Rauhensteingasse 8, 12:55 AM: The Death of Wolfgang Amadeus Mozart and a Parish Register That Does Not Diagnose

At about 12:55 AM on Monday 5 December 1791, in an apartment at Rauhensteingasse 8 in central Vienna, Wolfgang Amadeus Mozart died. He was thirty-five. His wife Constanze was at the bedside, her sister Sophie Haibel had come from across the city in the late afternoon, and Franz Xaver Süssmayr, the pupil who would later complete the Requiem, was present at points during the final hours. The attending physician Dr Thomas Franz Closset entered the cause of death in the St Stephen’s parish register as “hitziges Frieselfieber.” The phrase translates as severe miliary fever and described a high febrile illness with a fine skin rash like millet grains. In late-eighteenth-century Viennese registers it was a routine entry. It is a description of what the body looked like at the end. It is not a diagnosis.

That phrase, and what it does not say, is the case. For two hundred and thirty-four years the peer-reviewed medical-historical literature has tried to translate “hitziges Frieselfieber” into a modern diagnosis, and it has not converged. Rheumatic fever, post-streptococcal acute glomerulonephritis, trichinosis, kidney failure of broader etiology, chronic mercury exposure: each has been argued in print from the same surviving fragments. None has displaced the others. Closset’s case notes do not survive. The body was buried in a class-tiered common-grave plot at St Marx Cemetery; within a decade the plot was redug under the standard reuse cycle. A skull held by the Internationale Stiftung Mozarteum in Salzburg since 1902, with a provenance chain beginning with a gravedigger in 1801, was DNA-tested in 2004 to 2006 and returned an inconclusive result.

Sitting above that institutional record is a second case, the one the global cultural imagination knows. Pushkin in 1830, Rimsky-Korsakov in 1898, and above all Peter Shaffer’s Amadeus on stage in 1979 and Miloš Forman’s film of it in 1984 named the Imperial Kapellmeister Antonio Salieri as Mozart’s murderer. The film took eight Academy Awards. Salieri died in 1825 having, on the record, conducted Mozart’s works, attended Die Zauberflöte with apparent enthusiasm, and later taught Mozart’s surviving son. In 1823 he had told the visiting pianist Ignaz Moscheles on his own deathbed that the rumour was absurd. The standard scholarly accounts treat the poisoning theory as effectively dismissed.

This piece is the case of the case. The medical mystery is real and is sitting open in the journals. The Salieri mystery is folklore, and is treated here as folklore: named because the cultural accusation has named it, attributed to the works that carried it, weighed against the denial on record, and not adopted.

The documented account

Mozart was born on 27 January 1756 in Salzburg. The childhood tours are documented in the family correspondence in the Bauer-Deutsch edition at the Mozarteum. He settled in Vienna in 1781, married Constanze Weber at St Stephen’s on 4 August 1782, and composed prolifically through the 1780s. Of six children, two survived to adulthood: Karl Thomas, born 1784, and Franz Xaver Wolfgang, born July 1791, five months before his father’s death.

Mozart’s adult medical history is not robust. The correspondence preserved through Leopold’s death in 1787 records smallpox in childhood and recurrent rheumatic and febrile illnesses in 1762, 1765, and 1766, with stomach and rheumatic episodes through the Vienna years. He was not in robust health when 1791 began, and the year that followed was the most productive of his life.

The compositional landmarks of 1791 are documented. La clemenza di Tito, the coronation opera for Leopold II, premiered in Prague on 6 September. Die Zauberflöte premiered at Schikaneder’s Theater auf der Wieden in Vienna on 30 September. The Clarinet Concerto K. 622 was completed in early October. In July an anonymous commission for a Requiem mass had arrived through Anton Leitgeb, manager of Count Franz von Walsegg, who intended to pass the work off as his own in memory of his late wife. Mozart was working on the Requiem K. 626 through October and November. On 18 November he conducted the Masonic cantata Laut verkünde unsre Freude, K. 623, at his lodge Zur neugekrönten Hoffnung. It was the last work he directed.

The final illness is conventionally dated to on or about 20 November 1791, when he became bedridden. The symptoms recorded by later witnesses and in Nissen 1828 include severe swelling of the hands and feet (period accounts use “Geschwulst” and “Wassersucht,” the latter translating as dropsy), high fever, profuse sweating, vomiting, a rash, and pain on movement. The picture worsened over about two weeks. Closset attended throughout and called in Dr Mathias von Sallaba, a senior physician at the Vienna General Hospital, as consultant. Sallaba would himself be dead within six years, at thirty-three, of a hospital-acquired fever. Period treatment was applied, including at least one bloodletting in the final days. No case notes by either physician survive.

Mozart died at approximately 12:55 AM on Monday 5 December 1791 at the apartment in the Kleines Kayserhaus, Rauhensteingasse 8. The building was demolished in 1847; the site is now the Steffl department store.

The funeral was held on 6 December 1791 at the Crucifix Chapel (Kreuzkapelle) of St Stephen’s Cathedral. It was a third-class arrangement. A long popular tradition has read a pauper’s burial here; the documentary record does not support that reading. A third-class funeral in Vienna in 1791 was the standard middle-class arrangement under Joseph II’s reformed burial regulations. The 1784 edict had originally specified more radical provisions, including sack burial without an individual coffin; those were withdrawn after Vienna public protest before the edict took full effect. What remained was a class-tiered, time-limited common-grave system. Burial was at St Marx Cemetery on the city’s southern outskirts. The Vienna archival scholar Michael Lorenz has shown that the reusable-coffin and pauper’s-grave readings are documented myths. The exact location of the remains was lost when the plot was redug, around 1801. The present memorial at St Marx is the work of cemetery superintendent Alexander Kugler, erected c. 1859 and marking the approximate area.

Constanze remarried in 1809, to the Danish diplomat Georg Nikolaus von Nissen, who began a biography drawing on her papers and on testimony solicited from surviving witnesses. Nissen died in 1826 with the work unfinished; Constanze saw it through to publication as Biographie W. A. Mozarts, Breitkopf & Härtel, Leipzig, 1828. It is the foundational biographical document, and the source through which most of the famous deathbed details enter the record.

The evidence

The defensible evidence falls into a small number of classes, none of them a contemporaneous clinical document.

The parish register entry of 5 December 1791. The St Stephen’s register records the death with the cause “hitziges Frieselfieber,” signed by Closset. The original is at the Wien Stadt- und Landesarchiv, reproduced in Deutsch 1965/1995. As a diagnostic statement it is severely limited: “miliary fever” denoted a fever with a millet-grain-sized skin eruption and could attach to many distinct illnesses. The entry is the only contemporaneous official statement of cause that exists.

Constanze Mozart’s letters and recollections. Preserved in the Mozart-Briefe edition (Mozarteum, 1962-2005). Constanze also gave recollections during the 1829 visit of Vincent and Mary Novello to Salzburg; their diaries were published as A Mozart Pilgrimage in 1955. Her editorial role in Nissen 1828 is documented, and her recollections to the Novellos were given thirty-eight years after the events.

The “aqua tofana” remark belongs in this class and is often misattributed. Mozart in his final illness reportedly told Constanze that he was writing the Requiem for himself and that someone had given him aqua tofana, an arsenic-based poison of seventeenth-century Italian and Viennese folklore. The closest documentary trace is Constanze’s recollection to the Novellos in 1829, not Sophie Haibel’s 1825 letter, where it is sometimes wrongly placed. The attribution is at decades’ remove from the spoken words, and the clinical course is not consistent with arsenic poisoning.

Sophie Haibel’s letter of 7 April 1825 to Nissen. Written at Nissen’s request thirty-four years after the death. It is the source of the most vivid deathbed details and has to be read with the hedge that vintage requires. Sophie wrote that Mozart in the final days was so swollen he could barely move, that on the afternoon of 4 December some of the Requiem was rehearsed at his bedside with Mozart taking the alto line, and that during the Lacrimosa he attempted to imitate the timpani figure with his mouth before breaking off in tears. The letter is reproduced in Nissen 1828 and Deutsch 1965/1995. The scene cannot bear more weight than late-life recollection allows.

Closset’s contemporaneous observations. No case notes survive; what is known of his clinical reasoning is filtered through Nissen 1828 and subsequent secondary literature.

The Salieri cultural overlay sits alongside this material rather than within it. Antonio Salieri (1750-1825), Imperial Kapellmeister in Vienna, was the most prominent Italian composer there during Mozart’s career. The documentary record of the Mozart-Salieri relationship is one of normal professional contact. Salieri conducted Mozart’s works. He attended Die Zauberflöte with apparent enthusiasm, as Mozart reported in a letter to Constanze of 14 October 1791. After the death, the family sent Franz Xaver Wolfgang to study composition with Salieri, a decision difficult to reconcile with any family suspicion. A rumour that Mozart had been poisoned, with Salieri implicated, began to circulate in Vienna gossip in the early 1790s; the first explicit printed mention is in the Berlin Musikalisches Wochenblatt of late December 1791. The rumour cycle is traced in William Stafford’s The Mozart Myths (Stanford UP, 1991).

The literary afterlife is the part most readers know. Pushkin’s Mozart and Salieri (1830) crystallised the rumour as art; Rimsky-Korsakov’s 1898 opera set the Pushkin text; Shaffer’s Amadeus opened in London in 1979 and Forman’s film of it won eight Academy Awards in 1984. The closest documented contemporary report of Salieri’s response is from Ignaz Moscheles, who visited him in his final illness in 1823. Moscheles recorded that Salieri said, in substance, that he could assure him as a man of honour the report was absurd, that it was malice, sheer malice, and that Moscheles was to tell the world old Salieri, on his deathbed, had told him this. Reports of Salieri “confessing” during episodes of mental confusion late in life are at one remove.

The peer-reviewed medical-historical literature is where the open case actually lives. Carl Bär’s Mozart: Krankheit, Tod, Begräbnis (Mozarteum, 1966; 2nd ed. 1972), the standard German-language study, argued for rheumatic inflammatory fever with secondary cardiovascular failure, assessing the bloodletting as harmful but not the proximate cause. Jan V. Hirschmann’s “What Killed Mozart?” in Archives of Internal Medicine in June 2001 argued, as its primary thesis, for trichinosis from undercooked pork, citing a 7 October 1791 letter from Mozart that mentions pork cutlets and a trichinosis incubation of up to fifty days consistent with onset around 20 November. Rheumatic fever is among the alternatives Hirschmann considered; trichinosis is the thesis the paper defends. Richard H. C. Zegers, Andreas Weigl, and Andrew Steptoe, in Annals of Internal Medicine in August 2009, analysed the Vienna municipal mortality register for November 1791 through January 1792 against adjacent years, identified a relative excess of male deaths attributed to oedema and “Wassersucht,” and argued this was consistent with a streptococcal outbreak producing post-streptococcal acute glomerulonephritis. Peter J. Davies (JRSM, from 1983; Mozart in Person, 1989) and Lucien R. Karhausen’s The Bleeding of Mozart (2011) round out the principal contributions.

The Mozart-skull affair contributes no usable evidence to the medical question. A gravedigger named Joseph Rothmayer is reported to have retrieved a skull from the St Marx common-grave plot in 1801, claiming he had marked the body with wire. Its whereabouts are obscure until 1842; in 1868 it passed to the Viennese anatomist Joseph Hyrtl, and in 1902 it was donated to the Mozarteum by Hyrtl’s widow. A 2004 to 2006 DNA analysis by the Institute for Forensic Medicine in Innsbruck with the US Armed Forces DNA Identification Laboratory was inconclusive: the Salzburg family-grave reference samples were not consistent with each other. The skull’s authenticity is unresolved.

Hypotheses and open questions

Each of the six hypotheses below is labeled and source-attributed. None is asserted as fact. The parish register’s “hitziges Frieselfieber” is the only contemporaneous formal cause-of-death entry, and it is a symptomatic description.

1. Post-streptococcal acute glomerulonephritis with secondary cardiovascular complications (Zegers, Weigl, and Steptoe, 2009). The most-cited modern hypothesis. Based on the Vienna mortality data for 1791-92, read as an oedema-mortality cluster in young adult males consistent with a streptococcal outbreak possibly seeded from a nearby military hospital, and on a clinical course in which progressive swelling is consistent with renal failure. The inference from epidemic timing to individual case is necessarily probabilistic.

2. Rheumatic inflammatory fever with secondary cardiovascular failure (Bär 1966; Davies; others). Based on Mozart’s documented chronic illness history, including rheumatic episodes in childhood, and on the clinical course of sweating, swelling, joint involvement, and cardiovascular failure. The reading relies in part on Sophie Haibel’s late-life recollection and Nissen 1828’s editing of Constanze’s testimony, both of which carry hedges.

3. Trichinosis from undercooked pork (Hirschmann 2001). Hirschmann’s primary thesis. Based on Mozart’s letter to Constanze of 7 October 1791 referring to pork cutlets and a trichinosis incubation period of up to fifty days. The clinical course of progressive swelling and miliary rash is not a textbook trichinosis presentation, and the hypothesis was challenged in print shortly after publication.

4. Kidney failure of broader etiology. A meta-hypothesis encompassing the glomerulonephritis and rheumatic-nephritis variants above, accommodating the dropsy described by late witnesses without committing to a particular agent.

5. Chronic mercury exposure or other heavy-metal contribution. Discussed in Bär 1966 and Karhausen. Eighteenth-century practice routinely included mercury preparations; chronic exposure could plausibly explain residual symptoms without being the proximate cause.

6. Acute lethal poisoning, with Salieri or another party implicated. The cultural folklore. Its provenance is Vienna gossip from the early 1790s, the Berlin Musikalisches Wochenblatt of late December 1791, Pushkin 1830, Rimsky-Korsakov 1898, Shaffer 1979, and Forman 1984. It has no evidentiary basis in the documentary record. The clinical course does not match acute arsenic poisoning. Salieri’s documented relationship with Mozart was supportive; after the death the family sent Franz Xaver Wolfgang to study with him; the denial to Moscheles in 1823 is on the record. Modern scholarship (Stafford 1991; Robbins Landon 1988; Braunbehrens 1990; Solomon 1995) treats the theory as effectively dismissed. It is named here because the cultural accusation has named it; it is not adopted.

What remains unknown

Two hundred and thirty-four years on, the institutional file ends where Closset signed it. His case notes do not survive; Sallaba’s do not survive; the burial location was lost around 1801; the 2004 to 2006 DNA analysis was inconclusive.

The peer-reviewed literature has not converged. The Zegers, Weigl, and Steptoe 2009 streptococcal-glomerulonephritis reading sits at the centre of gravity of the field today without having displaced Bär’s rheumatic-fever reading, Hirschmann’s trichinosis, or the broader kidney-failure variants. Chronic mercury exposure remains on the table as a contributing factor. No single modern diagnosis has been established to the standard of scholarly consensus.

The cultural overlay resolves differently. Salieri has been named as Mozart’s murderer in global memory because Pushkin 1830, Rimsky-Korsakov 1898, Shaffer 1979, and Forman 1984 named him. The documentary record contains no evidence implicating him: no motive, no opportunity, no means, a supportive professional relationship, and Mozart’s own son later studying composition with him. The deathbed denial to Moscheles in 1823 is on the record. The reading is named here because it is the dominant popular one, and it is not adopted: it has no evidentiary basis.

What we can say, on the record that exists, is narrow. A composer with a chronic-illness history fell ill on or about 20 November 1791. Over two weeks his body swelled, he ran a high fever, sweated, vomited, and developed a miliary rash. He died at about 12:55 AM on 5 December 1791. The attending physician signed a parish-register entry that named the symptoms and not the cause. Two hundred and thirty-four years of inquiry have not closed the gap between that entry and a modern diagnosis. The case remains, in the literature, disputed and open.