Compulsory Labor: A Rehabilitative Measure Difficult to Distinguish from the Execution of Imprisonment

※ Originally published by The Reporter on November 16, 2021

Introduction

On the afternoon of October 12, 2021, the Justices of the Constitutional Court held oral argument in the constitutional case concerning compulsory labor[1], and the Court has since announced that it will issue its Interpretation on December 10, Human Rights Day. The case has attracted substantial public attention.

As early as 2001, the Justices issued J.Y. Interpretation No. 528holding that compulsory labor was constitutional. That conclusion, however, has long been questioned. Twenty years later, the system is finally before the Constitutional Court again. The petitioners raise several different objections. One concerns Article 3, Paragraph 3 of the Organized Crime Prevention Act[2]which, without differentiating among offenders or the seriousness of their conduct, required anyone convicted under Paragraph 1 to be sent to a labor establishment for three years of compulsory labor. The concern became even greater after the 2018 amendment broadened the definition of a criminal organization beyond coercive or violent crime to include structured organizations operating continuously or for profit. Other petitioners challenge the imposition of compulsory labor in addition to an ordinary criminal sentence where the two are not meaningfully different in substance, arguing that this effectively punishes a person twice for the same criminal act.

The Justices consolidated the petitioners' constitutional objections and reframed the issues around the compulsory-labor provisions in Article 90 of the Criminal Code[3], Article 3, Paragraph 3 of the Organized Crime Prevention Act, and Article 3, Paragraph 1[4] and Article 5, Paragraph 1[5] of the Rehabilitative Measures for Burglar and Fence Criminals Act. The questions include whether compulsory labor violates the prohibition against double punishment, infringes the liberty and security of person protected by Article 8 of the Constitution, or fails to satisfy constitutional due process in judicial decision-making. The case also raises the more fundamental question whether compulsory labor infringes human dignity. A central point of dispute is whether compulsory labor is "clearly differentiated," both in law and in practice, from the execution of imprisonment.

Compulsory labor is a rehabilitative measure imposed in addition to a criminal sentence. In general, when a court orders a rehabilitative measure on top of punishment, the justification is "special prevention": the offender is regarded as posing a risk of reoffending that punishment alone cannot adequately address, so the person is ordered to undergo treatment or another intervention aimed at reducing that risk. If a particular rehabilitative measure is not meaningfully distinguishable from punishment in the way it is carried out, however, the legitimacy of imposing it in addition to the sentence becomes doubtful. A closely related issue arose in J.Y. Interpretation No. 799issued on December 31, 2020. There, the Justices considered post-sentence compulsory treatment for sex offenders and held that the system must satisfy a constitutional requirement of "clear differentiation" from criminal punishment; otherwise it may conflict with principles tied to criminal punishment, including legality and the prohibition against double punishment. The Court explained that compulsory treatment is legally designed as a therapeutic rehabilitative measure for sex offenders assessed as presenting a high risk of recidivism, not as additional punishment. Yet years of actual operation had caused the system to resemble imprisonment. The responsible authorities were therefore ordered to make effective adjustments within three years so that compulsory treatment would, in practice as well as in law, be clearly differentiated from criminal punishment.

Compulsory labor raises the same structural concern. Men subject to compulsory labor are currently placed at Taiyuan Skill Training Institute, while women are placed in a skill-training unit attached to Kaohsiung Women's Prison. Both are correctional institutions under the Agency of Corrections, Ministry of Justice, and both also house ordinary inmates. The defining content of compulsory labor is mandatory work[6], but inmates serving prison sentences are likewise generally required to participate in work[7]. Persons subject to compulsory labor and people serving prison sentences are therefore housed within the same correctional system-and sometimes within the same institution-and both are required to work. What, then, actually distinguishes compulsory labor from imprisonment? That is the core of the constitutional objection.

This article argues that, under Taiwan's current system, compulsory labor is indeed difficult to distinguish clearly from imprisonment. That should not be surprising. The history of the prison helps explain why.

 

A History of Prisons and Forced Labor/Work

Forced Labor/Work[8] has long been a recurring feature of punishment. Historically, compulsory labor was often imposed primarily to subjugate offenders or inflict suffering. That history is only indirectly related to the compulsory-labor system now before Taiwan's Constitutional Court. More relevant is the role of compulsory labor and work in the modern prison, where it became a principal technique for "reforming the individual."

The history of the modern prison shows how closely confinement and compelled work developed together, although the relationship has never been free of controversy. Punishment is often imagined primarily as deprivation of liberty. Yet from the emergence of the modern prison and its gradual rise as the dominant punishment in the Western world, transforming or reforming the individual was also treated as a central purpose of imprisonment.

The idea that compulsory labor could reform the individual had religious roots. In seventeenth- and eighteenth-century Europe, labor was increasingly understood as a source of moral transformation: diligence could restore a person to divine grace, while productive work could teach skills useful to society. Modern prison labor also drew on the earlier European workhouse and house of correction. Those institutions were ostensibly established to address poverty, but authorities often viewed their inhabitants as people who stole or begged because they were lazy or unwilling to support themselves through legitimate work. Confinement and compulsory labor were therefore used to correct supposedly defective habits and instill diligence, obedience, and punctuality, while vocational training was intended to help people find work after release (Pierson et al., 2014).

The modern prison emerged around the turn of the nineteenth century. As Foucault observed, prisons appropriated disciplinary mechanisms already operating elsewhere in society and combined deprivation of liberty with techniques aimed at reshaping the individual. Labor was a central part of that project: regular work imposed order, reduced idleness, cultivated habits of obedience and diligence, and was presented as a means of correcting the prisoner's supposedly disordered way of life (Foucault, 2020: 445).

 

Evidence-Based Employment and Vocational-Training Programs in Contemporary Prisons

By the 1970s, American criminology had become deeply skeptical about whether correctional programs actually worked. Robert Martinson's well-known review was popularly summarized as the "nothing works" thesis. Techniques of individual reform that had once been taken for granted could no longer justify themselves on intuition alone, and prison work had to find a new basis if it was to be defended as a tool for reintegration. By the 1990s, "nothing works" increasingly gave way to the search for "what works": rather than assuming that correctional interventions are categorically effective or ineffective, researchers sought to identify, through scientific methods, which interventions reduce reoffending and under what conditions. Corrections thus became increasingly evidence-based (Cullen & Gendreau, 2001).

From an evidence-based perspective, the effectiveness of any correctional intervention-including prison employment or vocational training-must be demonstrated empirically. For many years the evidence on prison work remained inconclusive because studies often lacked rigorous designs. Experimental or quasi-experimental methods generally provide stronger evidence, but they can be difficult to conduct in prisons and may raise ethical concerns. In addition, programs all described as "work," "employment," or "vocational training" can differ dramatically in content, making their outcomes difficult to compare (Bushway, 2003).

As the research base has grown, studies suggest that prison employment and vocational programs are more likely to be effective when they contain elements such as the following (Smith et al., 2018):

• Programs targeted at people with medium or high risk of recidivism tend to show greater benefits than programs targeted at low-risk individuals. One theoretical explanation is that people at higher risk have greater intervention needs and may benefit from sufficiently intensive programming, whereas placing low-risk people in intensive programs can unnecessarily expose them to higher-risk peers.
• Programs with a community-transition component, including vocational training tied to actual labor-market demand, job-readiness preparation, concrete employment opportunities, and continued reentry services after release.
• Integration of employment programming with other correctional interventions. Reintegration needs are multidimensional and extend far beyond employment. Even stable employment depends on more than technical skill. The Risk-Need-Responsivity (RNR) model, for example, assesses multiple criminogenic needs; employment is only one dimension (Bonta & Andrews, 2020).

 

Prisons and Compulsory Labor in Taiwan

1. Work in Taiwan's Prisons

   From the beginning, Taiwan's prison law described imprisonment as having a reformative and reintegrative purpose. Article 1 of the 1945 Prison Act stated that imprisonment should cause inmates to repent and improve so that they could adapt to social life. Article 28 provided that prison work should take account of sanitation, correctional goals, economic considerations, sentence length, health, knowledge, skills, and the inmate's livelihood after release.

   The Enforcement Rules of the Prison Act, first promulgated in 1975, similarly provided in Article 37:
   Work assignments should be determined in light of the inmate's sentence, health, educational level, classification results, existing occupational skills, security needs, and plans for making a living after release.
   Except where otherwise provided by law, or where illness, security, or correctional considerations justified an exception, inmates were required to participate in work. Once assigned, they were not to be transferred to different work without a disciplinary or security reason.

   In other words, prison work in Taiwan was intended to have a correctional purpose, but participation was compulsory rather than voluntary. Seen against the history described above, that is unsurprising: compelled labor and work have long been integral to the operation of prisons.

   Taiwan also developed open prisons and outside-prison work. The Statute of Open Prisons was enacted in 1962. Its legislative explanation described open prisons as a means of training inmates in work skills, using their labor in public construction and economic-development projects, and creating a transitional "stepping stone" back into society before release-the kind of institution historically associated with the Irish intermediate-prison model. In 2017, the rules governing inmate work were amended to introduce "self-managed outside-prison work," moving outside work away from its earlier emphasis on supplying inmate labor for government projects and toward a model intended to help inmates "progressively adapt to social life."

   The principal problem with prison work in Taiwan is the gap between legal design and actual practice. In law, work is supposed to support reintegration. In practice, much in-prison work still consists of simple, low-skill paper processing-folding paper bags, paper lotus flowers, and similar items-and relatively few inmates can participate in work or training that develops marketable skills (Control Yuan Investigation Report No. 108-Ssu-Tiao-0014). Places in open prisons and self-managed outside-prison work are also limited. Escapes attract intense media attention and public tolerance for expanding these programs remains low (Control Yuan Corrective Measure No. 106-Ssu-Cheng-0008; Control Yuan Investigation Report No. 109-Ssu-Tiao-0075). Open prisons have also long faced criticism that people with connections or privilege are more likely to be selected, further undermining public confidence in the system.

2. How Compulsory Labor Is Executed in Taiwan

   Compulsory labor has existed as a rehabilitative measure since the early Criminal Code. Article 90 of the 1935 Code allowed a person who habitually committed crimes, made crime an occupation, or offended because of habitual vagrancy or idleness to be sent to a labor establishment for compulsory labor after serving the criminal sentence or receiving a pardon. Before martial law was lifted in 1987, compulsory labor was administered by the Taiwan Provincial Security Command and later the Taiwan Garrison Command. Its vocational-training corps used military management and imposed strict labor. Because compulsory labor was then carried out outside the prison system, a visible institutional distinction from imprisonment was easier to maintain.

   After martial law ended, the Taiwan Garrison Command was abolished and responsibility for compulsory labor was transferred to the Ministry of Justice. The Ministry had little practical experience administering the measure and faced an immediate problem: how could compulsory labor be distinguished from imprisonment? In 1988, the Ministry formally adopted a policy of "dual legal pronouncement, unified execution." Although imprisonment and compulsory labor remained legally distinct consequences, the substance of execution became essentially the same. A Ministry briefing at the time explained that prisons sought to cultivate diligent habits and occupational skills, while compulsory-labor facilities pursued the same goals; because their functions were identical, unified execution was said to protect the rights of persons subject to the disposition (Hsu, 2013: 198).

   The institutional history is revealing. Between 1987 and 1988, compulsory labor moved from a military disciplinary system into the prison system. Why could the Ministry of Justice absorb it so quickly? Because compulsory labor and work had already been central elements of the modern prison. Taiwan was no exception.

   In 2019, the Control Yuan issued a corrective measure concerning compulsory labor (Control Yuan Corrective Measure No. 108-Ssu-Cheng-0001). The Control Yuan observed that the statutes then authorizing compulsory labor all required the measure to be executed before imprisonment. Yet the Agency of Corrections' skill-training rules in effect from 2011 to late 2018 limited vocational-training courses to people who would be released or become eligible for parole within two years after completing the course. That structure showed that compulsory labor was not, in practice, being implemented as a genuinely "pre-sentence" measure. The Control Yuan further found that skill-training institutions treated people subject to compulsory labor in essentially the same way as ordinary inmates. In its view, that practice was inconsistent with the distinct nature of a rehabilitative measure and risked turning compulsory labor into a disguised extension of imprisonment, raising concerns under the principle of proportionality between offense and punishment.

   The Control Yuan therefore called on the Ministry of Justice and Agency of Corrections to carry out compulsory labor before imprisonment and to differentiate its execution from ordinary prison administration. But the historical problem remains: imprisonment itself absorbed compulsory labor and work as part of its correctional model. The difficulty of drawing a boundary is therefore unsurprising. Indeed, modern prison law now permits a wider range of work arrangements than compulsory labor and places greater emphasis on reintegration-for example through open prisons and self-managed outside-prison work. Compulsory labor, by contrast, lacks a staged work program designed to prepare a person progressively for return to the community[9].

   Why has compulsory labor fallen behind prison law in using work to support reintegration? One reason may be that comparable compulsory-labor regimes have gradually disappeared from other legal systems, leaving little modern legislation to borrow from. Another is that Taiwan's legislature has often paid too little attention to how punishment and rehabilitative measures actually operate in practice, sometimes invoking "special prevention" while legislating primarily for deterrent effect. The gap between legislative purpose and actual execution is especially visible in the sequencing of compulsory labor. Taiwan has alternated between pre-sentence and post-sentence execution, as summarized below:

   Legal Basis	Order of Execution
   Article 90 of the Criminal Code	Originally post-sentence; amended in 2005 to require pre-sentence execution.
   Article 3 of the Rehabilitative Measures for Burglar and Fence Criminals Act[10]	Pre-sentence.
   Article 19 of the Controlling Guns, Ammunition and Knives Act (deleted)	Post-sentence; deleted by amendment in 2001.
   Article 3 of the Organized Crime Prevention Act	Originally post-sentence; amended in 2017 to require pre-sentence execution.

   All compulsory-labor measures still in force in Taiwan are therefore executed before imprisonment. Yet the evidence-based literature described above suggests that work programs with a community-transition function are more likely to succeed, and that vocational training makes most sense when it occurs close to release and can connect directly to post-release employment. Why, then, does Taiwan do the opposite? The answer lies in the rigid sequencing of post-sentence rehabilitative measures. Taiwan generally interprets a measure ordered after punishment as beginning only after the criminal sentence has been fully executed, and unlike German law, Taiwan does not provide a flexible mechanism for executing a rehabilitative measure during the sentence (see § 67 of the German Criminal Code[11]). Parole under supervision does not count as completion of the sentence. Thus, if imprisonment is executed first and compulsory labor second, a person released on parole would have to complete the entire supervision period and then return to a correctional institution to begin compulsory labor. That sequence destroys any plausible reintegrative function. Nevertheless, when compulsory labor was added to the firearms statute in 1996 and when the Organized Crime Prevention Act was enacted the same year, both still provided for post-sentence compulsory labor-showing how disconnected legislation could be from actual correctional practice.

   The Control Yuan's corrective measure emphasized pre-sentence execution and insisted that compulsory labor, as a rehabilitative measure, be carried out differently from imprisonment. But the more fundamental question is whether the compulsory-labor regime itself remains defensible. The answer should not be to force correctional authorities to manufacture distinctions merely so that the system can continue to exist.

 

Assessing the Ministry of Justice's Arguments in the Compulsory-Labor Constitutional Case

The Ministry of Justice's arguments for constitutionality-and even its proposed substitute if compulsory labor were abolished-again reveal how difficult it is, in the contemporary correctional system, to distinguish compulsory labor clearly from imprisonment.

1. The Ministry of Justice's Reliance on Human-Rights Treaties and Comparative Law

   The Ministry of Justice argued that the ICCPR, the European Convention on Human Rights, and the U.S. Constitution all contain rules analogous to Taiwan's compulsory-labor regime, and that the Swiss Criminal Code likewise permits young offenders to be placed in institutions providing vocational training. On that basis, it argued that compulsory labor is constitutional.

   But the Ministry did not identify the specific treaty or comparative-law provisions during oral argument. A closer examination shows that the ICCPR, the European Convention on Human Rights, and the Thirteenth Amendment to the U.S. Constitution address a different proposition: work required as part of a criminal sentence or lawful detention is not, merely for that reason, prohibited forced labor or involuntary servitude. Article 8(3)(a) of the ICCPR prohibits forced or compulsory labor, while Article 8(3)(b) expressly preserves hard labor imposed pursuant to a criminal sentence by a competent court. Article 4(3) of the European Convention excludes work normally required during lawful detention or conditional release from the definition of forced or compulsory labor. The Thirteenth Amendment similarly prohibits slavery and involuntary servitude except as punishment for crime after conviction. These provisions reflect the historical connection between imprisonment and compulsory prison work. They do not establish a separate rehabilitative measure equivalent to Taiwan's compulsory labor, and the United States does not impose a distinct compulsory-labor disposition in addition to the criminal sentence. Treating these provisions as if they were comparative-law counterparts to Taiwan's system is therefore misleading.

   For readers who wish to verify the materials directly, the relevant Chinese and English provisions are set out below:

   International Covenant on Civil and Political Rights, Article 8 1. No one shall be held in slavery; slavery and the slave-trade in all their forms shall be prohibited. 2. No one shall be held in servitude. 3. (a) No one shall be required to perform forced or compulsory labour; (b) Paragraph 3 (a) shall not be held to preclude, in countries where imprisonment with hard labour may be imposed as a punishment for a crime, the performance of hard labour in pursuance of a sentence to such punishment by a competent court; (c) For the purpose of this paragraph the term "forced or compulsory labour" shall not include: (i) Any work or service, not referred to in subparagraph (b), normally required of a person who is under detention in consequence of a lawful order of a court, or of a person during conditional release from such detention; (ii) Any service of a military character and, in countries where conscientious objection is recognized, any national service required by law of conscientious objectors; (iii) Any service exacted in cases of emergency or calamity threatening the life or well-being of the community; (iv) Any work or service which forms part of normal civil obligations.	International Covenant on Civil and Political Rights, Article 8 1. No one shall be held in slavery; slavery and the slave-trade in all their forms shall be prohibited. 2. No one shall be held in servitude. 3. (a) No one shall be required to perform forced or compulsory labour; (b) Paragraph 3 (a) shall not be held to preclude, in countries where imprisonment with hard labour may be imposed as a punishment for a crime, the performance of hard labour in pursuance of a sentence to such punishment by a competent court; (c) For the purpose of this paragraph the term "forced or compulsory labour" shall not include: (i) Any work or service, not referred to in subparagraph (b), normally required of a person who is under detention in consequence of a lawful order of a court, or of a person during conditional release from such detention; (ii) Any service of a military character and, in countries where conscientious objection is recognized, any national service required by law of conscientious objectors; (iii) Any service exacted in cases of emergency or calamity threatening the life or well-being of the community; (iv) Any work or service which forms part of normal civil obligations.
   European Convention on Human Rights Article 4 1. No one shall be held in slavery or servitude. 2. No one shall be required to perform forced or compulsory labour. 3. For the purpose of this article the term "forced or compulsory labour" shall not include: (a) any work required to be done in the ordinary course of detention imposed according to the provisions of Article 5 of this Convention or during conditional release from such detention; (b) any service of a military character or, in case of conscientious objectors in countries where they are recognised, service exacted instead of compulsory military service; (c) any service exacted in case of an emergency or calamity threatening the life or well-being of the community; (d) any work or service which forms part of normal civic obligations.	European Convention on Human Rights Article 4 No one shall be held in slavery or servitude. No one shall be required to perform forced or compulsory labour. For the purpose of this article the term "forced or compulsory labour" shall not include: any work required to be done in the ordinary course of detention imposed according to the provisions of Article 5 of this Convention or during conditional release from such detention; any service of a military character or, in case of conscientious objectors in countries where they are recognised, service exacted instead of compulsory military service; any service exacted in case of an emergency or calamity threatening the life or well-being of the community; any work or service which forms part of normal civic obligations.
   Constitution of the United States Amendment XIII Neither slavery nor involuntary servitude, except as a punishment for crime whereof the party shall have been duly convicted, shall exist within the United States, or any place subject to their jurisdiction.	The Constitution of the United States of America, Amendment XIII (1865) Neither slavery nor involuntary servitude, except as a punishment for crime whereof the party shall have been duly convicted, shall exist within the United States, or any place subject to their jurisdiction.

   The Ministry of Justice also invoked Article 61 of the Swiss Criminal Code, concerning placement of young adults in an institution and vocational training, but again did not identify the provision during oral argument. The relevant text is reproduced below:

   Swiss Criminal Code Article 61 If the offender was under 25 years of age at the time of the offence and suffers from a serious developmental disorder, the court may commit the offender to an institution for young adults if: a. the developmental disorder was a factor in the felony or misdemeanour committed; and b. the measure is expected to reduce the risk of further offences in which the developmental disorder is a factor. Institutions for young adults must be managed separately from other institutions and facilities under the Code. The offender should be taught the skills needed to live independently and without further offending. In particular, the offender should be encouraged to undergo basic and advanced vocational and professional training. The deprivation of liberty associated with the measure may last no more than four years. If the offender is recalled to custody following parole, the total may not exceed six years. The measure must be revoked no later than when the offender reaches the age of 30. If the offender was convicted of an offence committed before the age of 18, the measure may be implemented in an institution for minors.

   Swiss Criminal Code Art. 61 If the offender was under 25 years of age at the time of the offence and if he is suffering from a serious developmental disorder, the court may commit him to an institution for young adults if: a. the offender's developmental disorder was a factor in the felony or misdemeanour that he committed; and b. it is expected that the measure will reduce the risk of further offences being committed in which his developmental disorder is a factor. Institutions for young adults must be managed separately from other institutions and facilities under this Code. The offender should be taught the skills needed to live independently and without further offending. In particular, he should be encouraged to undergo basic and advanced vocational and professional training. The deprivation of liberty associated with the measure amounts to a maximum of four years. In the event of the recall to custody following release on parole, it may not exceed a maximum of six years. The measure must be revoked when the offender reaches the age of 30. If the offender was convicted of an offence committed before he was 18 years of age, the measure may be implemented in an institution for minors.

   The Swiss provision is not a close analogue to Taiwan's compulsory-labor regime. Its focus is on young adults whose offending is connected with a serious developmental disorder and on teaching the skills necessary for independent, law-abiding life. Taiwan's compulsory labor, by contrast, historically targeted people described as habitual offenders or as having offended because of habitual vagrancy or idleness and sought to compel them to acquire occupational skills in a correctional institution.

2. Whether Compulsory Labor Is Clearly Distinguished from the Execution of Imprisonment

   The Ministry of Justice argued that compulsory labor is meaningfully different from ordinary imprisonment. Two of the principal distinctions it advanced before the Constitutional Court were:

   • (1) Compulsory labor centers on "skill training." Except where factual circumstances make participation impossible, persons subject to compulsory labor must undergo vocational training intended to produce occupational certification. Ordinary prison work, the Ministry said, often consists only of simple labor such as folding paper boxes.
   • (2) Taiyuan Skill Training Institute devotes substantial resources to professional instructors, facilities, and equipment for compulsory labor-resources that, according to the Ministry, far exceed those available to ordinary inmates.

   Using those differences to defend the constitutionality of compulsory labor is deeply ironic. The state already has a responsibility to invest in meaningful prison work that supports reintegration. Taiwan has long underinvested in vocational training for ordinary inmates, leaving most people to perform simple, low-skill work while only a minority gain access to self-operated prison industries or genuine vocational training. The Ministry should improve prison work rather than cite its own underinvestment as a reason why compulsory labor deserves to survive.

   The second argument also raises an obvious question: what about the skill-training unit attached to Kaohsiung Women's Prison? If unusually well-funded vocational training at Taiyuan is what supposedly distinguishes compulsory labor from imprisonment and makes the former constitutional, does compulsory labor at Kaohsiung Women's Prison become constitutionally suspect simply because the same distinction is less visible there?

3. Whether Compulsory Labor Reduces Recidivism More Effectively Than Ordinary Imprisonment

   The Ministry of Justice also cited recidivism statistics suggesting that compulsory labor produces better outcomes than ordinary imprisonment. Put aside for the moment the absence of the rigorous research design ordinarily required to establish the effectiveness of a correctional intervention. Even if compulsory labor truly produced lower recidivism, the reason might simply be the Ministry's own description of the two regimes: people subject to compulsory labor receive vocational training, while ordinary inmates are often assigned low-skill work such as folding paper boxes. Appropriate vocational training that helps people reintegrate after release is already a core responsibility of prison administration. Underinvestment in prison work cannot serve as a constitutional justification for imposing a separate compulsory-labor measure.

4. Alternatives After Abolishing Compulsory Labor

   The Ministry of Justice proposed that, if compulsory labor were abolished, courts could instead increase the sentence on a case-by-case basis. That proposal itself underscores how difficult the Ministry finds it to identify a substantive distinction between compulsory labor and punishment.

 

Conclusion: Where Does Compulsory Labor Go from Here?

Compulsory labor may once have served particular functions in particular stages of social and economic development-for example, when states sought to cultivate a more productive workforce, or when industrialization created large demand for low-skill factory labor and compulsory work was used to train people in industrial schedules, discipline, and bodily routines. Taiwan is plainly no longer at that stage of development.

The deeper problem is that compulsory labor has always been difficult to separate from imprisonment. The modern prison itself absorbed the idea of reforming individuals through labor. As correctional theory changed, the religiously inflected notion that labor itself reforms character gave way to a focus on employment and vocational programs supported by empirical evidence. Training tied to contemporary labor-market needs can indeed be one element of an effective correctional program. But the same evolution has occurred within ordinary prison administration. Taiwan's prisons may still rely heavily on low-skill paper-processing work because of inadequate public investment, but that is a defect to be corrected-not an inherent institutional distinction between imprisonment and compulsory labor.

Finally, the legal design of compulsory labor has fallen behind the development of prison administration. Contemporary correctional theory calls for interventions of different types and intensities based on an individual's risk and needs. Vocational training may be one component, but it cannot be the entire treatment model. Even if we look only at employment, prison administration can provide vocational training near release and use staged reintegration mechanisms such as open prisons and self-managed outside-prison work. Those arrangements align far more closely with the evidence on effective employment programs than a compulsory-labor regime that, under current law, can only be executed before imprisonment.

Viewed from every angle, compulsory labor is a system left behind by the times. Whether-and how-it will survive constitutional review remains to be seen.

References

• Bonta, James, and D. A. Andrews. 2020. The Psychology of Criminal Conduct. Chinese translation by Chung Hung-Pin, Tung Tzu-Yi, Ku Yi-Chien, Tsou Hui-Ying, Tsai Hsi-Chen, Huang Hsiao-Wen, Peng Wei-Ning, and Hsia Mei-Ting. Taipei: Yeh Yeh Book Gallery.
• Foucault, Michel. 2020. Discipline and Punish: The Birth of the Prison. Chinese translation by Wang Shao-Chung. Taipei: China Times Publishing.
• Ho Lai-Chieh and Lin Yu-Hsiung, review editors; Li Sheng-Chieh and Pan Chih-Hung, compilation editors; Wang Shih-Fan et al., translators. 2017. German Criminal Code. Taipei: Angle Publishing.
• Hsu Heng-Da. 2013. "On Compulsory Labor as a Rehabilitative Measure." Taiwan Law Review, no. 214: 193-212.
• Bushway, Shawn D. 2003. "Reentry and Prison Work Programs." In Employment Dimensions of Reentry: Understanding the Nexus Between Prisoner Reentry and Work.
• Cullen, Francis T., and Paul Gendreau. 2001. "From Nothing Works to What Works: Changing Professional Ideology in the 21st Century." The Prison Journal 81(3): 313-338.
• Pierson, Anthony, Keith Price, and Susan Coleman. 2014. "Prison Labor." Politics, Bureaucracy, and Justice 4(1): 12-23.
• Smith, Paula, Lindsey M. Mueller, and Ryan M. Labrecque. 2018. "Employment and Vocational Programs in Prison." In The Oxford Handbook of Prisons and Imprisonment, ed. John Wooldredge and Paula Smith, 454-472. New York: Oxford University Press.