A Working Battery With Nothing To Power

What We Know

Wilhelm König was working through a shipment of finds in the storeroom of Baghdad’s National Museum in 1938, objects pulled two years earlier from Khujut Rabu, a low mound of ruins near the ancient city of Ctesiphon, when one clay jar made him stop. König, the Vienna-born archaeologist running the museum’s laboratory, realized this one didn’t belong with the rest. Most of what came out of that site was ordinary: pottery, coins, the debris of a vanished settlement. This jar held something else. Inside its neck sat a cylinder of rolled copper sheet, sealed at the bottom with a soldered copper plate. Suspended inside the cylinder, held by a plug of asphalt so it never touched the copper walls, was a rod of iron.

König recognized the shape. Strip away the clay, and what remained was the internal anatomy of a simple electric battery: two different metals, kept apart, ready to sit in a liquid. He published the observation that same year, in a two-page paper whose title translates as “A Galvanic Element from the Parthian Period?” The question mark was doing real work. König had a hunch, not proof.

Later tests bore the hunch out. Fill the copper vessel with an acidic liquid available anywhere in the ancient world (vinegar, wine, even grape juice) and the jar produces a small, genuine electric current. Every replication since has confirmed it, most recently a January 2026 study that found a second, previously unnoticed cell hiding in the design itself: the jar’s own unglazed clay, damp against a strip of solder on the copper’s exterior, forming a separate battery in series with the first. Together they measured over 1.4 volts, enough to visibly electroplate metal or split water into bubbles of gas.

Ten broadly similar jars turned up nearby at Seleucia and Ctesiphon in the years after. Same clay body, same sealed metal tube inside. Not one of them had an iron rod.

The Roadblock

Nobody argues about what the assembly does. The argument is entirely about intent, and intent is exactly what the physical evidence can’t supply. The original excavation notes from Khujut Rabu were thin to begin with, and the jar itself has been effectively unreachable for renewed study for decades, its precise present whereabouts within Iraq’s museum system inconsistently reported since a 2003 looting scattered the collection it belonged to. Nobody has run modern residue analysis, precision dating, or high-resolution imaging on the original object in living memory.

That leaves scholars arguing from comparisons instead of the thing itself. The jar’s pottery style points to a period centuries later than König’s own dating, but pottery style isn’t a signature. It places the object in a broad tradition, not on a specific day. The ten sibling jars from Seleucia and Ctesiphon prove this wasn’t a one-off design, then complicate the case further by omitting the one part, the iron rod, that makes the electrochemistry work at all. And the culture that built these things left no text, no inscription, no workshop record explaining any of it. Every theory here is reconstructing a missing instruction manual from the hardware alone.

Best Guesses

Electroplating

König’s own theory, from 1938: fill the assembly with acid and use the current to plate a thin layer of real gold or silver onto a cheaper base object, a plausible ancient trade, since gilded-looking goods sold as readily then as now. The trouble is what’s missing everywhere else. Paul Craddock, a metallurgist at the British Museum who has examined gilded objects from this region and period, says every one he’s seen uses one of two documented techniques, gold leaf or mercury-fire gilding, with no trace of electroplating’s signature. No electroplated artifact from Parthian or Sasanian Iraq has ever turned up, from any excavation, anywhere.

Scroll storage

St John Simpson, of the British Museum, offers the plainest reading: nothing here is wired to anything. The jar is a document case. What that reading must explain first is the one detail no sibling jar shares: an iron rod, insulated from the copper by asphalt, doing nothing a scroll case would need. Simpson’s answer starts with the dating: he argues the pottery reads centuries later than König’s, and that the original excavation records were too thin to support his date in the first place. Roll a scroll around the metal cylinder for rigidity, seal it, and bury it: a use with real company, since ten nearly identical jars turned up nearby holding the same sealed tube and nothing else.

Electrotherapy

Paul Keyser, writing in the Journal of Near Eastern Studies in 1993, ties the jar to a documented medical habit rather than a jewelry trick: first-century Mediterranean physicians prescribed live electric torpedo fish, pressed against the skin, for headaches and gout. The trouble is voltage. Keyser’s own reconstruction produces a real but faint current: with vinegar or lemon juice, the sensation barely registers, nothing like the popular image of “ancient electricity.” The 2026 reconstruction of a stronger, second cell only sharpens that problem, since a shock strong enough to etch metal is rougher medicine than a mild tingle. What still argues for the theory is the fish themselves: a portable cell delivers a comparable numbing jolt without keeping one alive to press against the skin.

Ritual corrosion

The newest entry belongs to Alexander Bazes, the independent researcher behind the 2026 reconstruction of the jar’s hidden second cell. His case inverts the whole electrical framing: the point was never to power anything outside the jar. Seal a written prayer or curse around the iron rod, let the assembly’s own acid and current corrode it, and the visible decay becomes proof that something unseen passed through the charm. That reading fits a documented regional habit: jars like this one turn up buried in house foundations alongside incantation bowls, ceramic bowls used for household protection. William Hafford, of the Penn Museum, argues independently that the iron rod reads as a ritual nail, not an electrode. Neither has explained why a symbolic gesture needed asphalt insulation and soldered joints precise enough to still register on a voltmeter two thousand years later.

The Verdict

The Khujut Rabu jar isn’t coming back for retesting anytime soon, so the evidence that would settle this has to come from somewhere else: another single jar, from a documented and undisturbed context, complete with an iron rod and something legible still inside it, the actual prayer, the actual scroll, or copper-salt staining where a plated object once sat. Ten near-identical jars are known. Not one offers that.

Absent a new find, the honest answer isn’t which theory is correct. It’s that the people who built this thing left behind a device that plainly works and not one sentence about why they made it. The electrochemistry stopped being in question in January 2026, at 1.4 volts and climbing. The reason it exists at all is exactly as open as it was in 1938.