaty.¹⁸ It is therefore not surprising that calls become stronger for both change in world politics¹⁹ and the revision of the United Nations (“UN”) treaty system as such. After all, the UN recognizes the unsatisfactory effectiveness of the treaty system and has embarked on a process of reform.²⁰

It may be believed that to enhance the impact of the Convention, it is essential to ensure that local actors at the national level comply with the treaty,²¹ and that such an approach does not depend on international enforcement.²² Byrnes has underscored that “the battles for human rights have to be won at a national level”.²³ His remark confirms the earlier argument put forth by Heyns and Viljoen who have insisted that it is necessary to “harness the treaty system to domestic forces” in order to ensure the practical realization of human rights.²⁴ Additionally, Chen has suggested that neither the international nor domestic constituencies should be overestimated, and the correct approach is to view the treaty implementation as “a joint project co-owned by local as well as global actors”.²⁵

This paper aims to enhance the ongoing debate on treaty compliance by examining the implementation of the Convention situated outside the UN treaty system. Focus is given to the Republic of China (Taiwan), which is not a UN member State and is thus neither obliged nor politically urged to observe human rights obligations under the treaty regime. Surprisingly, the Taiwan government (“the Executive Yuan”) has recently adopted a bill on the implementation of the UNCAT and its Optional Protocol (“OPCAT”)²⁶, including the creation of the National Preventive Mechanism (“NPM”).²⁷ In addition, multiple stakeholders have already taken steps to ensure the implementation of these documents in practice.

It has to be emphasized in the first place that as a non-recognized State,²⁸ Taiwan is formally excluded from the international human rights framework. It is prevented from having any official dialogue with the UN, including, among other things, a deposition of ratification documents at the UN General Assembly.²⁹ As a result, over 23 million people living in Taiwan, including residents and foreigners, remain beyond the reach of UN human rights protection.

Given this legal vacuum, whether and to what extent these international human rights obligations will be incorporated into Taiwan's domestic law solely depends on the willingness of the local actors to do so. This wide discretion naturally increases risks that the local government decides to grant the protection only to some ‘favourable’ human rights and to adopt an interpretation that may be contrary to established human rights discourse. This quandary provides a unique opportunity to examine the motivation of local actors to adhere to the UN human rights treaties as well as to assess their impact on national law outside the UN treaty system. This analysis further responds to the calls for systematic studies on the effects of international treaties outside the treaty system.³⁰

The specific position of Taiwan in the international community sets ambitious goals, such as to examine whether the scope of the prohibition of torture is the universal paradigm shared equally by the UN members and those ‘standing outside’; whether the normative framework of the Convention could ‘live independently’ outside the UN treaty system or whether de facto compliance rather than formal ratification is the direction in which the UN treaties should move. However, the aim of this article is more modest, as it seeks to investigate the legal way in which the UNCAT is going to be ratified (incorporated in law)³¹ by non-member State, and the advantages and challenges of this uncommon process. Particular attention is paid to the implementation of the obligation to criminalize torture and to establish universal jurisdiction. Nevertheless, such a narrow focus does not give up on the ambition to look beyond this case study of Taiwan, which could provide some directions for future research.

The paper is fragmented into four principal parts. The first Part begins by reviewing the concept of torture and the specific character of the Convention against Torture. It argues that although the Convention has an invaluable advantage in establishing the concept of torture in Taiwan by setting a clear definition of torture, which is deemed the first step in torture prevention, its implementation faces challenges requiring rigorous examination of the State's implementation capacity.

The following Part explores the ratification process of the UNCAT in Taiwan. Building on a path-dependence model for ratifying UN treaties developed in Taiwan over the years, it examines the current Implementation Act. It also highlights the strengths and weaknesses of this model and underscores future challenges.

The paper follows with an assessment of a State's core obligations, namely, to criminalize torture and to establish universal jurisdiction for that crime. This article maintains that these commitments are pivotal Taiwanese government's duties under the Convention and a litmus test of the implementation process. This Part also reveals the pitfalls of Taiwan's international position and its impact on the fight against torture in Asia and the world.

The final Part summarizes the benefits and challenges of UNCAT's ratification process in Taiwan and outlines a direction for improvement of both the local legislative endeavour and the international struggle against torture.

I. Concept of Prohibition of Torture, Benefits and Challenges of Implementation

The Convention against Torture is undoubtedly one of the most challenging human rights treaties. Unlike catalogue-like treaties³² or subject-related treaties³³, the Convention is generally understood as a specialized “single-issue” treaty³⁴, which sets the aim to strengthen the prohibition against torture.³⁵ Given its importance, ratification of the UNCAT is today a matter of priority³⁶ and it could be said that ratification of the Convention is a State's minimum commitment to international human rights law.

As the preamble suggests, the drafters of the Convention had not intended to reiterate the requirement of the prohibition of torture and other ill-treatment that have been already outlawed under international law.³⁷ On the contrary, they wished to move on with the fight against torture and to develop concrete measures to make the fight against torture more effective.³⁸ Given its strong emphasis on the response of criminal law, some scholars question the traditional view of the UNCAT as a human rights instrument. Malcolm Evans, for example, argues that its primary purpose is not to outlaw torture as a practice contrary to human rights, but rather to assert jurisdiction over acts of torture, and most importantly, to lay the basis for universal jurisdiction.³⁹ Against this background, Carver and Handley deduced that the Convention “has a stronger family resemblance to UN crime suppression treaties than to other human rights instruments”.⁴⁰ After all, the criminalization of torture provided by Article 4 was directly inspired by similar articles in UN Conventions regarding hijacking, sabotage against aircraft, attacks on diplomats, and the taking of hostages.⁴¹

The UNCAT provided a legal definition of torture for the first time in treaty history. It determines torture as an absolutely prohibited act that cannot be justified under any exceptional circumstances and provides a clear definition of torture with a definite set of constituent elements.⁴² Article 1 of the Convention determines that an act could be labelled as “torture” if, and only if, severe pain or suffering, whether physical or mental, is purposefully and intentionally inflicted on a person by the State's representatives or on their behalf.⁴³ This definition is often accompanied by the element of the victim's powerlessness, as torture is believed to occur only in situations of physical control over the person.⁴⁴ Although the Convention's definition is often contested by the scholars⁴⁵ who point out that some elements should better be clarified⁴⁶ or even deleted,⁴⁷ it goes without saying that the definition of torture in Article 1 of the Convention is now taken as a model definition worldwide.⁴⁸

Against this background, it is apparent that ratification of the Convention is all but an easy endeavour. The Asia-Pacific region belongs to that of the lowest ratification rate as of 64 States,⁴⁹ 13 countries have made no action toward ratification⁵⁰ and three States have signed but not ratified.⁵¹ With regard to the ratification of the OPCAT, the situation is even more lamentable as it has been ratified by only six Asia-Pacific States. Hence, the unprecedented move taken by Taiwan to ratify both UNCAT and OPCAT could have some knock-on effect on its neighbours. As Goodliffe and Hawkins have demonstrated, the commitments to the Convention affect the likelihood that the country's neighbours will sign or ratify: “If most neighbours have signed or ratified, then that country is more likely to sign or ratify, and do so sooner.”⁵²

II. Ratification of the Convention in Taiwan

It goes beyond doubt that ratification of the UNCAT in Taiwan is a highly challenging endeavour. Unlike some countries where the prohibition of torture has a certain legal tradition and is even considered as the key constitutional principle,⁵³ neither Taiwan's Constitution nor the ordinary legislation contains an explicit prohibition of torture adhering to its absolute and non-derogable character. Although the law randomly refers to the notion of “torture” or similar terms (there are two terms of “torture”⁵⁴ and six terms referring to other forms of ill-treatment in legislation⁵⁵), it is evident that an act of torture has not been established in the legal terms, not alone as a distinct criminal offence. As a result, when allegations of torture appear, they are commonly seen as something elusive and incomprehensible—difficult to deal with by traditional legal measures. This lacuna causes ineffective investigation of torture allegations and widespread impunity of torture perpetrators in Taiwan.⁵⁶ Apart from a legal revision, an overall mindset of law-enforcement agents must be changed. To illustrate the latter, Professor Yung-Lien Lai from the Central Police University⁵⁷ noted that given the humanization of the prison system⁵⁸, many prison guards today feel uncertain about their competencies and afraid of “losing their powers”. He explained that the external supervision of prisons and the new “service-oriented approach” might be rather challenging for them to accept as they were used to working for a long time in an environment without external supervision and with emphasis on the use of force.⁵⁹ This illustrates that the development of anti-torture safeguards is still in the embryonic phase in Taiwan, and the implementation must thus not be limited to changes of statutory law, but concern administrative and judicial apparatus, preventive monitoring, training, and human rights education.⁶⁰

Other significant challenges concern Taiwan's exclusion from the UN and forty years of a totalitarian era under martial law.⁶¹ The international legal vacuum dates back to the year 1971 when Taiwan (more precisely, the Republic of China) lost its seat at the UN. As a non-UN member State, Taiwan was no longer allowed to officially sign and ratify the UN human rights instruments. Although Taiwan has signed the International Covenant on Civil and Political Rights (“ICCPR”), including its first optional protocol and the International Covenant on Economic, Social and Cultural Rights (“ICESCR”) in 1967, it was unsuccessful in ratifying these documents prior to its exclusion from the UN.⁶²

In the aftermath of the martial law era, the then-president Chen Shui-bian announced his intention to “ratify” the International Bill of Human Rights⁶³ and initiated the process of voluntary adaptation of domestic law to human rights commitments under the UN treaties. Since that time, a plethora of human rights guarantees has been adopted, including procedural safeguards of a criminal trial (in particular a mandatory recording).⁶⁴ Democratization process along with “domestication” of international human rights instruments have tempted many to believe that torture no longer occurs in Taiwan today.⁶⁵

However, this inference seems premature. Given the lack of independent external monitoring of detention places and missing concept of torture in Taiwanese law, torture remains present in some disguised forms and situations. Evidently, there have been a number of documented cases of abuse, ill-treatment, and deaths in some areas of public administration, such as deaths of soldiers in military settings,⁶⁶ forced labour and inhumane treatment of migrant workers on fishing vessels,⁶⁷ overcrowding of prisons,⁶⁸ excessive use of restraints,⁶⁹ or the extradition of foreign citizens to countries where they might face a risk of torture.⁷⁰ These examples illustrate what Nils Melzer calls, the “torturous environment”,⁷¹ which is a dangerous climate of potential torture or ill-treatment. This concern is fuelled by the suggestion of the Minister without Portfolio and Spokesperson of the Executive Yuan, Lo Ping-cheng, that more research is needed to determine the occurrence of psychological torture despite significant improvements in eradicating physical torture.⁷²

A. Legislative History and Reasons for Ratification

Since Taiwan is prevented from formally ratifying the UN treaties and depositing the ratification instruments with the UN,⁷³ it has developed its own ratification model based on the adoption of a domestic implementation act⁷⁴ that grants the treaty's provisions a domestic legal status with the legal force of domestic statute.⁷⁵ The treaty provisions then come into force on the date of promulgation by the President of Taiwan.⁷⁶ Once the treaty is ratified, its provisions gain the domestic legal effect and could be directly invoked before the Taiwanese administrative and judicial bodies.⁷⁷ To date, five United Nations treaties have been implemented in domestic law,⁷⁸ including the ICCPR and its Article 7 (prohibition of torture). The process of treaty “domestication” by means of an implementation act has established a path-dependent legislative model that has been followed by the subsequent implementation acts as by the present UNCAT Implementation Act.

On top of that, as Taiwan is unauthorized to engage in formal dialogue with UN treaty bodies, a “self-made international review process” introduced by domestic stakeholders to hold the governments accountable for their treaty commitments.⁷⁹ Parallel to the UN reporting mechanism, the Taiwanese government has obliged itself to prepare a State report on treaty implementation every four years and to invite a group of independent international experts such as the International Review Committee (“IRC”) to review this report. A coalition of non-governmental organizations, the Covenants Watch, has become the key watchdog for the implementation process—observing the Government's responses and preparing its own parallel report to IRC.⁸⁰

In 2013, the IRC reviewed the initial reports on the ICCPR and ICESCR for the first time. Notwithstanding several concerns and recommendations, the IRC praised the democratic and human rights progress in Taiwan as an “exemplary commitment to the process of monitoring compliance with the relevant human rights obligations”.⁸¹ In order to further strengthen human rights protection, the IRC recommended initiating processes aimed at an early acceptance of the obligations under several human rights treaties including the UNCAT, and setting up a national preventive mechanism under the OPCAT.⁸² The same recommendations were reiterated in 2017.⁸³

To satisfy the said recommendations, internal debate on UNCAT and OPCAT ratification was initiated within the Executive Yuan.⁸⁴ In 2013, the Criminal Investigative Bureau of the Ministry of Justice (“CIB”) was appointed as the agency responsible for the implementation of the Convention and the Optional Protocol. It has been tasked to commission research activities, organize expert seminars, and coordinate preparatory work on implementation. Two Taiwanese universities⁸⁵ have been also conducted research projects on preventing torture.⁸⁶ In particular, research carried out by the Central Police University identified eight ministries and agencies that are likely to be tasked with implementation once the Implementation Act is passed. The research further developed several recommendations for revising current law.⁸⁷ It is surprising, however, that unlike advocacy for implementation of the human rights treaties in recent years,⁸⁸ civic groups have not launched any visible campaigns on UNCAT and OPCAT implementation.⁸⁹

A turning point in the UNCAT advocacy occurred in 2019 when the Executive Yuan drafted and sent the UNCAT and OPCAT Implementation bill to the Legislative Yuan for deliberation. Regrettably, due to the parliamentary elections in January 2020, the bill remained undiscussed and later expired. Nevertheless, the newly formed government remained committed to ratification and a new bill was prepared by the Executive Yuan on the basis of the previous draft law.

On 10 December 2020, the draft was approved by the government and sent to the Legislative Yuan for deliberation. When approving the bill, the Taiwan premier Su Tseng-chang noted that “[w]hile the days of cruel and torturous treatment are far in the past, the nation must remember those lessons learned and avoid repeating the same mistakes by creating a legal basis for the convention to be applied”.⁹⁰ In addition, premier Su emphasized that the present ratification “demonstrates to the world Taiwan's commitment to becoming a nation founded on human rights and its keenness in aligning domestic human rights laws with those of the international community”.⁹¹

These remarks aptly explain the three main incentives for the implementation of anti-torture documents in Taiwan. First, it is evident that the ratification of the Convention is an important historical milestone that strives to make up for the torturous past and pledges for the future of human treatment and respect for dignity. Second, it can breathe a new life for addressing current policies and practices that increase the risk of torture and ill-treatment, such as harsh military discipline, forced labour on fishing vessels, and a plethora of other systemic deficiencies as demonstrated above. Third, it goes beyond doubt that the Taiwan government views the present ratification as another step in Taiwan's ambition to stand shoulder to shoulder with the international community and showing the world that it emerges into a democratic State respecting a rule of law and human rights.

B. Path-Dependence Model of the Implementation Act and its Pitfalls

As noted above, on the basis of Article 2 of the UNCAT and OPCAT Implementation Act,⁹² both the Convention and its Optional Protocol are granted legal force as domestic statute and might be directly invoked before judicial and administrative bodies. In order to ensure that the interpretation of the treaty provisions is in compliance with the spirit of these documents, Article 3 of the Implementation Act calls on the Taiwan authorities to apply treaty provisions in accordance with “the object and purpose of the UNCAT and OPCAT” and to refer to “general comments and resolutions of the CAT and its affiliated committees, the review opinions of national reports, decisions on individual complaints and the identification of international legal documents”.

On top of that, the explanatory report provides that a wide range of anti-torture standards, including the decisions of the European Court of Human Rights (“ECtHR”), shall be taken into account.⁹³ Article 3 of the Implementation Act could therefore be regarded as a clause that opens the door to a plethora of anti-torture standards and interpretations developed not only by the CAT and the SPT under the UN, but also by the ECtHR when interpreting Article 3 of the European Convention of Human Rights (“European Convention”).

However, this praiseworthy regulation includes several pitfalls. First, given the different definitions of torture under the UNCAT and the European Convention,⁹⁴ it is not clear whether a higher standard of protection provided by the European Convention shall trump the standards established under the UNCAT.⁹⁵ Second, it has not yet been established how precisely the law-enforcement agencies in Taiwan will apply the abundance of relevant anti-torture standards, in particular, what mechanisms will take place to resolve a potential conflict between international standards and domestic law.

An important provision of the UNCAT Implementation Act is Article 9, which introduces the “gradual adaptation model”⁹⁶ and obliges the government agencies at all levels to carry out an overall legal reform. First, this provision requires governmental agencies to forward a checklist of relevant laws and administrative measures within one year after the implementation of the Act. Second, within the period of three years after the ratification, they are requested to complete the formulation (enactment), amendment or repeal of laws and regulations and the improvement of administrative measures. Bearing in mind that the ratification of the Convention could be a costly issue, the Implementation Act also stipulates that the governmental agencies shall prioritize the expenditures required to implement the provisions of the UNCAT and OPCAT.⁹⁷

Evidently, the above provisions of the Implementation Act could be regarded as a clear-cut roadmap that is built on the pledge to carry out a comprehensive legal assessment and to adopt a variety of legislative, administrative, judicial, and other measures to ensure compliance with the Convention.⁹⁸ They also show a commitment to follow the international human rights standards as part of the process of interpretation and application of the Convention. However, two caveats are worth noting here.

As with previous implementation acts, the UNCAT Implementation Act is governed exclusively by a top-down model,⁹⁹ meaning that a moment of ratification is the impetus that triggers the review process, and as a consequence, no legislative changes can be envisaged before that date.¹⁰⁰ This provides State authorities with an excuse not to adopt any torture-relevant regulations until the Implementation Act is passed.¹⁰¹ This implementation mode has been criticized by scholars. For example, Teng Yean-Sen has pointed out that this model raises doubts about the obligatory nature of human rights obligations in Taiwan. He argues that, on the basis of this model, the human rights obligations are related more to the concept of legality, rather than to be linked to their legitimacy.¹⁰² This is a fortiori true for the obligation to combat torture, as it is a pivotal duty of all States emerging from the character of torture as a principle of customary international law and jus cogens norm. Against this background, the Implementation Act has rather a declaratory than a constitutive character¹⁰³ of the obligation to prohibit torture. Moreover, as Taiwan is already bound by this commitment under Article 7 of the ICCPR (ratified in 2009), it is obliged to adhere to a plethora of anti-torture safeguards today.¹⁰⁴

In addition, scholars Baccini and Urpelainen have noted that the treaty implementation is closely linked to the momentum of ratification, the true nature and content of treaty obligations could be blurred¹⁰⁵ and a State thus may fail to examine properly its implementation capacities and ratification demands.¹⁰⁶

The second caveat concerns the simultaneous implementation of the UNCAT and OPCAT in a single Implementation Act. Article 27 of the OPCAT stipulates that the Optional Protocol is open for ratification only to States that have first ratified the Convention. This is a common character of the optional protocols as they are intended to enhance human rights protection first established by the text of the “main treaty”¹⁰⁷ and it is, therefore, not surprising that there is frequently some time lag between the ratification of the UNCAT and the OPCAT.¹⁰⁸ Having regard to travaux préparatoires of the Convention, it appears that Article 27 was subject to a vivid discussion within the Working Group on a draft Convention against Torture and significant support was given to adopt both documents at the same time and to perceive the OPCAT as a free-standing treaty that could be ratified at any time.¹⁰⁹

However, given the concerns that construction of the Optional Protocol might complicate the elaboration of the Convention or even “consume” the Convention's provisions contained within the Swedish draft of the Convention, the concept was abandoned.¹¹⁰ It is plausible that Article 27 of the OPCAT is obsolete today, as the vast majority of States have already ratified the Convention and nothing indicates that concurrent ratification of the two documents by the non-State Parties might have any adverse effects on domestic or international protection of torture.

Arguing the opposite, as a State is willing to ratify the OPCAT and open the detention places to independent oversight by the NPM and the SPT, the more it may be expected to be engaged in the adoption of other preventive measures arising from the text of the Convention. Seeing from this angle, the parallel ratification of the two documents in Taiwan sends a clear message to domestic authorities to promptly begin with monitoring torture and ill-treatment in order to obtain a true picture of the conditions and treatment in detention centres. In turn, early understanding of the situation in detention places could help to better design other measures to combat torture within the framework of the Convention.

However, when the two documents are combined within a single Implementation Act, due diligence must be paid not to intertwine the OPCAT preventive mandate with a broad range of other preventive anti-torture measures under the UNCAT. In particular, it should be recognized that the mandate of the NPM under the OPCAT is exclusively limited to preventive monitoring and should not replace other general purpose mechanisms.¹¹¹ Against this background, the present Article 5 of the Implementation Act is a bit vague, as it authorizes the NHRC to set up the NPM but omits to provide a clear legal basis and organization for the NPM.¹¹²

Taking together the benefits and challenges of the present Implementation Act, it is plausible to believe that, despite several pitfalls, this model establishes a coherent path toward compliance with the Convention. It recognizes the binding effect of the UN anti-torture framework, acknowledges the authoritative interpretation of the Committee against Torture, endorses a need for broad legislative reform, and establishes a path (including a time frame) for conducting such a reform. Further, it anticipates creating the NPM to start promptly with preventive monitoring of places of detention. Notably, the NHRC has already allocated personnel for the NPM mandate, has elaborated manuals and checklists for visits and has already carried out a pilot program including several NPM-like visits to places of detention.¹¹³

III. International Effect of Ratification of the Convention in Taiwan

It has been said earlier that the pivotal aim of the Convention is an assertion of jurisdiction over acts of torture, in particular laying the basis for universal jurisdiction. Hence, the question arises as to whether the ratification of the UNCAT in Taiwan could contribute to that aim and foster the accountability of torture perpetrators at domestic level, in Asia-Pacific, and worldwide. The three conditions must be met to effectively address the international crime of torture. First, torture must be criminalized as a specific criminal offence. Second, a State must adopt laws and policies authorizing its judicial bodies to exercise jurisdiction over those who allegedly commit that crime. Third, a certain level of international cooperation is needed to effectively bring torture perpetrators to justice. These prerequisites shall be examined in some detail.

A. Criminalization of Torture in Taiwan

The Constitution of the Republic of China (Taiwan) establishes criminal liability for public officials who infringe “freedoms or rights of persons”. On the statutory level, criminal liability for civil servants is provided for in criminal law, in particular, the Criminal Code and the Criminal Code of the Armed Forces.¹¹⁴

The Criminal Code contains provisions that are deemed to serve as a legal basis for sanctioning public officials for crimes falling under the definition of torture. The most relevant provisions are those concerning abusive powers, using the threat of violence with the purpose to extract a confession, violence or cruelty (Articles 125 and 126). In a similar vein, the Criminal Code of the Armed Forces prohibits commanders from abusing their subordinates (Article 44).

It appears that the term “abuse” is the closest term to that of torture or ill-treatment. The Criminal Code defines abuse as “an act of abuse or maltreatment of another person in a violent, coercive or inhumane way” (Article 10). The Criminal Code of the Armed Forces provides for a similar definition of abuse, being equal to “a range of military actions which are considered as “inhuman treatment to soldiers” (Article 44).¹¹⁵ In addition to these provisions, there is no other criminal law rule that can be invoked as a criminal response to torture.

There is no doubt that the said provisions do not meet the requirement that torture must be criminalized as a separate offence, which expressly incorporates the definition of torture and all of its elements (Articles 1 and 4 of the Convention).¹¹⁶ Pointing to this lacuna, the IRC has repeatedly recommended to Taiwan to “insert the crime of torture, as defined in Article 1 of the UN Convention against Torture, as a separate crime with adequate penalties in its Criminal Code”.¹¹⁷ On the contrary, in the response to the IRC, the government of Taiwan expressed a conviction that the above provisions are those that “expressly prohibit and impose penalties on torture and cruel treatment in Taiwan”¹¹⁸ and that “Taiwan already has laws that prohibit crimes described as torture in Article 7 of the ICCPR and Article 1 of the Convention against Torture”.¹¹⁹

The understanding of a crime of torture under the above provisions apparently shows a “Janus face”, as it suggests that torture is already a criminal act, but in fact, no legislation establishes a crime of torture in Taiwan in accordance with Article 1 of the Convention (not even the notion of “torture” is used by the legislative texts). Four caveats emerge from this gap.

First, the above provisions do not respect the absolute nature of the prohibition of torture, as the existing crimes (that allegedly constitute the legal basis for torture) are subject to a 20-year statute of limitations¹²⁰, presidential pardons,¹²¹ and amnesties.¹²²

Second, due to the missing regulation on part of a “public official” in accordance with Article 1 of the Convention, there is a narrow understanding of “public officials” in Taiwan. This problem has been identified by Qian Jianrong who has criticized the approach of the Taiwan Supreme Court¹²³ for adhering to the narrow interpretation of public officials embodied in Article 125 of the Criminal Code as to be prosecutors and judges only.¹²⁴

Third, the criminal punishment for crimes reaching the level of torture is not addressed by the severe penalty. Articles 125 and 126 of the Criminal Code provide for a punishment of imprisonment between one to seven years (and Article 44 of the Criminal Code of the Armed Forces provides punishment ranging between three and ten years). Although the penalty of imprisonment may be increased if certain aggravating circumstances (e.g., serious bodily harm or death), the principal punishment is considerably lenient given the grave nature of torture and when comparing it with other crimes. For example, in the case of an offence of serious physical injury, the Criminal Code (Article 278) stipulates between five to twelve years of imprisonment.

Fourth, given the absence of the crime of torture, there is no credible data concerning allegations of torture and criminal response. For example, the initial report on the ICCPR states that 24 civil servants were indicted between 2005 to 2008 for various crimes, in particular, for abusing their powers¹²⁵ and of those five were found guilty.¹²⁶ Such general information provides no clue on the nature of these abuses, in particular, whether these cases reached the level of torture or remained below its threshold.

Although the establishment of the crime of torture is not in itself sufficient to effectively fight against impunity as underscored by Carver and Handley,¹²⁷ recent empirical data demonstrates that there exists a correlation between the establishment of a specific offence of torture and the reduction of torture (police torture).¹²⁸ This observation has been further supported by country-specific studies,¹²⁹ frequent CAT's recommendations,¹³⁰ and UN Special Rapporteurs against Torture¹³¹ as well as by recent ECtHR's jurisprudence.¹³² In the same vein, the IRC has made it clear to the Taiwan government that the incorporation of a separate and specific crime of torture with adequate penalties into the Criminal Code of Taiwan is essential to fighting impunity for perpetrators of torture, which is deemed to be one of the most effective means to eradicate torture and other forms of ill-treatment.¹³³ Despite the above evidence, to date, the IRC recommendations remains unheard of by Taiwan authorities.

Apparently, the current UNCAT Implementation Act does not in itself provide a sufficient legal basis for the crime of torture, and overall legal reform of criminal legislation is thus needed¹³⁴ with the aim of establishing torture as a separate offence.¹³⁵ Evidently, the new crime shall be also accompanied by the entire reform of criminal procedure adhering to a plethora of rules concerning prompt, impartial and effective investigation¹³⁶ and rapport-based interviewing.¹³⁷ Moreover, law-enforcement officials, including judges, must be continuously trained to be able to recognize the torture and adopt an appropriate response. As Nils Melzer, the UN Special Rapporteur on Torture explains, a holistic approach must be adopted to integrate the fight against torture into national legislation, policies, implementation procedures, and mechanisms of accountability and oversight.¹³⁸ The main task would be to change the mindset of the judicial branch to show a zero-tolerance of evidence and confessions extracted using torture and to demonstrate a proactive approach regarding to torture investigation.¹³⁹ Other factors that perpetuate violations and impunity should be also examined.¹⁴⁰ Carver and Handley aptly suggest that since the law is a minimum standard and the “practice is often dependent or consequent upon it”,¹⁴¹ what actually matters is “not laws on the statute book but practice in the police stations.”¹⁴²

B. Universal Jurisdiction and International Cooperation

Criminalization of torture as a specific offence in accordance with Article 1 of the Convention and establishing jurisdiction over the suspected tortures are two sides of the same coin providing minimal safeguards against impunity. On the basis of Article 5 of the Convention, which is considered a cornerstone of the Convention,¹⁴³ State Parties shall establish jurisdiction over the crime of torture that occurred within its territory¹⁴⁴ or committed by (or against) its nationals.¹⁴⁵ Furthermore, for the first time in a human rights treaty,¹⁴⁶ the Convention against Torture requires States Parties to establish jurisdiction over the crime of torture committed by any offender in any territory (universal jurisdiction).¹⁴⁷

The Convention adheres to the well-established principle that as the prohibition of the torture imposes upon States obligations erga omnes, the perpetrators of torture are regarded as hostes humani generis (the enemies of all mankind),¹⁴⁸ that must be brought to justice regardless of their whereabouts or nationality.¹⁴⁹ This principle subjects all members of the international community to the same correlated rights and duties,¹⁵⁰ and ensures that there is no safe haven for alleged perpetrators of torture.¹⁵¹

On top of that, as the prohibition of torture enjoys a higher rank in the international hierarchy than a treaty and even “ordinary” customary rules,¹⁵² it is plausible to believe that the principle of universal jurisdiction has attained the character of international custom and therefore obliges all members of the international community, regardless of the status of ratification of a particular treaty.¹⁵³ Additionally, the Convention does not limit itself to “only” recognizing the principle of universal jurisdiction, but also establishing a system¹⁵⁴ to exercise such jurisdiction in practice. It also expressly provides the international obligation to provide each other with necessary assistance.¹⁵⁵

States Parties have a legal obligation under the Convention to take legislative, executive, and judicial measures to establish universal jurisdiction over the crime of torture.¹⁵⁶ The Committee against Torture requires States to make “clear legislative provisions” on universal jurisdiction in their domestic law.¹⁵⁷ This requirement may be demonstrated by the landmark case of Hissène Habré, where the Committee found a violation of Article 5(2) of the Convention, on the ground that the Senegalese authorities failed to take the legislative measures necessary to establish the legal possibility for the Senegalese courts to exercise the universal jurisdiction.¹⁵⁸

In Taiwan, although Article 5 of the Criminal Code lists offences for which national judicial bodies have universal jurisdiction (such as the offences of sedition, piracy, treason or drug-related offences), neither crime of torture nor “torture-like” crimes under Articles 125 and 126 of the Criminal Code are included in this enumerative list.¹⁵⁹ In a nutshell, there is no legal provision in Taiwan's criminal law that establishes a jurisdiction over the foreign torturers for torture committed abroad. Therefore, it is plausible that if a foreign torturer such as Hissène Habré appears on Taiwan's territory with the plea for a safe haven, he will likely succeed. This is not only a hypothetical concern but may have direct practical implications as observed in the recent case of human trafficking, which failed to bring the perpetrators to justice.¹⁶⁰

However, even if the legislation fully complies with the principle of universal jurisdiction, it could not provide a sufficient guarantee of the effective investigation and prosecution of foreign torturers in practice unless there is close cooperation in the fight against torture worldwide. In this regard, Article 9 of the Convention obliges States Parties to provide one another the greatest measure of assistance in connection with criminal proceedings, namely, provide the evidence required for the criminal proceedings such as witness testimony or documentary evidence.¹⁶¹

Schmidt explains that this provision obliges both a State in which torture has been committed in and a State in which an alleged torturer is a citizen to provide a forum State with all evidence needed to proceed with the prosecution.¹⁶² Burgers and Danelius have further noted that while respect should be given to domestic procedural rules (such as those concerning the admissibility of evidence), rules and procedures that are excessively burdensome or make cooperation virtually impossible do not comply with Article 9's obligation.¹⁶³

Regrettably, international cooperation regarding universal jurisdiction is a thorny issue. It commonly requires an active engagement not only from the parties concerned, but also from a whole array of actors, particularly, regional organizations.¹⁶⁴ Taiwan's case is even more problematic, as it maintains official diplomatic relations with only fourteen countries.¹⁶⁵ Moreover, it has concluded bilateral treaties on mutual criminal cooperation with only seven countries.¹⁶⁶ Besides these limited official relations, cooperation with other countries is conducted on an ad hoc basis¹⁶⁷ and is governed by domestic law provisions.¹⁶⁸ Where the responding State refuses to cooperate, Taiwan has neither a mechanism to enforce such cooperation nor a mandate to refer the problem to an international forum.

Overall, international cooperation between Taiwan and other States is established on an ad hoc informal basis and is very circumspect in addressing Taiwan's international status.¹⁶⁹ Moreover, if any kind of cooperation is established, it is mainly justified by the protection of the parties’ own interests, such as the smooth return of one another's fugitives or criminal prosecution of individuals (private offenders). This could be well illustrated by the mutual judicial assistance between Taiwan and mainland China, which has purely practical reasons and provides very little room for human rights consideration, as Chen and Cohen have demonstrated by the example of the cross-strait extradition procedure.¹⁷⁰

Against this background, it is plausible to believe that given the “non-recognized” status of Taiwan, there will be very limited (if any) international cooperation under the principle of universal jurisdiction. It is conceivable that local judicial bodies will be reluctant to prosecute foreign torturers (let alone high-ranked foreign officials), as this might involve a risk of a destabilizing effect on international relations.¹⁷¹ In the same vein, the UNCAT States Parties will unlikely be keen to provide Taiwan judicial assistance to trial their citizens. Besides the political concerns, the lack of experience of Taiwanese courts concerning international criminals could also play a significant role because they have not yet initiated any criminal proceedings based on the universal jurisdiction under Article 5 of the Criminal Code.¹⁷² Despite these practical difficulties, which might render a substantial part of the provisions of the Convention inapplicable, neither the CIB nor any other government agency has so far analysed these implementation perils.¹⁷³ The future developments in UNCAT implementation will shed more light on these concerns and implementation dilemmas.

IV. Conclusion

Herman Burgers and Hans Danelius, one of the drafting fathers of the text of the Convention, have aptly emphasized that “in the struggle for the protection of human rights it is necessary to be both a sceptic and a believer”.¹⁷⁴ These antipoles are being observed on the road toward eradicating torture in Taiwan.

One might be overwhelmed by scepticism, as Taiwan's current legal framework has very little in common with the basic anti-torture safeguards, including the core safeguards such as the criminalization of torture. It seems that despite praiseworthy democratic processes, allegations of torture or other ill-treatment are not taken seriously by law-enforcement entities and the criminal justice system. On top of that, it seems that the Taiwan government is resistant to respond to some systemic deficiencies despite repetitive criticism by domestic and international actors.¹⁷⁵

To sum up, despite the local government's lofty aim to put in place anti-torture safeguards by ratifying the Convention and its Optional Protocol, a bee in its bonnet appears as to whether the present endeavour could truly achieve its objective. From an affirmative angle, one may consider the current ratification process as an official acknowledgement of the need for a change and a pledge of that change. By placing it in a broader context, Taiwan has shown a colossal achievement in pursuing human rights and fostering democratic foundations in recent years.

Above all, taking into account the text of the Implementation Act and related preparatory activities, the following points suggest that the current ratification process is not an instance of “window-dressing” ratification: First, ratification of the Convention is made without any reservations. Second, the Convention will be a part of domestic law. Third, implementation of the Convention shall adhere to the practice of the UN treaty bodies. Fourth, the OPCAT will also be ratified including the commitment to establish the NPM. Fifth, there were already several government-sponsored research studies and consultations. Sixth, the NHRC has already started with some preparatory works for the new NPM agenda. Seventh, the Implementation Act stipulates a timely-limited legislative assessment and consequent legislative reform. Eighth, the national budget is prioritized for the implementation process.

Although it is premature to conclude on the success of the UNCAT and OPCAT project in Taiwan, the present ratification is undoubtedly a significant step forward in creating an anti-torture framework. For the first time in Taiwan's history, the government has dedicated itself to setting up a plethora of anti-torture safeguards that are believed to become an important and indispensable element in the fight against torture.¹⁷⁶ Needless to say, however, the prompt revision of criminal law for the establishment of a distinct crime of torture and universal jurisdiction will be the watertight evidence that Taiwan is serious about the UNCAT and OPCAT implementation.

On top of that, some implications may be distilled for the UN treaty system and the enhancement of the UNCAT and OPCAT implementation efforts in Asia. Evidently, the case of Taiwan shows that the formal membership in the international organization is not a conditio sine qua non for the implementation of the treaties of that organization. Taiwan's exclusion from the UN had constituted neither a legal obstacle for domestic stakeholders to voluntarily incorporate the treaty obligations nor did it undermine their political will to do so.

Against this background, it appears that an act of ratification says very little about a State's willingness to truly comply with a given set of commitments. Hence, besides the international impetus on treaty ratification and international enforcement, the emphasis should be put more on the actual compliance with treaty provisions on the domestic level by investing in local actors to become internally motivated to voluntarily adapt their laws and practices to universal standards.

Nonetheless, encouraging domestic actors to voluntarily and genuinely strive for human rights does not undermine the importance of international treaties because they are vital for the enhancement of human rights commitments on the domestic level. It is clearly observed in Taiwan's case that the current pledge for changes in legislation and practice is directly inspired by the two UN treaties and without such an incentive, neither the establishment of anti-torture safeguards nor the monitoring of places of detention would be on the agenda in today's Taiwan.

Remarkably, as there exists an overlap between the Convention and the customary international law,¹⁷⁷ it is worth exploring whether extending the normative scope of the Convention beyond the UN influence contributes to the emerging process of new rules under customary international law. As aptly noted by Erika Wet, the exact scope of the peremptory norm of prohibition of torture is controversial and it is highly disputed, for example, whether it includes an obligation to prosecute torture perpetrators or grant victims the right to redress.¹⁷⁸ Hence, it is believed that the academic discourse about the international customary law would benefit from an analysis of the impact of the treaties on non-States Parties and States standing completely outside the UN treaty system. Undoubtedly, the case of Taiwan would be helpful to initiate such an analysis.

In any event, it is credible to believe that the Convention should no longer be regarded as the exclusive project of the UN, but could be rather considered as the universal platform for all members of the international community.¹⁷⁹ In this context, the voluntary and unreserved adherence to the UN treaty by the non-UN member is also vital for fostering the legitimacy of the UN treaty project and could be a welcomed advance for its future development and the promotion of human rights ideas elsewhere, particularly, in countries that are so far reluctant to ratify.

Moreover, a practical consequence to the universal struggle against torture shall be pointed out. Since the moment of ratification, not only Taiwanese citizens but also foreigners are expected to enjoy fundamental guarantees against torture in Taiwan, such as guarantees against refoulement, protection against cruel interrogations during a criminal investigation, or human conditions in places of detention. In particular, it is hoped that the abuses of many migrant workers from South-East Asia and protect foreigners of many categories against refoulement could be addressed.¹⁸⁰ Furthermore, Taiwan strives to become a partner in bringing foreign torturers to justice in accordance with the principle of universal jurisdiction.

It should be further underscored that despite Taiwan’ goodwill to implement all the commitments of the Convention, its international isolation could render some provisions inapplicable. In particular, where there is no official international assistance in criminal matters, Taiwan is likely to be unable to exercise universal jurisdiction over torturers, which might increase the potential loophole of impunity and reduce the protection against torture worldwide.¹⁸¹

At the end of the day, the eradication of torture is a shared responsibility of all members of the international community. If Taiwan truly succeeds and ratifies these challenging treaties, it shall then be up to the international community to pick up the gauntlet and make the most of this welcomed initiative.