Flaws in Forensic Science and Criminal Investigation Exposed by the Chiang Case: Comparing the Reasoning in the Guilty and Not-Guilty Verdicts

On September 13, 2011, Chiang Kuo-Ching's acquittal became final. Pursuant to Article 225 of the Military Trials Act mutatis mutandis with Article 440 of the Code of Criminal Procedure (Note 1), the Ministry of National Defense immediately published the full text of the retrial acquittal judgment in the Executive Yuan Gazette and posted it online (Note 2). Following a final death sentence in a military trial, overturning the case through retrial proceedings made Chiang Kuo-ching the first such instance in 60 years.

This was the first time the public could directly read an authentic military justice judgment. What judicial flaws did it expose, and how did it explain the causes of the error? The author recommends that all friends concerned with judicial justice browse it online themselves, as follows:

Document Type	Search Website	Search Method
Retrial Acquittal Judgment	Executive Yuan Gazette Information Network http://gazette.nat.gov.tw	Enter the keyword "Chiang Kuo-ching" in quick search
Original Guilty Judgment	Judicial Reform Foundation http://www.jrf.org.tw	Enter the keyword "Chiang Kuo-ching Case" in full-text search

This article introduces to the public the critical yet little-known issues of forensic and criminal investigation. It calls on the government to learn its lesson, weed out unqualified personnel, and establish forensic and criminal investigation systems that meet international standards.

The Root of Error: Lie Detector Testing

"Pity it is that the defendant in this case, Chiang Kuo-Ching, was preferentially targeted as a major suspect back then simply because he failed the Investigation Bureau's lie detector test. The reasons why the defendant was originally convicted were none other than an over-reliance on the defendant's confession, and the truthfulness of that confession was not verified purely through forensic evidence, thereby giving rise to unreliable corroborating evidence..." (Note 3)

The acquittal directly pointed out that the erroneous lie detector test conducted by the Investigation Bureau on October 1, 1996, led military personnel to resort to torture and coercion to extract confessions, and further influenced subsequent forensic science to make "unreliable" judgments. In contrast, the guilty judgment fully endorsed the lie detector test and cited it as a legitimate basis for breaking the defendant's psychological defenses. What was considered right yesterday is wrong today—how lamentable!

Lie detector tests may seem to use many scientific instruments, but in reality, lie detector results cannot be replicated or verified at all, because every test yields different reactions, and question design and numerical interpretation rely almost entirely on the subjective discretion of the examiners. Therefore, "German practical principles prohibit any form of lie detection on defendants primarily because lie detection inevitably affects the defendant's 'freedom of decision and activity of will.' Because the principle of lie detection relies on probing information that the tested person may wish to conceal through involuntary and uncontrollable physiological reactions, German case law holds that this violates personal freedom." (Note 4) Furthermore, other advanced judicial nations are equally strict and cautious regarding the use of lie detector tests.

The current severity of the Taiwanese legal community's abuse of lie detector tests far exceeds societal imagination. Among the 12 cases finalized with death sentences by the Supreme Court this year, at least 7 involved lie detector tests (Note 5). In 3 of these cases, the lie detector results were unfavorable to the defendant and were cited as proof of guilt; in the other 4 cases, the results were favorable to the defendant but were not cited as proof of innocence. The approach of judicial practice in adopting lie detector tests is unscientific, unfair, and even violates the presumption of innocence.

The Error That Should Least Have Happened: Fingerprint and Palm Print Analysis

On September 12, 1996, the day the incident occurred, 46 fingerprints and palm prints were recovered from the scene and submitted to the Criminal Investigation Bureau (CIB) of the National Police Agency for analysis. On November 1, 1996, the CIB replied that no matches were found with the fingerprints and palm prints collected from the scene. Thereafter, the bureau was repeatedly asked to re-examine the prints and consistently insisted that no such person was found. It was not until October 6, 2010, that a prosecutor from the Taichung District Prosecutors Office issued an appraisal permit to re-collect the palm prints of suspect Hsu Jung-Chou (許榮洲) that the CIB admitted its mistake, overturned its original conclusion, and confirmed that the 42nd palm print belonged to suspect Hsu Jung-chou.

To ascertain the reasons for the fingerprint and palm print analysis errors, the Taipei District Prosecutors Office summoned the original CIB analysts to testify in court on February 17, 2011. The acquittal judgment records their testimony:

"Initially in this case, the Criminal Investigation Division of the Taipei City Police Department submitted the rolled palm prints for comparison, but most of them had omissions or deficiencies. Some lacked the upper half of the palm, and some were unclearly printed. Because the focus at the time was primarily on comparing Chiang Kuo-ching, if the subject was not a specific target, we would not request a re-sampling when their palm prints were unclearly printed or had omissions..." (Note 6)

From the above testimony, it can be seen that the problem stemmed from the laziness and negligence of the analysts, who failed to re-sample unclear fingerprints and palm prints for comparison in accordance with regulations, thereby letting Hsu Jung-chou's palm print slip through. This was an error that should least have happened. Ensuring that fingerprint and palm print samples are complete, clear, and match the body parts left at the scene is the most basic precaution in criminal forensics.

Currently, domestic fingerprint databases are uniformly managed by the CIB, yet the flaws exposed by the Chiang Kuo-ching case have still not been reviewed or held accountable. Taking the case of Chiu Ho-Shun (邱和順), whose death sentence was finalized this year as an example, 7 fingerprints of the perpetrators were collected from the ransom envelope, but it remains unknown to whom they belong to this day. Is the CIB's long-standing conclusion that no such person could be found reliable? The current mechanism makes verification fundamentally impossible. If human negligence like that in the Chiang Kuo-Ching case were to happen again, besides the general public once again bearing high compensation costs, the dead cannot be brought back to life—who can compensate an innocent person for their life!

Errors Driven by Preconceived Conclusions: DNA, Cutting Tools, and Bloodstain Analysis

The main evidence on which the guilty judgment relied to convict Chiang Kuo-ching consisted of: 1. Tissue paper stained with the victim's blood and Chiang Kuo-Ching's semen. 2. A serrated knife stained with human blood. 3. Military trousers stained with human blood. These pieces of evidence were all augmented by the halo of scientific analysis, covering up society's doubts at the time about the military's torture and coercion of confessions. Today, the acquittal strips away the truth behind these "sciences":

"Judging from the bloodstain splatter pattern on tissue paper No. 3-11-1, this tissue paper was originally inside packaging. After the victim was attacked, blood was spattered onto the wall and onto the opened tissue paper packaging on the ground. Chiang Kuo-Ching subsequently used it, thereby leaving Chiang Kuo-Ching's biological traces on the same piece of tissue paper..." (Note 7)

The truth turned out to be that when the murderer killed the victim, trace amounts of blood splattered into the packaging of clean tissue paper placed inside the restroom. Chiang Kuo-Ching, who went to the restroom afterward, pulled out and used tissue paper from this package stained with trace blood and discarded it in the wastebasket. After the incident, forensic and investigative personnel detected Chiang Kuo-Ching's DNA from this tissue paper, and thus the most aggrieved scapegoat was created.

Investigation Bureau DNA Type Table of October 7, 1996 (Note 8)

Type	DQα	LDLR	GYPA	HBGG	D7S8	GC
Tissue Paper 11-1	1.1,3,4	BB	AB	AB	AB	ABC
Victim	3,3	BB	AA	BB	AA	AB
Chiang Kuo-ching	4,4	BB	AA	AB	AB	AC

Note: Highlighted sections represent DNA types belonging to neither the victim nor Chiang Kuo-Ching.

This was not an unavoidable error, because after this tissue paper was crumpled and thrown into the wastebasket, it was also contaminated with someone else's feces or bodily fluids. The Investigation Bureau's DNA analysis results back then mismatched the defendant's or victim's DNA in at least two places (see the table above), meaning that the traces shown on the tissue paper must have included the DNA of a third, or even a fourth person. Calculated using Taiwan's population genetic statistical data, out of every 4 "three-person combination DNAs," 1 set of the DNA type detected by the Investigation Bureau would appear (Note 9). According to genetic analysis standards, this tissue paper should not have been used to prove Chiang Kuo-Ching's crime, but the Investigation Bureau skillfully worded the analytical conclusion as: "The DNA mixture type presented by exhibit No. 11-1 (tissue paper) includes the DNA of the victim and the DNA type of suspect '18-J' (Chiang Kuo-Ching)." This presupposed "inclusion" caused Chiang Kuo-Ching to take his grievances to the grave.

"One fingerprint collected from the seized serrated knife was found, upon analysis, to match the right thumbprint of Yen Ko-Fu (顏柯夫), and no traces related to Chiang Kuo-Ching were collected. This is available for reference in CIB Fingerprint Analysis Report No. 506 dated October 24, 1996..." (Note 10)

Although as early as October 24, 1996, a fingerprint belonging to someone other than Chiang Kuo-Ching had already been discovered on the serrated knife, the ROC Armed Forces Forensic Center concealed the fact that there were other people's fingerprints on the knife, drew the analytical conclusion that the serrated knife was the murder weapon, and repeatedly defended this conclusion until the flaws were exposed during the retrial. Criminal analysis that intentionally ignores physical evidence favorable to the defendant is not science. True science cannot explain why Chiang Kuo-Ching managed to leave someone else's fingerprints while washing away his own.

"The military trousers worn by the defendant that were seized were stated by him to have been worn on the day of the incident, and he also claimed that the bloodstains on these military trousers were stained when he stabbed the victim with a knife. However, after the CIB (hereinafter referred to as the Criminal Investigation Bureau) re-examined these military trousers, only the defendant's DNA was detected, and the victim's DNA was not detected. Thus, it is evident that the defendant's confession indeed does not match the facts."

Fifteen years ago, the CIB rashly inferred from very weak chemical reactions that there was human blood on the serrated knife and military trousers. Fifteen years later, new technology proved that there was no human blood on the serrated knife, and while there was human blood on the military trousers, it was Chiang Kuo-Ching's own blood. Admittedly, technology 15 years ago was insufficiently advanced to perform more precise analyses. However, what killed Chiang Kuo-ching was not technology, but rather the analysts' violation of the scientific rule to "speak only to the evidence at hand," exaggerating the credibility of their experiments and downplaying the possibility of error, thereby committing a grave mistake.

The definition of science is that experiments conducted through strict scientific methods yield conclusions within an effective scope where no counter-examples are found, and possesses repeatability. Conversely, pseudoscience refers to claims of scientific backing where the asserted conclusions are usually based on preconceived stances, selecting only one among multiple possible interpretations while failing to rule out counter-examples or achieve repeatable verification. The forensic and criminal analysis errors in the Chiang case were victimized by pseudoscience masquerading as science. The main forensic flaws and responsible analytical units are summarized as follows:

Forensic Item	Finding in Acquittal Verdict	Finding in Guilty Verdict	1996 Analytical Unit	2011 Analytical Unit
Lie Detector Test	Lie detector misled the direction of the investigation	Lie detector uncovered the defendant's heavy involvement	Investigation Bureau	None (re-analysis impossible as the subject was already executed)
Palm Print	Palm print No. 42 belonged to Hsu Jung-Chou	Failed to account for the unidentified person's bloody palm print at the scene	Criminal Investigation Bureau	Criminal Investigation Bureau
Semen	No sperm of the defendant found	Contained the defendant's semen	ROC Armed Forces Forensic Center Investigation Bureau	Criminal Investigation Bureau Institute of Forensic Medicine, Ministry of Justice
Tissue Paper	Bloodstains were spattered while the tissue paper was inside its packaging; Chiang Kuo-ching subsequently used it	Stained with blood after wiping the victim's vulva
DNA	Investigation Bureau's DNA analysis conclusion was erroneous	Did not exclude DNA types mixed with the defendant
Serrated Knife	Fingerprint on the knife belonged to Yen Ko-Fu; no human blood reaction; victim's wounds were not caused by a sharp instrument; serrated knife was not the murder weapon	Knife had human blood and could cause the wound morphology of the victim; serrated knife was the murder weapon	ROC Armed Forces Forensic Center	Graduate Institute of Forensic Medicine, National Taiwan University College of Medicine
Military Trousers	Only trace bloodstains; DNA analysis identified them as the defendant's blood	Stained with human blood, inadvertently left behind during the crime	Criminal Investigation Bureau	Criminal Investigation Bureau
Crime Scene Reconstruction	Reconstructed blood source location and bloodstain splatter patterns did not match the defendant's confession	Defendant's confession matched the scene inspection and crime scene simulation evidence	Taipei City Police Department	Not re-analyzed

Conclusion: An External Review Mechanism for Scientific Forensics Should Be Established

Chiang Kuo-Ching was targeted as the perpetrator because he failed a lie detector test, and he was later acquitted because fingerprint and palm print analyses generated reasonable doubt in favor of the defendant. The miscarriage of justice in the Chiang case began with forensic science and ended with forensic science, demonstrating that water can both float and sink a boat.

Scientific conclusions must withstand repeatable verification; conclusions that cannot be verified are not science. Because domestic forensic science and criminal investigations are entirely controlled by closed official systems free from external review and supervision, and because judicial personnel are accustomed to accepting expert evidence at face value, the most fundamental scientific principle of repeatable verification has been reduced to a dead letter. The deceased are gone; the only way forward is to reform these systemic flaws and prevent future tragedies, thereby comforting the spirits of Chiang Kuo-ching and his father in heaven.

Postscript

The vast majority of forensic issues discovered in the Chiang Kuo-ching case today had long been pointed out by defense attorneys, yet military justice authorities turned a deaf ear. The author pays tribute to the original defense counsel, Mr. Hsueh Po-San (薛柏三, who passed away in 2002) and Ms. Liang Sui-Chang (梁穗昌); the retrial defense counsel, Mr. Yu Po-Hsiang (尤伯祥), Mr. Huang Ta-Yuan (黃達元), and Ms. Huang Hung-Hsia (黃虹霞); as well as all friends who participated in the work to exonerate this case under the convocation of the Judicial Reform Foundation.

---

Notes

1. Article 440 of the Code of Criminal Procedure: "Cases in which a retrial is petitioned for the benefit of a convicted person and an acquittal judgment is pronounced shall have the judgment published in the gazette or other newspapers."
2. Executive Yuan Gazette Information Network http://gazette.nat.gov.tw, Executive Yuan Gazette, Vol. 17, Issue 180, September 22, 2011.
3. Section "VII" of the reasoning in the acquittal judgment. Executive Yuan Gazette, Vol. 17, Issue 180, p. 30544, September 22, 2011.
4. Lin Yu-hsiung (林鈺雄), Code of Criminal Procedure (Vol. 1), p. 183, May 2007 edition.
5. Namely the 7 cases involving Lien Kuo-wen (連國文), Wang Hsin-fu (王信福), Huang Chu-wang (黃主旺), Cheng Chin-wen (鄭金文), Kuo Chun-wei (郭俊偉), Hsieh Chih-hung (謝志宏), and Ou-yang Jung (歐陽榕). Queries can be made via the Judicial Yuan's Jurisprudential Data Retrieval System.
6. Section "VI.3.(4).2.(4)" of the reasoning in the acquittal judgment. Executive Yuan Gazette, Vol. 17, Issue 180, p. 30541, September 22, 2011.
7. Section "VI.3.(1).3.(2)" of the reasoning in the acquittal judgment. Executive Yuan Gazette, Vol. 17, Issue 180, p. 30528, September 22, 2011.
8. Source: Hsu Jung-chou Indictment, p. 71, www.tpc.moj.gov.tw/public/Data/152493422243.pdf, last browsed October 11, 2011.
9. Cited from the testimony of Li Chun-i (李俊億), Director of the Institute of Forensic Medicine, Ministry of Justice, on February 11, 2011. Section "VI.3.(1).2.(2)" of the reasoning in the acquittal judgment. Executive Yuan Gazette, Vol. 17, Issue 180, p. 30526, September 22, 2011.
10. Section "VI.3.(2).3.(1)" of the reasoning in the acquittal judgment. Executive Yuan Gazette, Vol. 17, Issue 180, p. 30531, September 22, 2011.