Annual Review | Commentary on the 2021 Q1-Q3 Correctional External Inspection Reports and the Establishment of External Inspection Panels

Introduction

The amended Prison Act and Detention Act took effect on July 15, 2020. One important feature of the amendments was the requirement that every prison and detention center (collectively, the "institutions," or, where appropriate, "correctional institutions") establish an external inspection panel. The system received particular attention during the reform process because correctional institutions are closed facilities: outsiders have limited visibility into their internal operations, and problems may therefore remain unknown outside the institution. The amended Acts sought to introduce independent external inspection panels to increase transparency and protect the rights of inmates and defendants. Issuing inspection reports is one of the panels' core duties. Through those reports, the public can better understand actual conditions inside correctional institutions and can also assess how the external inspection panels themselves are operating.

By law, each institution's external inspection panel must issue a report every quarter. The inaugural panels first issued reports for Q4 2020, which were published on February 8, 2021. On March 25, 2021, JRF published its "Commentary and Recommendations on the Q4 2020 External Inspection Reports for Prisons and Detention Centers." However, the external inspection reports for Q1 through Q3 2021 were not all released until December 2021, when they were published at once. Civil-society organizations therefore had to wait months before they could use the reports to assess how the panels were operating. Expectations were high, and the wait was long. Some reports were clearly prepared with considerable care, but many others were disappointing in how little they conveyed.

The panels have now been operating for more than a year, with less than a year remaining in their current terms. This is therefore an appropriate time to review their performance and look ahead. JRF offers the following observations on the Q1-Q3 2021 external inspection reports and the operation of the panels.

1. Toward More Independent External Inspection Panels

   Assessing Whether Panels Are Genuinely External and Independent: Member Composition

   The law requires each institution's external inspection panel to be external and independent, but the composition of the current panels suggests that this design has not been realized. According to data provided by Legislator Chiu Hsien-Chih, as of August 30, 2021 there were 253 external inspection panel members. Of those, 180 had been "recommended by the institution," while 51 were not included in the Expert and Scholar Talent Pool for External Inspection Panels at all. In other words, as many as 91% of members had not entered the talent pool through outside recommendations. Of the 51 correctional institutions, 34-67%-had panels composed entirely of people selected by the institutions themselves, with no externally recommended member.

   This outcome stems from the design of the member-selection mechanism under the Regulations for Implementing the External Inspection Team of the Prison and the Detention Center. Article 4, Paragraphs 3 and 4 require the Agency of Corrections, Ministry of Justice to establish an expert-and-scholar talent pool for external inspection panels and to solicit recommendations broadly from relevant agencies and institutions, schools, and professional organizations, as well as self-nominations from qualified experts and scholars, as a reference for selecting panel members. Article 5, Paragraph 1 provides that an institution is to submit a proposed list of appointees, which is then forwarded by the supervisory authority to the Ministry of Justice for approval and appointment. In practice, this has been interpreted to allow correctional institutions to prepare their own nominee lists. They may draw from the talent pool, but they may also identify candidates outside the existing pool and, at the same time, ask those candidates whether they agree to be added to it. Article 5, Paragraph 2 further provides that the Agency of Corrections is not bound by the institution's proposed list when it submits prospective panel members to the Ministry of Justice. The data nevertheless show that most correctional institutions have relied on their own networks to identify experts and scholars they consider suitable, and that the Agency of Corrections has generally allowed those nominations to proceed. As a result, outside recommendations to the talent pool have little practical effect, and correctional institutions exercise overwhelming influence over panel selection.

   We believe this design is problematic as a matter of ordinary human behavior: members who are supposed to be external and independent may be selected by the very institution they are to inspect, without an effective mechanism for examining the relationship between the institution and the member. It is only natural for an institution to prefer people it knows and finds easy to work with. The current selection process therefore cannot ensure that panel members are genuinely external and independent. With less than a year remaining before the next term must be selected, the Ministry of Justice should promptly amend the member-selection provisions of the Regulations for Implementing the External Inspection Team of the Prison and the Detention Center so that the next generation of panels can be more independent and transparent.

   Assessing Whether Panels Are Genuinely External and Independent: Who Writes the Inspection Reports?

   The statutory design assumes that external inspection panels will be external and independent. Because inspection reports are one of the panels' core duties, those reports should be written by panel members themselves, not by the institutions on their behalf. If a panel described as external and independent can delegate even this core task to the institution it oversees, how can it persuade the public that it is truly external and independent?

   The current inspection-report template places "Inspection Content and Handling" alongside "Institutional Handling" in the same table. That format necessarily requires the institution to write part of the document, making it difficult to tell which portions were written by the external inspection panel and which by the institution.

   Given this limitation, we would at least expect the "Inspection Content and Handling" section to be written by the external inspection panel members themselves. Yet some reports plainly read as institutional records of panel members' activities and statements; in others, it is difficult to tell whether the text was written by the panel or by the institution on the panel's behalf. Consider the following example:

   "The members heard operational reports from Chiayi Prison's various divisions concerning the inspection priorities for Q2 and Q3, and accepted questions from the members." The first half of the sentence records the inspection activity from a third-person perspective, while the final phrase-"accepted questions from the members"-reads as though the institution, rather than the panel, is speaking.

   This kind of shifting voice is not uncommon. At times the text reads as a recommendation made from the external inspection panel's perspective; at other times it sounds like a third party recording the panel's activities or members' remarks; elsewhere it reads like an explanation by the institution. The result is confusion over who actually wrote the inspection report. Because current reports do not identify their authors, we cannot determine with certainty whether institutions are writing reports on behalf of the panels. JRF raised the same concern in its "Commentary and Recommendations on the Q4 2020 External Inspection Reports for Prisons and Detention Centers," but the problem has persisted.

   Because inspection reports do not currently identify their authors, we cannot confirm whether correctional institutions are in fact writing reports on behalf of external inspection panels. The competent authorities should investigate and explain whether this practice exists, so that public concerns can be resolved.

   We reiterate that external inspection reports should be written by the panel members themselves. Under Article 7, Paragraph 3 of the Prison Act and Article 5, Paragraph 3 of the Detention Act, a panel must issue a report each quarter, after which the relevant authorities respond and take follow-up action. "Institutional Handling" is therefore part of the authorities' response after a report has been issued, not part of the inspection report itself. Because the report is to be prepared and submitted by panel members, its template should not include a section such as "Institutional Handling," which members cannot reasonably be expected to complete. Institutions may, of course, still assist with minutes of inspection-panel meetings and provide presentations or other materials, and the panel may incorporate excerpts from those materials into its report as appropriate.

2. Panel Members Should Disclose Their Names and Reports Should Explain How the Panels Operate

   A basic function of institutional external inspection panels is to promote transparency. Disclosing panel members' names is an important part of that transparency and reflects a willingness to take public responsibility for the report under one's own name. This is therefore one of the points we focus on when reading inspection reports.

   None of the Q4 2020 inspection reports disclosed panel members' names. In the Q1-Q3 2021 reports, 24 panels still disclosed no member names, but 10 disclosed some names and 17 disclosed all names. The respective shares were 47%, 20%, and 33%.

   We commend panel members who are willing to disclose their names and encourage more external inspection panel members to do the same.

   Beyond disclosure of member names, some Q1-Q3 2021 reports also describe how the panels operate and the difficulties they encounter during inspections, which helps make the system itself more transparent. Taiyuan Skill Training Institute's Q1 2021 report devotes substantial space to improving panel effectiveness and operations, including advance preparation for inspection meetings and the management and preservation of inspection-related materials. Those details help the public understand how the panel works. Dun Pin High School's Q1 2021 report (the school was formerly the Taoyuan Branch of Chengjheng High School) states that panel members sought to select students at random for interviews, but the institution refused to allow interviews with students whose parents had not given prior consent, explaining that all students were juveniles and that parental consent was legally required. This too helps the public understand practical obstacles encountered during inspection.

   Very few panels explain in their reports either how they operate or what difficulties arise during inspection, but both are important to the public's understanding of the external inspection system. We encourage more panels to address these matters in their reports.

3. Inspection Reports Should Advance Correctional Transparency

   A central task of an external inspection panel is to use its reports to show the public what conditions are actually like inside the institution. The reports are valuable precisely because they come from an external, independent body that can describe those conditions from outside the institution and thereby improve public understanding of correctional facilities.
   Yet many current inspection reports reveal little about what information the panel actually obtained through its inspections.

   Keelung Prison's external inspection report is one example. Under "Inspection Content and Handling," it states that, regarding guidance for housing prisoners with HIV, "after the meeting, the panel will propose that the Agency of Corrections, Ministry of Justice establish uniform and detailed standards for all correctional institutions in order to improve the treatment of prisoners with HIV," and that "our current review found no problem with the institution's application of the Reference Guidelines on Reasonable Accommodation for Inmates with Disabilities in Correctional Institutions; any future questions can be discussed as they arise." The report, however, does not make clear what information the Keelung Prison panel actually obtained through inspection that led it to those conclusions.

   Other reports do make clear what information panel members obtained through inspection. The reports issued by the external inspection panels of Yunlin Prison and Dun Pin High School are useful examples.

   Yunlin Prison's Q1 2021 report, for example, summarizes information obtained from interviews with the head of the work section and with inmates who actually participated in self-managed outside-prison work. In Q2 and Q3, the panel also attempted to move beyond the constraints of the prescribed report format and present its inspection findings in the form it considered most appropriate.

   Dun Pin High School's Q1 2021 report shows the panel attempting to verify information supplied by the institution through multiple sources. The institution had stated on February 24 that when a student committed a rule violation, a counseling teacher would interview the student and a record would be made. During an on-site inspection, the panel selected two disciplinary cases at random and asked the institution to produce the counseling records for both. It found that one case contained only a group-counseling record and no record of individual case counseling. The institution explained that, under prior practice, only instructors and specialized counseling staff kept written records, while homeroom teachers had been required to create written records of counseling for students who violated rules only since March of that year. The report clearly sets out both the panel's verification process and the institution's response, making it particularly useful.

   We hope more external inspection panels will use their external and independent position to explain what they actually learned through inspection, so that the public can better understand conditions at each institution. We also encourage panel members to read reports from other institutions and learn from those that have been prepared with particular care.

4. External Inspection Should Rely More Heavily on Interviews with Inmates and Defendants

   External inspection panels may conduct on-site inspections, interviews, requests for written opinions, document review, and other forms of inspection. We are particularly concerned with whether panels interview the people confined in the institutions, because such interviews are among the most basic and important tools of correctional external inspection. Yet among the Q1-Q3 2021 reports, only seven panels-14%-mentioned that members had interviewed inmates or defendants. Most panels appeared to rely mainly on institutional briefings, question-and-answer sessions with staff, and on-site inspections accompanied by institutional personnel, leaving them with few independent sources of information.

   The handling of petitions likewise shows that some panels give insufficient weight to collecting the experiences and views of people in custody. JRF once forwarded to Hsinchu Prison's external inspection panel an inmate's petition concerning poor ventilation and unbearable summer heat in the housing units. In the forwarding letter, JRF suggested that the panel interview the petitioner and other inmates to understand their experiences and views. The panel's Q3 2021 report instead stated: "Based on an on-site inspection of the housing units and a review of the investigation report and photographs, ventilation in the prison housing units appears to comply with Article 52, Paragraph 1 of the Prison Act, which requires housing units, workplaces, and other areas to maintain the space, lighting, and ventilation necessary for health. From an objective perspective and from the standpoint of the inmate population as a whole, this panel considers the reasonableness of the submission. If it reflects only an inmate's subjective perception, counseling by the prison may suffice; there is therefore no need to interview the petitioner surnamed Chou or other inmates." The panel thus treated a one-time inspection conducted in the presence of prison staff, together with the prison's own explanation, as sufficient to complete its inquiry. It did not take account of the importance of interviews for gathering inmates' experiences and views or of the need to verify inspection information through multiple methods.

   By contrast, Dun Pin High School's external inspection report shows that members interviewed detained students before the COVID-19 situation worsened and, even after Taiwan's domestic outbreak, continued selecting students at random and interviewing them by video. The panel's emphasis on interviews-and its effort to preserve them during the pandemic-is a practice worth learning from.

   We urge external inspection panel members to give appropriate weight to interviews with inmates and defendants. First, the people confined in an institution are an essential part of that institution, and their experiences and views deserve attention. Second, panels currently rely too heavily on information supplied by the institution and therefore need independent means of verifying what they are told.

5. The Receipt and Handling of Petitions Need to Be Strengthened

   Article 92, Paragraph 1 of the Prison Act and Article 84, Paragraph 1 of the Detention Act provide that inmates and defendants may petition an external inspection panel. Receiving a petition is, in substance, a way for the panel to receive views and complaints directly from people in custody; handling a petition is the action the panel takes after receiving those views. These processes therefore show how a panel carries out its statutory duty to protect the rights of inmates and defendants.

   Among the Q1-Q3 2021 reports, only four panels-8%-received petitions specifically directed to the external inspection panel. Although few panels received such petitions, five discussed how petitions should be received. The Zihciang Minimum-Security Prison panel discussed establishing a postal mailbox and an electronic complaint mailbox; the Keelung Prison panel asked the prison to inform inmates that they could request an in-person meeting with an external inspection panel member; and the Taiyuan Skill Training Institute panel resolved that submissions placed in the complaint box and addressed to the external inspection panel would not be opened by the institution but would instead be sent by express registered mail to the panel convener for handling. It also asked the institution to announce inspection dates in advance by appropriate means so that inmates could petition the panel in person. These discussions show a willingness by some panels to collect inmates' views proactively.

   The Q1-Q3 2021 reports also reveal petition-handling procedures that may be inconsistent with governing regulations. In Kaohsiung Women's Prison's Q3 2021 report, inmate Wen ○-Ting sought to make an oral petition to panel members regarding what she regarded as unfair treatment resulting from a complaint lodged against her by a cellmate acting out of a grudge. The institution investigated the matter first and then reported the result at the external inspection panel meeting; the panel subsequently decided that the case did not concern the institution's overall operations and recommended that it be handled through internal prison procedures. Article 16 of the Regulations for Implementing the External Inspection Team of the Prison and the Detention Center, however, provides that after receiving a petition under Article 92 of the Prison Act or Article 84 of the Detention Act, the external inspection panel may decide-based on factors including the specificity of the petition, whether it falls within the panel's authority, its relationship to the inspection priorities, and the panel's staffing capacity-whether to handle the matter through external inspection procedures or refer it to the institution. The regulation therefore contemplates that the panel first receives the petition and then decides whether to handle it itself or refer it to the institution, not that the institution investigates first and reports back so that the panel can simply endorse the institution's handling.

   We hope more external inspection panels will actively receive and handle petitions and will do so in accordance with the governing rules.

Appendix: Summary of External Inspection Reports by Institution

Correctional Institution	No. of Reports	Name Disclosure Status	Inspection Methods (Excluding Petitions)	Reception and Handling of Petitions	Recommended?
Taipei Prison	2	Surnames only	Briefings; on-site inspections; Q&A with institution staff	No petitions
Taoyuan Prison	3	Surnames only	Q&A with institution staff	No petitions
Bade Minimum-Security Prison	3	One member disclosed their name; two declined to disclose theirs.	Briefings; on-site inspections; review of written materials	1 petition, although it is unclear whether it was submitted to the external inspection panel (Q1 2021)
Taoyuan Women's Prison	2	Surnames only in Q1 2021; full names listed in Q3 2021	Briefings; on-site inspections; brief conversations with prisoners accompanied by their children; Q&A with institution staff	No petitions
Hsinchu Prison	2	Full names listed	Unclear from the report; only recommendations are described.	1 petition (forwarded by JRF). Handling consisted of an on-site inspection of the housing units and review of the investigation report and photographs; the panel declined to interview the petitioner or other inmates about whether ventilation was inadequate and summer heat unbearable. The report states: "From an objective perspective and from the standpoint of the inmate population as a whole, this panel considers the reasonableness of the submission. If it reflects only an inmate's subjective perception, counseling by the prison may suffice; there is therefore no need to interview the petitioner surnamed Chou or other inmates."
Taichung Prison	2	Full names listed	On-site inspections; interviews with division heads and responsible staff. The report repeatedly says the panel "learned" or "understood" matters without explaining how the information was obtained or what it consisted of.	No petitions
Taichung Women's Prison	2	Surnames only	On-site inspections. Because the reports focus mainly on recommendations, it is often unclear what inspection activities were actually carried out.	No petitions
Changhua Prison	2	All members declined to disclose their names.	On-site inspections. The reports generally say only that the panel "learned about" matters from the prison, making the actual inspection methods unclear.	5 petitions; petitioners were not interviewed.
Yunlin Prison	3	Five members disclosed their names; one declined to disclose theirs.	On-site inspections; requests for materials; interviews with staff and inmates; Q&A with institution staff	No petitions	Highly recommended. 1. Clearly organizes and presents information obtained through inspection from the perspective of outside panel members. 2. The appendices to the Q2 and Q3 2021 reports are particularly recommended. They appear closer to the panel's original reports but could be included only as appendices because their format did not conform to the format prescribed by the Agency of Corrections.   Links: Q1 2021 Report Q2 2021 Report Appendix Q3 2021 Report Appendix
Yunlin Second Prison	3	Surnames only	Briefings; on-site inspections; Q&A with institution staff	No petitions
Chiayi Prison	2	Four members disclosed their names; one member's surname only was shown.	Briefings; Q&A with institution staff	No petitions
Tainan Prison	2	Surnames only	Q&A with institution staff; on-site inspections	No petitions
Tainan Second Prison	2	Surnames only	Q&A with institution staff; requests for materials; on-site inspections	No petitions
Mingde Minimum-Security Prison	2	Full names listed	Q&A with institution staff	No petitions
Kaohsiung Prison	2	Full names listed	Q&A with institution staff; on-site inspections	No petitions
Kaohsiung Second Prison	2	Full names listed	Unclear from the report.	No petitions
Kaohsiung Women's Prison	2	Surnames only	On-site inspections	1 petition. The institution investigated first and reported the matter at the external inspection panel meeting; the panel then decided that the case did not concern the institution's overall operations and recommended handling it through internal prison procedures.
Pingtung Prison	2	Full names listed	"On-site written inspection" (it is unclear whether this means an on-site inspection combined with review of written materials, or only reviewing materials at the institution without inspecting the site).	No petitions
Taitung Prison	2	Surnames only.	On-site inspections; Q&A with institution staff	No petitions
Hualien Prison	3	Four members disclosed their names; one member's surname only was shown.	Briefings; Q&A with institution staff	No petitions
Zihciang Minimum-Security Prison	2	Full names listed	Still discussing effective inspection methods	No petitions. The panel discussed establishing a postal mailbox and an electronic complaint mailbox for receiving petitions.
Yilan Prison	2	Full names listed	Unclear from the report.	No petitions
Keelung Prison	2	Full names listed	Briefings; interviews with inmates	No petitions. The panel discussed how petitions could be received, including allowing inmates to request an in-person meeting.
Penghu Prison	1	Surnames only	On-site inspections	No petitions
Green Island Prison	2	Three members initially disclosed their names, but one name was later concealed again. Two members' given names were partially masked with "O" for de-identification.	On-site inspections; document review	No petitions
Kinmen Prison	1	Surnames only	Unclear from the report.	No petitions
Taipei Detention Center	2	Surnames only in Q1 2021; full names listed in Q3 2021	Q&A with institution staff; other methods are largely unclear from the report.	No petitions. The report mentions establishing an external inspection comment box so that petitions can be monitored on an ongoing basis.
Taipei Women's Detention Center	2	Surnames only	Q&A with institution staff; on-site inspections	No petitions
Hsinchu Detention Center	3	Surnames only	Written materials provided in advance by the institution; on-site inspections	No petitions	Recommended. The report indicates that the institution provides panel members with written materials in advance so they can prepare before the meeting-an approach worth encouraging. Link: Q2 2021 Report
Miaoli Detention Center	2	Full names listed	On-site inspections; other methods are largely unclear from the report.	No petitions
Taichung Detention Center	2	Surnames only	Briefings; on-site inspections; interviews with inmates	No petitions
Changhua Detention Center	2	All members declined to disclose their names.	Briefings; on-site inspections	No petitions
Nantou Detention Center	2	All members declined to disclose their names.	Q&A with institution staff; other methods are largely unclear from the report.	No petitions
Chiayi Detention Center	2	Full names listed	On-site inspections; "inquiries into handling procedures" as an inspection method (meaning unclear).	No petitions
Tainan Detention Center	3	Full names listed	Briefings; Q&A with institution staff; on-site inspections	No petitions	Recommended. The "Inspection Content and Handling" section presents the information obtained through inspection in fairly concrete terms. Link: Q3 2021 Report
Pingtung Detention Center	2	Full names listed	Q&A with institution staff; on-site inspections	No petitions
Hualien Detention Center	3	Three members disclosed their names in Q1 2021; full names were listed in Q2 and Q3 2021.	Q&A with institution staff; other methods are unclear.	No petitions
Keelung Detention Center	3	One member disclosed their name; four did not disclose theirs.	Briefings; Q&A with institution staff; on-site inspections	No petitions. The panel discussed petition boxes and decided that inmates could use the existing comment boxes in each housing area rather than establishing a dedicated external inspection panel box.
Sindian Drug Abuser Treatment Center	2	Surnames only	Q&A with institution staff; on-site inspections	No petitions
Taichung Drug Abuser Treatment Center	2	One member disclosed their name in Q1 2021; by Q3 2021 only surnames were shown.	On-site inspections; Q&A with institution staff; consultations with division heads	No petitions
Kaohsiung Drug Abuser Treatment Center	1	All members declined to disclose their names.	On-site inspections; interviews with inmates. Inmates volunteered for interviews through an open sign-up process.	No petitions
Taitung Drug Abuser Treatment Center	2	Surnames only	On-site inspections; Q&A with institution staff	No petitions
Taiyuan Skill Training Institute	2	Four members disclosed their names in Q1 2021; full names were listed in Q3 2021.	Q&A with institution staff; requests for materials. The panel planned to select and interview inmates with different types of disabilities in November.	1 petition; the petitioner was interviewed. In Q1 2021, the panel resolved that, to allow inmates to petition without concern, written submissions could be mailed to Taiyuan Skill Training Institute or placed in a complaint box with the external inspection panel designated as the recipient. The institution would not open such submissions but would send them by express registered mail to the panel convener for handling. To allow inmates to petition or communicate with the panel in person, the institution was also asked to announce inspection dates in advance by appropriate means and arrange meetings as needed.	Recommended: the Q1 2021 inspection report. It records resolutions concerning inspection procedures and methods, helping the public understand how the panel operates. Link: Q1 2021 Report
Dongcheng Skill Training Institute	3	Surnames only	Briefings; Q&A with institution staff; on-site inspections	No petitions
Yanwan Skill Training Institute	1	Surnames only	On-site inspections; other methods are unclear from the report.	No petitions
Taipei Juvenile Detention Center	3	Full names listed in Q1 and Q2 2021; names not disclosed in Q3 2021	On-site inspections; Q&A with institution staff. Other methods are unclear from the report.	No petitions	Recommended: the Q2 2021 report. The "Inspection Content and Handling" section presents the information obtained through inspection in fairly concrete terms. Link: Q2 2021 Report
Tainan Juvenile Detention Center	2	Surnames only	On-site inspections; requests for materials	No petitions	Recommended. The "Inspection Content and Handling" section presents the information obtained through inspection in fairly concrete terms. Link: Q3 2021 Report
Chengjheng High School	2	Surnames only	Unclear from the report.	No petitions
Ming Yang High School	2	Two members disclosed their names; one declined to disclose theirs.	On-site inspections; Q&A with institution staff	No petitions
Dun Pin High School (formerly the Taoyuan Branch of Chengjheng High School)	3	Full names listed	Requests for materials; on-site inspections; Q&A with institution staff. The panel also interviewed students, teachers, pharmacists, security officers, and other personnel.	No petitions	Highly recommended. 1. Uses diverse inspection methods and gives significant weight to interviews. 2. Presents concrete information obtained by the external inspection panel. 3. Describes difficulties encountered during inspection.   Links: Q1 2021 Report Q3 2021 Report
Li Zhi High School (formerly the Changhua Branch of Chengjheng High School)	2	Surnames only	Briefings; on-site inspections	No petitions