Rice Paddies and Red Clothing: The Hwaseong Serial Murders That Haunted Korea

The First Body in the Field

On 19 September 1986, the decomposing body of a seventy-one-year-old woman named Lee Wan-im is found in a pasture near Taean-eup, Hwaseong County, Gyeonggi Province. She has been strangled. Her clothing has been used to bind her limbs. She had disappeared four days earlier while walking home after visiting her daughter. The police in Hwaseong, a quiet agricultural district of rice paddies and small towns roughly sixty kilometers south of Seoul, treat it as an isolated crime — a terrible event, but contained.

They are wrong.

One month later, on 20 October, twenty-five-year-old Park Hyun-sook vanishes after stepping off a bus while returning home from Songtan. Her body is found three days later in a canal. She has been raped, bound with her own clothing, and strangled. On 12 December, twenty-five-year-old Kwon Jung-bon disappears in front of her own house. Her body will not be found for four months, when it turns up on an embankment the following April.

Three women dead in three months. The pattern is unmistakable: all are female, all are attacked at night or early morning while traveling alone, all are bound with items of their own clothing — pantyhose, socks, bra straps — then raped and strangled. The rural landscape of Hwaseong, with its unlit farm roads and scattered villages, is the hunting ground. The killer operates in the spaces between streetlights.

---

The Killing Years

The murders accelerate through 1987 and 1988. Between January 1987 and September 1988, seven more women die. Their ages range from fourteen to seventy-one, demolishing any profile based on victim type. A twenty-four-year-old factory worker is found in a rice paddy. A teenager walking home from a friend's house never arrives. A middle-aged woman vanishes from a road she has walked a thousand times before. The fourteenth-year-old Park Sang-hee is attacked in September 1988 — the eighth victim — and her case will later become the most legally consequential of them all.

The method remains consistent but not identical. Most victims are strangled with their own garments — pantyhose knotted around the throat, a bra strap twisted into a ligature — but some are stabbed or bludgeoned. The killer uses what is available: a rock, a bladed tool, the victim's own clothing turned into instruments of death. In several cases, items of clothing are found stuffed into the victim's mouth or genitals. The sexual violence is systematic, occurring before or during the killing. Forensic analysis of crime scenes recovers semen samples, hair fibers, and traces of blood, but in 1986 South Korea, DNA profiling is still years away from practical application. The National Forensic Service can determine blood type — type B, a detail that will prove significant decades later — but little more. The biological evidence is catalogued and stored, its investigative value deferred to a future that no one can yet imagine.

One detail haunts the investigation with particular persistence: the apparent correlation between victims and the color red. Several of the women attacked are wearing red clothing at the time of their deaths. Whether this represents a genuine fixation of the killer — a trigger color that selects victims from the anonymous stream of pedestrians on rural roads — or a statistical artifact of a small sample size is never determined. But the rumor spreads through Hwaseong and beyond. Women stop wearing red. Stores report declining sales of red garments. An entire color becomes associated with death.

By 1988, the terror in Hwaseong has become a national obsession. Parents escort daughters everywhere. Women avoid walking alone after dark. The government deploys officers in unprecedented numbers. Night patrols are stationed on the narrow farm roads. Checkpoints are erected at key intersections. Plain-clothes policewomen are used as decoys, walking alone through the killing zones with backup teams hidden in the surrounding paddies and drainage ditches. On at least one occasion, a decoy officer is actually attacked — but the suspect escapes into the darkness of the rice paddies, vanishing into a landscape he clearly knows better than the officers pursuing him.

The ninth killing occurs in December 1988. Then a gap — over two years of silence — before the tenth and final murder: a woman found strangled on 3 April 1991. She is the last victim attributed to the Hwaseong series. Then the murders stop. The killer does not gradually wind down; he simply ceases. The silence is more disturbing than the violence. Either the murderer has died, been imprisoned for an unrelated crime, or moved elsewhere. The police do not know which. They will not know for twenty-eight more years.

---

The Scale of Failure

The Hwaseong murders generate the largest criminal investigation in South Korean history. The numbers are staggering: 2.05 million police man-days are deployed over the course of the investigation. Over 21,280 suspects and witnesses are interrogated. 40,116 individuals have their fingerprints taken. 570 DNA samples and 180 hair samples are analyzed. The investigation records fill five large plastic bags of documents. More police resources are devoted to this single case than to any other criminal matter in the nation's history.

None of it works.

The investigation is hobbled by multiple factors, each compounding the others. South Korea in the late 1980s is still under the shadow of authoritarian rule — President Chun Doo-hwan's military government only gives way to democratic transition in 1987, the same year the murders are intensifying. The police force is structured as an instrument of state control, not civilian protection. Officers are trained in crowd suppression, political surveillance, and the extraction of confessions — not in forensic evidence collection, crime scene preservation, or behavioral analysis of serial offenders. The concept of serial murder itself is largely foreign to Korean criminal investigation; the Hwaseong case is recognized as the country's first identifiable string of murders with a consistent modus operandi.

The forensic infrastructure is primitive by international standards. There is no national DNA database. There is no criminal profiling unit. Crime scene protocols are inconsistent — evidence is handled without gloves, scenes are trampled by responding officers, biological samples are collected but stored under conditions that will degrade their quality over time. The National Forensic Service possesses rudimentary capabilities for blood typing and hair comparison but nothing approaching the DNA analysis that Western law enforcement agencies are beginning to deploy.

Investigators pursue thousands of leads that go nowhere. They examine blood types, compare shoe prints, canvas neighborhoods, set up roadblocks, and interview every known sex offender in the province. They bring in military reservists to help search the rice paddies. They test a theory that the killer targets women on rainy nights — a pattern that appears in the data but proves impossible to operationalize. They generate suspect after suspect, clear them one by one, and return to the same empty starting point.

The institutional pressure to solve the case — from the media, from politicians, from the terrorized public — produces what institutional pressure always produces when competence is lacking: a scapegoat.

---

The Wrong Man

On 27 July 1989, police arrest a twenty-two-year-old man named Yoon Sung-yeo in connection with the eighth murder — the rape and killing of fourteen-year-old Park Sang-hee. Yoon is a local resident with a mild intellectual disability. He has no criminal record. He has no connection to any of the other murders. But he lives near the crime scene, and in the confession-driven culture of Korean policing, proximity is enough to begin the process.

Under interrogation, Yoon confesses.

The confession is a lie, extracted through torture. The district prosecutors' office will later confirm that Yoon endured what official language euphemistically describes as "cruel treatment" at the hands of police — a term covering sleep deprivation, physical assault, and electrocution. Officers deprived him of sleep for extended periods, held him in illegal detention beyond the permitted timeframe, and subjected him to physical violence until he produced the statement they needed. A forensic report from the National Forensic Service claims that pubic hair samples found at the crime scene show a 40 percent match with Yoon's. This report is later found to have been fraudulently written — the analyst manufactured a conclusion to support the confession already obtained.

Yoon is convicted and sentenced to life imprisonment. He will serve twenty years.

During those twenty years, he maintains his innocence from inside a prison cell. He writes letters. He appeals. He tells anyone who will listen that he did not kill Park Sang-hee. No one listens. During those twenty years, the actual killer — a man living within the same community, breathing the same air, walking the same roads — continues his life undetected. He marries. He works. He raises children. He attends neighborhood gatherings in the same district where he strangled ten women.

The wrongful conviction is not a footnote to the Hwaseong case. It is its center of gravity: the point at which the investigation stops seeking the truth and starts defending a fiction. Once the system has produced a conviction, the institutional incentive shifts from solving the crime to protecting the verdict. Every new piece of evidence, every lingering doubt, every continuing murder becomes a threat not to public safety but to institutional credibility.

The remaining nine murders are never charged to anyone. The investigation stalls, then effectively ceases. In 2006, the original fifteen-year statute of limitations expires on the earliest killings. The case enters the category of the permanently unsolved — or so everyone believes.

---

The DNA in the Drawer

For thirty years, the biological evidence collected from Hwaseong crime scenes sits in storage at the National Forensic Service. Semen samples, hair fibers, traces of skin and blood — preserved in the hope that future technology might extract what 1980s science could not.

The future arrives in 2019.

The intervening decades have transformed forensic science beyond recognition. DNA amplification techniques — polymerase chain reaction methods refined far beyond their 1980s capabilities — can now extract viable genetic profiles from degraded, decades-old biological samples that would have been considered useless a generation earlier. Short tandem repeat analysis can build an individual genetic fingerprint from quantities of material invisible to the naked eye. And crucially, these profiles can now be compared against databases that did not exist when the Hwaseong evidence was first collected. South Korea's national criminal DNA database, established in the years since the Hwaseong murders, now contains profiles of convicted criminals across the country. The National Forensic Service begins a systematic re-examination of cold case evidence.

On 18 September 2019, the Gyeonggi Nambu Provincial Police announce a result that sends a shockwave through South Korea: DNA extracted from the underwear of one of the Hwaseong victims matches a man already in the criminal database. His name is Lee Choon-jae. He is fifty-six years old. He is currently serving a life sentence at Busan Prison for the 1994 rape and murder of his sister-in-law.

Subsequent DNA testing links Lee to four additional Hwaseong crime scenes. The evidence is unambiguous. After thirty-three years, the largest unsolved murder case in Korean history has a name.

---

The Man Behind the DNA

Lee Choon-jae was born on 31 January 1963 in Hwaseong — the very community he terrorized. He grew up there, was remembered by neighbors as a quiet and polite child. At some point in his youth, his sister drowned, an event that by his own account affected him profoundly. He later claimed to have been sexually assaulted by an older sibling, though investigators could not verify this.

After graduating high school in February 1983, Lee entered the Republic of Korea Army, serving as a tank driver. His military records describe excellent technical skills; he was selected to drive the company commander's lead tank, a role that afforded him a rare sense of authority and purpose. He was discharged on 23 January 1986 — eight months before his first known murder.

The transition from military precision to civilian monotony appears to have been the inflection point. Lee described feeling bored with his post-discharge life. He worked at an electrical parts company in Hwaseong, then moved to construction work as an unlicensed crane operator. He married, had children, and maintained a facade of ordinariness that his neighbors never penetrated.

The killing timeline makes this dual existence explicit. Lee was murdering women in Hwaseong between 1986 and 1991 while living in the community, holding employment, and presenting a face of unremarkable normalcy. His neighbors did not suspect him. His family did not suspect him. The 2.05 million police man-days deployed to find him did not reach him.

In 1994, three years after the Hwaseong murders ceased, Lee raped and murdered his sister-in-law. For this crime — committed in Cheongju, Chungcheongbuk Province — he was arrested, convicted, and sentenced to life imprisonment with the possibility of parole after twenty years. He entered Busan Prison, where he remained for a quarter century before his DNA betrayed him.

---

The Nine Interrogation Rounds

Confronted with the DNA evidence in September 2019, Lee initially denies involvement. Over the course of nine interrogation sessions, he gradually provides details that match non-public investigative information: the bindings used on victims, the disposal methods, the specific locations. Details that only the killer could know.

On 2 October 2019, Lee fully confesses. He admits to perpetrating all ten Hwaseong serial murders between 1986 and 1991. He describes targeting women walking alone at night, raping them, and strangling or stabbing them with available materials — nylon stockings, hammers, whatever was at hand. He admits to concealing bodies in fields, streams, and embankments.

But Lee does not stop at ten. He confesses to fourteen murders in total — the ten Hwaseong victims plus four additional killings that had not been connected to the series. He also admits to more than thirty rapes and attempted rapes committed across Gyeonggi Province.

On 2 November 2020, Lee appears in court as a witness in the retrial of Yoon Sung-yeo — the man wrongfully convicted of the eighth murder. In the courtroom, he publicly confirms his guilt. He tells the court he is surprised he was not caught sooner. He apologizes to Yoon.

Yoon Sung-yeo is acquitted. After twenty years in prison for a murder he did not commit, the formal machinery of justice finally acknowledges what he has been saying since 1989: he is innocent.

---

Justice Expired

The DNA evidence is incontrovertible. The confession is detailed and verified. The case is solved in every meaningful sense except the legal one.

Lee Choon-jae cannot be prosecuted for any of the Hwaseong murders. The statute of limitations — originally fifteen years, extended to twenty-five in 2007 — had expired on all ten cases before the 2019 identification. South Korea abolished the statute of limitations for murder entirely in 2015, but the constitutional principle of non-retroactivity prevents this reform from applying to crimes committed before the law changed.

The irony is corrosive. The Hwaseong murders were one of the primary catalysts for the abolition of the statute of limitations. The public outrage over these unsolved killings directly fueled the legislative push to ensure that future murderers could never escape prosecution through the passage of time. But the very case that inspired the reform cannot benefit from it.

Lee remains in Busan Prison, serving his life sentence for the 1994 murder of his sister-in-law. He will be eligible for parole after twenty years — a date that has already passed. Whether he is ever released will depend on a parole board's assessment of a man who has confessed to fourteen murders and more than thirty sexual assaults but can only be legally held for one.

On 2 July 2020, police officially close the Hwaseong serial murder investigation, confirming Lee Choon-jae as the perpetrator of fourteen murders and nine rapes connected to the case. The investigation that consumed 2.05 million police man-days ends with a press conference thirty-four years after it began.

---

The Film and the Final Shot

In 2003, while the case was still unsolved, director Bong Joon-ho released "Memories of Murder," a film based on the Hwaseong killings. Bong spent six months researching the case, interviewing investigators and studying crime scene reports. The film depicts the investigation with brutal honesty: the incompetent rural detectives, the torture-based interrogations, the forensic failures, the military-era corruption that made genuine police work nearly impossible.

The film's final shot became one of the most celebrated sequences in Korean cinema. Detective Park, played by Song Kang-ho, returns years later to one of the murder sites. A child tells him that another man recently visited the same spot and stared into the same drainage ditch. Park asks what the man looked like. The child says: ordinary. Just an ordinary face. Park turns and looks directly into the camera — directly at the audience — as if searching the viewers for the killer who might be sitting among them.

When the real killer was identified in 2019, Bong — by then an international figure who would win the Palme d'Or and the Academy Award for Best Picture with "Parasite" — said he felt a complex mixture of relief and sorrow. The film had captured something true about the case: that the killer was ordinary, that he lived among the community he terrorized, that his face was one you could pass on any street without a second glance.

Lee Choon-jae reportedly watched "Memories of Murder" in prison. He told investigators he found it entertaining.

The rice paddies of Hwaseong have been consumed by Seoul's relentless suburban expansion. The narrow farm roads where women once walked alone at night are now multilane avenues lined with apartment towers. The landscape that concealed a killer for five years no longer exists. But the questions the case raised — about the reliability of confessions, the preservation of evidence, the limits of justice delayed — persist long after the fields have been paved over. The Hwaseong serial murders changed South Korea's criminal justice system, its forensic capabilities, and its understanding of what ordinary people are capable of. The cost of that education was ten women's lives, one man's stolen decades, and a truth that arrived thirty-three years too late for the law to act upon it.

The Anatomy of Institutional Failure

The Hwaseong serial murders are frequently narrated as a triumph of DNA technology — a story about science finally delivering justice after decades of darkness. This framing, while emotionally satisfying, obscures the more uncomfortable truth at the case's core: the thirty-three-year gap between the murders and the identification of Lee Choon-jae was not primarily a failure of technology. It was a failure of institutions.

**The Confession-Based Paradigm**

South Korean criminal investigation in the 1980s operated under what criminologists call a confession-based paradigm. The institutional incentive structure rewarded officers for obtaining confessions, not for building forensic cases. When physical evidence pointed nowhere, the system did not pause and wait for better tools — it found a suspect and extracted an admission. Yoon Sung-yeo's torture and wrongful conviction was not an aberration. It was the system functioning exactly as designed.

The 570 DNA samples collected during the investigation were catalogued but essentially shelved. They were not useless — they contained the genetic fingerprint of the killer — but the institutional framework had no mechanism for treating biological evidence as the primary investigative pathway. The confession of a suspect, however obtained, was the goal. Everything else was supplementary.

**The Proximity Problem**

Lee Choon-jae lived in Hwaseong throughout the killing period. He was a local. He was, by every account, unremarkable. This proximity created what behavioral analysts call the "neighbor blind spot" — the psychological resistance of a community to believing that extreme violence originates from within its own membership. Officers were looking outward, searching for an outsider, a drifter, someone visibly abnormal. Lee was none of these things.

The police decoy operations illustrate this blind spot perfectly. Officers dressed as women walked the farm roads, hoping to draw out the killer. On at least one occasion the killer attacked a decoy — and escaped. The operational assumption was that the killer was an outside predator who could be lured and trapped. The possibility that he might recognize the operation, identify the backup teams, and adjust his behavior was not adequately considered. A local resident would have understood the rhythms of police activity in his own community in ways an outsider could not.

**The Statute of Limitations Paradox**

The legal outcome of the Hwaseong case represents a genuine paradox in criminal justice reform. The case was among the most powerful arguments for abolishing the statute of limitations on murder — and yet the reform it inspired cannot be applied to the case itself. This is not a flaw in the law. The principle of non-retroactivity exists to prevent governments from retroactively criminalizing conduct or extending punishments after the fact. It is a protection against state overreach.

But the practical consequence is that Lee Choon-jae has confessed to fourteen murders and more than thirty sexual assaults for which he faces no legal accountability beyond his existing life sentence. The formal closure of the case in 2020 was an investigative resolution, not a judicial one. The families of the ten Hwaseong victims received the truth but not the legal validation that a trial and conviction would provide.

**The Wrongful Conviction as Central Narrative**

Yoon Sung-yeo's twenty-year imprisonment is often treated as a tragic sidebar to the main story of the serial killings. This framing inverts the actual significance. Yoon's case is not peripheral — it is the mechanism by which the investigation was terminated. Once Yoon was convicted of the eighth murder, the institutional pressure to solve the Hwaseong case was partially relieved. One conviction, however flawed, allowed the system to claim partial success and redirect resources.

The fraudulent forensic report — the fabricated 40 percent hair match — demonstrates that the corruption extended beyond the interrogation room into the laboratory. The National Forensic Service, the institution later credited with solving the case through DNA in 2019, was the same institution that manufactured evidence to convict an innocent man in 1989. The thirty-year arc from fraudulent forensics to legitimate forensics is not simply a story of technological progress. It is a story of institutional reform — slow, incomplete, and purchased at the cost of a man's life.

**The Red Clothing Question and Investigative Tunnel Vision**

The reported correlation between victims and red clothing illustrates a broader pattern of investigative tunnel vision. Whether the connection was genuine or coincidental, its public prominence consumed investigative bandwidth that might have been directed elsewhere. Officers devoted resources to monitoring women wearing red, establishing surveillance around clothing stores, and analyzing the color composition of victims' wardrobes. This focus on a single behavioral variable — one that the killer himself may never have articulated or even consciously recognized — exemplifies how pattern-seeking in serial murder investigations can generate confident hypotheses from insufficient data. The red clothing theory became a media fixation, a public safety warning, and ultimately a distraction from the systematic forensic work that would have been required to identify the killer through his biological evidence rather than his supposed preferences.

**What the Case Teaches**

The Hwaseong case has become a standard reference point in discussions of forensic reform, wrongful conviction, and the limitations of time-based justice. But its deepest lesson may be simpler than any of these institutional critiques. For thirty-three years, a killer lived among the people who were searching for him. He was not hidden. He was not disguised. He was simply ordinary — and ordinariness, in a system designed to detect the extraordinary, proved to be the most effective camouflage of all.