The Silver Needle That Turned Black / 银钗验毒：宋慈的砒霜检测术

Forensic Chemistry Six Hundred Years Before Chemistry

From The Coroner's Notebook — a retelling of cases from Xi Yuan Ji Lu (洗冤集录), Volume IV: On Poisoning, Case 30

By Song Ci (宋慈, 1186–1249) · Retold by Cathay Tales

Hunan, 1247. A magistrate kneels beside a corpse with a thin silver hairpin in his hand. He rinses it in soapberry water, slides it past the dead man's teeth, and seals the mouth with paper. An hour later he pulls the pin back out. It comes up black — and won't wash clean. The verdict is murder. The chemistry that just convicted the killer will not be properly explained anywhere on earth for another six hundred years.

The Cold Open

Picture a Song dynasty courtyard. A merchant has been dead since the night before. His wife is composed — too composed, the neighbors say. She claims he fell ill after dinner. The county coroner has been called.

The coroner does not look at the wife. He kneels beside the body. He notices the fingernails: a dull, deep blue at the tips. The corners of the mouth are stained dark. The lips are split and curled. Under the tongue, when he lifts it with a wooden pick, he sees tiny dark blisters.

He has read his Song Ci.

He calls for soapberry water and a silver pin (银钗, yín chāi) — the official forensic tool of his office, kept in a wooden case at the county yamen^{[1]Yamen (衙门, yámén): The combined office, courthouse, and residence of a county magistrate in imperial China. Forensic tools like silver pins were kept under official seal at the yamen and were not personal property — losing one was a punishable offense.}. He washes the pin twice. He slides it carefully past the dead man's teeth, through the throat, and seals the mouth with mulberry paper.

He waits an hour, drinking tea. The wife stops being composed.

When the magistrate withdraws the pin, it is the color of a thundercloud — a deep, oily blue-black. He washes it in soapberry water again. The color does not come off.

In Song China, that is enough to begin an arrest.

The Technique

The full procedure occupies a few terse lines in Chapter 30 of Xi Yuan Ji Lu (洗冤集录), the world's first systematic forensic manual, completed by Song Ci in 1247. Translated literally:

"To test for poisoning: take a silver pin, washed in soapberry water (皂角水, zàojiǎo shuǐ), and insert it down the corpse's throat. Seal the mouth with paper. After a long while, withdraw it. If it shows blue-black and the color does not wash off, there was poison. If it remains bright white, there was none. If the deceased ate after taking the poison, so that the substance has passed into the bowels and the throat yields no result, then test from the anus instead."^{[2]Chapter 30, Fú Dú (服毒, "Ingesting Poison"): Located in Volume IV of Xi Yuan Ji Lu. Song Ci composed the work between 1241 and 1247 while serving as a judicial commissioner in southern China. The chapter on poisoning is one of the longer technical sections; it includes the silver-pin test, the rice-bag absorption test, body-symptom diagnostics, and a separate procedure for testing bones that had already been buried. Translation paraphrased from the Han Dian Classics edition; see source link in the original-text section below.}

This is not magic, and Song Ci does not call it that. He treats it as one tool among many. The same chapter walks the coroner through reading the body itself: blue-black gums, split lips, contracted tongue, blisters under the skin, swollen ears, distended belly, dark fluid leaking from the seven orifices. These are the symptoms of arsenic poisoning (砒霜中毒, pīshuāng zhòngdú) — though Song Ci has no word for arsenic in our sense. He has a word for the poison, and a list of what the poison does to the body, and a steel hairpin made of silver that will not lie.

He also notes the failures. If the corpse has been dead too long — already entering decomposition — the pin will turn black anyway, fooled by the body's own gases. If the deceased ate a heavy meal afterward, the test in the throat may come up clean while the poison sits two feet lower in the gut. Song Ci's solution to that: a second technique using a small bag of glutinous rice steamed inside the corpse's mouth, which absorbs whatever the silver missed. He files this in the same chapter. He is, in 1247, already two steps ahead of the obvious failure modes.

A Case from Song Ci's Files

Song Ci himself does not narrate individual cases the way a modern true-crime writer would. Xi Yuan Ji Lu is a manual, not a casebook — it teaches the technique and leaves the cases to the magistrate's own court records. But the silver-pin test survived for six centuries because magistrates kept using it on real corpses.

The most famous documented case comes from much later — 1808, in Huai'an (淮安), Jiangsu province. A junior official named Li Yuchang (李毓昌, a Qing-dynasty inspector sent to audit local flood relief funds) was found hanging in his lodgings. The local magistrate, Wang Shenhan (王伸汉, the man Li had come to investigate), ruled it suicide. The family did not believe it.

When Li's body was exhumed a year later under orders from the Jiaqing Emperor, the silver-pin test was performed on the long-decayed remains. The pin came back black — and stayed black. Combined with traces of arsenic in the bones (骨黪黑色, gǔ cǎn hēi sè — "the bones grow a dull, smoky black," in Song Ci's words), it confirmed poisoning. Wang Shenhan had bribed Li with poisoned tea, then strung the body up to fake a suicide. He and his accomplices were executed. The case became one of the most-cited vindications of forensic procedure in Qing legal history.

What convicted Wang Shenhan, six hundred years after Song Ci put pen to paper, was a paragraph in Chapter 30.

How It Holds Up Today

Modern toxicology has, of course, caught up. The black coating that formed on the magistrate's silver pin is silver sulfide (硫化银, Ag₂S) — a black, insoluble compound produced when metallic silver reacts with sulfur or sulfur compounds. The chemical equation is taught in any high-school chemistry class: 2Ag + S → Ag₂S.

The catch is that the test does not actually detect arsenic. Arsenic trioxide (As₂O₃), the active poison in pīshuāng, contains no sulfur and does not react with silver at all. What the silver pin was detecting, all those centuries, was the sulfide impurities left over from the crude smelting of arsenic ore — chiefly orpiment (As₂S₃) and realgar (As₄S₄) — which Song-dynasty workshops were unable to fully separate. The poisoner brought sulfur in with the arsenic, and the silver caught the sulfur.

This means Song Ci's test had two known failure modes that modern forensic chemistry has since classified:

1. False positives from sulfur-containing foods (egg yolk, garlic, mustard) and from putrefactive hydrogen sulfide gas released by any decaying corpse.
2. False negatives if a modern, highly refined arsenic compound is used — which is precisely why the silver-pin test was finally abandoned in the Republican era, after several wrongful acquittals in the 1930s where pure arsenic produced no reaction at all.

But for six hundred years, against the kind of impure arsenic available to a Chinese poisoner, the test worked often enough to be the legal standard. The modern equivalent — atomic absorption spectroscopy (AAS) on hair, nails, or stomach contents — replaced it not because the silver pin was wrong, but because the pin could only tell you yes or no, while AAS can tell you how much, in what tissue, over what time period.

For comparison, Western forensic science did not reliably test for arsenic poisoning until the Marsh Test, developed by the British chemist James Marsh in 1836 — almost six centuries after Song Ci's manual went into use. And the Marsh Test, ironically, was developed in response to a public scandal in which a known arsenic poisoner walked free because contemporary tests were too unreliable to convict her.

The Song dynasty coroner with his silver pin would not have been impressed.

Translator's Notebook

The first thing that struck me, reading Song Ci, is how unromantic he is. There is no theatrical preamble about the seriousness of his task, no invocation of Heaven, no warning about the wickedness of poisoners. He is a man writing a manual. He tells you what to do, what to watch for, what to do if the first thing doesn't work, and what colors to look for under different conditions. The whole chapter on poisoning runs to about three pages. Most of it reads like a checklist.

The second thing is that he is constantly aware of being wrong. Every technique has a paragraph attached describing the cases where it fails — old bodies, recently fed bodies, sick bodies, infant bodies. He gives the coroner backup procedures and tells him to record everything in detail because the case may be reviewed at the prefectural level. Song Ci is writing in a legal system that, at least in principle, audited its own results.

The third thing is harder to say without sounding sentimental. The silver-pin test was developed by people who did not know what arsenic was, did not know what silver sulfide was, and did not have the word "chemistry." They had a problem — wives being poisoned by husbands, husbands being poisoned by wives, merchants being poisoned by partners — and they solved it by paying attention. They noticed that silver got darker when it touched certain things. They noticed that the discoloration would not wash off. They noticed that the contamination came from inside the body. They wrote it down, taught it to the next generation of coroners, and used it for six hundred years.

That is not pre-scientific. That is just science before anyone had bothered to name it.

I think about that a lot, actually. The story we tell ourselves in the modern West is that real forensic science begins in Europe in the 1800s. The story Song Ci tells, without ever bothering to argue it, is that forensic science begins whenever you have a magistrate, a corpse, and someone willing to write down what worked.

Next case: The Red Umbrella Trick — how a Song dynasty coroner read invisible bone fractures using nothing but sunlight and a sheet of oiled red paper. → [Coming soon]