Vampires, and the corpse that looked
fed.

The boundary

Three meanings of real, written down before any source was read.

The run refused to answer “were vampires real” as one question. It split it into documented events, misread natural causes, and genuine supernatural occurrence, and kept the metaphysics itself out of scope.

The primary period runs from the Petar Blagojević case in 1725 to Maria Theresa’s 1755 decree ending state-sanctioned exhumations. Medieval revenants and Slavic and Greek traditions sit behind it as background. The 1892 Mercy Brown case in Exeter, Rhode Island, is a cross-check from another continent and century, not a core geography.

Fifteen subquestions were written first: the earliest textual sources, what Johann Flückinger’s report actually records, whether rabies or porphyria symptoms appear in any case file, how long the exhumed had been buried, whether the Polish “deviant burials” were about revenants at all, and how Stoker’s Dracula relates to any of it.

One rule mattered more than the rest. Tabloid claims about specific exhumed bodies that no primary record or scholar corroborates were to be flagged as unverifiable, not used as findings.

The question was not whether vampires exist. It was which kind of real the record can support.

Competing explanations

Eight explanations, and one that every passage leaned against.

Each explanation was written down with the observation that would destroy it. Thirty-seven passages were then scored against all eight.

Decomposition (H1) collected nine supporting passages and none against. Frontier administration (H6) collected seven for and one against. Tuberculosis for the Brown family (H4) collected five for and none against. Rabies (H2) managed two weak supports. Porphyria (H3) and premature burial (H5) drew only opposing passages.

The supernatural reading (H8) is the outlier: twenty passages against, none for. That is not one refutation. It is cumulative coverage, because every class of sign in the ledger has a named natural or social account: blood at the mouth and bloating from decomposition, apparent non-decay from cold and lack of air, death clusters from contagion, and the paper trail from frontier officials.

Most passages were neutral to most explanations, which is how it should be. A passage dating the word “vampire” to an Old Russian text says nothing about rabies.

No single passage killed the supernatural. Every sign it relied on was claimed by something else.

Where the lore came from

The belief was old. What changed in the 1720s was the paperwork.

The earliest reference the run found is an Old Russian text of 1047. The panics that made “vampire” a European word came almost seven centuries later, when an empire started writing village cases down.

In 1718 the Habsburgs took northern Serbia from the Ottomans and settled the frontier with soldier-farmers. Military officials now investigated village deaths and filed reports. Petar Blagojević was exhumed at Kisiljevo in 1725, ten weeks after burial. Arnold Paole was accused after his death and dug up forty days in. In January 1732 the army surgeon Johann Flückinger filed his report on Medvegia, Visum et Repertum, which still sits in the Austrian State Archives.

Those reports were printed and passed around, and that is how the word reached German and English. The run supports the frontier as an explanation of documentation and diffusion, not of belief: the belief predates the frontier by centuries, and van Swieten was still being sent to stamp it out in the 1750s.

Dracula is a separate, later layer. Stoker drew on Emily Gerard’s 1885 “Transylvanian Superstitions” for garlic, the stake and the word Nosferatu. The run judged the link to Vlad III thin and found it has nothing to do with the panics of the 1700s.

The frontier explains why these villages were written down, not why their people believed.

Were they real?

Real events, real causes, no undead.

Two moments in the record come closest to a controlled test of the supernatural reading. Both came back negative.

In 1892 a trained physician, Dr Metcalf, examined the exhumed Mercy Brown and found the decay expected for her time in the ground. The blood in her heart was what a body in a cold grave keeps. Her brother Edwin drank the ashes of her burned heart and liver in water as a cure, and died two months later anyway. A cure built on the vampire theory was tried, and it failed exactly as tuberculosis would predict.

The Brown deaths were sequential over nearly a decade: Mary Eliza in December 1883, Mary Olive in June 1884, then Mercy and Edwin falling ill in 1891. That slow, household-by-household wasting is consumption’s signature, and it is what looked like someone draining the living.

In Serbia, gas pressure from decomposition pushes blood into a corpse’s mouth, and bloating and skin slippage are routine. Villagers who dug up a suspect saw ordinary decay and read it as a body that had been feeding. Even the Habsburg investigator van Swieten put “undecayed” corpses down to fermentation and lack of air in the grave.

The cure was tried. The patient died on tuberculosis’s schedule.

The weak link

The strongest explanation rests on an investigator sent to end the belief.

Decomposition wins on the counts. But the one document that could test it corpse by corpse came back as a title, and the historical witness for it had a job to do.

Decomposition’s own falsifier asks for a sign-by-sign check of Flückinger’s report against burial depth, coffin type, soil and season. The run could not do it. The Austrian archive record came back with the report’s title and date, not its text, and a Spanish transcription site was assessed as anecdotal and contributed nothing.

The run also flagged its 18th-century witness. Maria Theresa sent van Swieten with the stated aim of eradicating the belief as a “barbarism of ignorance”, so his naturalistic conclusion is not a neutral test. And it noticed something it could not check: the Medvegia inquest is dated 7 January 1732, so the exhumations were in midwinter, when cold slows decay and “undecayed” corpses are expected rather than strange. That would strengthen the decomposition case if the burial dates and weather were found.

Winning on the counts is not the same as passing your own test.

The popular theories

Rabies is a coincidence of timing. Porphyria has no patient.

Two medical explanations circulate widely as the “real” vampire disease. The run tested both against the case records and neither found a case.

Rabies has a timing link: an outbreak in 18th-century Europe coincided with the hysteria, and insomnia and aversion to light and water resemble vampire traits. But the Blagojević and Paole passages record burial intervals, accusations and deaths, with no hydrophobia, salivation, bite history or incubation period for any victim. The decisive test could not be run, and the strongest rabies argument actually leads to werewolves, through dog bites and wolf fear.

Porphyria is doubly disfavoured. No passage describes photosensitivity, excess hair, discoloured urine or reddish teeth in any accused person or family. The run’s caution is that this is absence from the narratives, not a genealogical record ruling it out.

Premature burial is excluded for the best-dated cases. Blagojević had been buried ten weeks and Paole forty days, and the Brown bodies showed ordinary decay. The run notes the exclusion does not yet cover the individual Medvegia corpses, whose burial dates it never saw.

A disease that fits the legend is not a disease that appears in the case files.

Trying to break it

Eight searches for the other side, and two that came back.

After the first pass, the run went looking for the evidence that would hurt each explanation most. Six searches returned nothing. Two returned a source.

The premature-burial search returned the Smithsonian’s account of burial intervals, which cut against live burial rather than for it. The deviant-burial search returned the PLOS One isotope study of Drawsko, which found the people buried with sickles and stones were locals, not outsiders. That rules out the outsider explanation but not an epidemic one, since cholera kills too fast to mark a skeleton.

The reading that survives at Drawsko folds the two together: locals distrusted on some ground, often the first to die in an outbreak, buried so they would stay down. The burials were never dated against the Serbian panics, so they cannot yet corroborate them.

Three passages were refused by the verbatim gate because their wording could not be found in the stored bytes. They are in the transcript and count for nothing.

The isotopes cleared the outsiders. They did not clear the epidemic.

What the number means

Fourteen out of a hundred, and why the number is that low.

Twenty findings asserted, one conjecture, six open questions. Every one computed to unknown confidence.

Corroboration and provenance scored zero because no passage was traced to an independent origin. The sources are mostly secondary: PBS, the Smithsonian, Wikipedia articles, a Frontline interview, and one peer-reviewed isotope study. Agreement among them may be one story retold, and the run does not pretend otherwise.

Falsification scored 15 of 20 for the eight counter-searches. Saturation scored zero: the second retrieval round found seven new source families where the first found three, which means the search had not run dry. Twelve stress probes were recorded against the findings, and counter-evidence was found, which caps the result in the unknown band.

A low score here is the run admitting that its sources may all be repeating each other.