Chicago · the night of 8 October 1871
The Great Chicago Fire
How did the O'Leary barn on DeKoven Street catch fire on the evening of Sunday 8 October 1871, at what hour, and can anyone be named for it?
What happened. On the warm, dry evening of Sunday 8 October 1871, at the end of a summer with little rain, a fire started in the cow barn behind the two small cottages of Patrick and Catherine O'Leary, an Irish couple who kept five cows and sold the milk, at 137 DeKoven Street on the West Side of Chicago. By Patrick O'Leary's account his wife went to bed about eight and he half an hour after her; on the neighbourhood's time that put him in bed about ten minutes, and her about forty, when the barn caught, though the O'Learys cited no clock. Their tenants the McLaughlins, who rented the front cottage while the O'Learys lived in the rear one, had company that evening for a cousin newly arrived from Ireland, with two tunes on a fiddle and two dances. Daniel Sullivan, a neighbour from across the street who drove a delivery wagon for a living and had a wooden leg, had sat with the O'Learys until they turned in; some time after half past eight by the neighbourhood's clocks he saw fire in the barn, ran across, shouted fire, cut two cows loose and got out as the loft burned above him. Within half an hour five buildings were burning; the fire crossed the river that night and burned through the centre of the city for two days.
Why it is a mystery. Nobody saw the fire start, and within a day the newspapers had blamed Mrs O'Leary and a cow that kicked over her lamp, in two extras (special editions) of 9 October, one of them the Chicago Evening Journal's, an afternoon paper, a story that Michael Ahern, a reporter for the Chicago Republican in 1871, boasted forty years later he had made up; the Board of Police and Fire Commissioners, the city board that ran both the police and the fire department, held an inquiry in November and December 1871, questioned fifty-one witnesses and found no proof that anyone had been in the barn after dark. H. A. Musham, whose 1940 study is the longest older account of the fire's first hours, kept the cow story; Richard Bales, a lawyer with the Chicago Title insurance company, which holds the only land records to have survived the fire, argued in a 1997 article and a 2002 book that Sullivan could not have seen the barn from where he said he sat, that he lied to hide his own presence in it, and that he started the fire, alone or with Dennis Regan, another neighbour who had called that evening; and on the strength of Bales's research the Chicago City Council exonerated Mrs O'Leary and her cow by resolution in October 1997. Others have blamed a guest from the McLaughlins' party going to the barn for milk, boys gambling in the loft by lantern light (a claim first made public in 1944 on behalf of Louis M. Cohn, a Chicago businessman who died in 1942 and said he had been one of them), a spark from a chimney, the new hay heating by itself, and fragments of Biela's comet, a comet that had broken up and been lost to sight some twenty years earlier; A. T. Andreas, who interviewed the neighbours and the firemen for his History of Chicago in 1884, and Carl Smith, the Northwestern University historian whose 2020 book is the most recent full history of the fire, both concluded that the cause cannot be known. Three things keep the question open: the witnesses' words survive but disagree with one another by up to an hour; the shorthand record of the inquiry survives only as 1,100 handwritten pages at the Chicago History Museum; and Bales's argument about Sullivan's view, the blocked view, the only evidence that names anyone, has been repeated in print for nearly thirty years without a published re-examination.
How it was judged. The evidence read was the inquiry as the Chicago Tribune, one of the city's morning papers, reported it day by day, the affidavits (the sworn statements of the O'Learys and Sullivan, taken a week after the fire on 15 October 1871), the interviews Andreas printed, Bales's own transcription of the witnesses (read in fragments, since his book could not be read in full), the city directories, an 1891 property atlas, a 1906 fire-insurance map, the present-day lot lines and one photograph of the O'Leary cottage. From the last five the block was rebuilt to the foot, and a 3-D model was run for 7,200 draws (runs with every uncertain dimension set at random within its range) to ask, for each of the four places the record puts Sullivan, whether the barn or a flame above it could be seen; the 46 time statements (what witnesses said about the hour) were tabulated and sorted by the clock each rested on; and every theory was set against the record and given a share, as a judgment, with the shares recomputed under the readings that change them most. Two findings tipped the balance: the yard and barn were in complete darkness, so whoever was inside carried a light or struck one, and Bales's blocked-view argument succeeds for the one seat he chose for Sullivan, the midpoint of the west parcel of William White, a neighbour who owned two parcels of land on Sullivan's side of the street and at whose fence Sullivan said he sat, and fails for the three other places the record puts him, the seats differing by one 25-foot parcel and the outcome changing with about three feet in building height, which is within the uncertainty of every dimension in the reconstruction. The verdict is a flame carried into the barn by a person, about 85 percent, with the fire beginning at about 8.40 by the neighbourhood's clocks, and the person cannot be identified, because in the central assessment (the ranking with each piece of evidence at the weight the report gives it) no candidate reaches 30 percent; the inquiry transcript read in the original, or a record fixing the width and height of the house of James Dalton, the neighbour next door east, which decides what Sullivan could see, would change it.
Verdict
A flame carried into the barn by a person, about 85 percent; the person cannot be identified from the surviving evidence. Among the possible carriers, Daniel Sullivan and a member of the O'Leary household stand first, close to even. The fire began inside the barn, on the side where wood and shavings were kept, at about 8.40 by the neighbourhood's clocks.
The 85 percent for a carried flame is a judgment with its sensitivity (how far it moves under other readings of the evidence) printed beside it, and the candidates below it are ordered by judgment with ranges: the two leading candidates are close to even and none reaches 30 percent in the central assessment. The inquiry transcript was not read in the original and Bales's book only in fragments, so the ranking rests on the Tribune's reports of the inquiry, the affidavits and Andreas's interviews.
The fire began inside the O'Leary barn on the side where wood and shavings were kept and was in the hay loft when the first neighbour reached it (about 95 percent). It began about 8.40 p.m. by the neighbourhood's clocks (range 8.30 to 8.50); the 9.20 to 9.30 of the Board of Police and the press is 35 to 45 minutes late on every clock, the fire department's included. The cause was most probably a flame someone carried into the barn, a lamp, a lantern, a match or a pipe (about 85 percent); a chimney spark blown in at the open doors is about 8 percent, another accident 5, arson 3, spontaneous heating of the new hay 2, and Biela's comet is excluded.
The percentages are judgments made from the evidence. They depend on the list of scenarios in section 7 of the report being complete and on the weight given to each piece of evidence in sections 3 to 7. In the central assessment no candidate reaches 30 percent, and the two leading candidates, Daniel Sullivan and a member of the O'Leary household, are close to even. Each of three alternative readings in section 8 (accept Bales's blocked view as proof of a lie; accept the Dalton scream, Dalton's memory of a woman's scream from the direction of the O'Leary premises about ten minutes before the fire's light, and the neighbours' affirmations, the neighbours who told a Cincinnati newspaper's correspondent within days that the cow story was true; accept the Cohn story) lifts one candidate to between 38 and 45 percent, and none of the three rests on evidence that can be checked, which is why the central figures stand.
The one argument that names anyone is Bales's case against Sullivan: that he could not have seen the barn from where he said he sat, so his story covers his own presence inside. The sightline argument holds for the one seat Bales chose (the midpoint of White's west parcel, where the model shows nothing under about 16 ft above the barn's ridge, the top of its roof: median 15.6 ft with the eye at the height of a man seated on the raised sidewalk, 14 to 16 ft across the three eye heights tried) and fails for the three other placements the record gives (his affidavit's "a few lots east", the Board's "nearly opposite", Andreas's "in front of his own house", which by Bales's own deeds is the next parcel west, where a flame within 5 ft of the ridge is visible in 61 percent of the model's draws and the barn itself in 17 percent). The difference is one 25-ft parcel and about three feet of house height, inside the error of every dimension the block can be given. His time argument rests on an error about the hour that the official record as a whole makes, so it tells nothing about Sullivan in particular. Sullivan remains a candidate, level with the O'Leary household at the head of the ranking, and the evidence does not place him above that.
The shares are judgments from the evidence in sections 3 to 7 of the report, ranked in section 8; the carried-flame rows together give the headline figure of about 85 percent. Three readings move them most, each applied alone with the others held at their central values: accepting Bales's blocked view as proof that Sullivan lied about his seat lifts him to 45; accepting the Dalton scream as reliable and the 1871 neighbours' affirmations as independent of the newspapers lifts the O'Leary household to 38; accepting Cohn's bequest story (the claim published with the bequest he left) or the Evening Journal's report of a "strange man" as evidence lifts the alley scenario to 42. None of the three rests on evidence that can be checked, which is why the central figures stand.
What was done
The 1871 inquiry of the Board of Police and Fire Commissioners was read as the Chicago Tribune reported it day by day, with the sworn statements taken on 15 October 1871 from the O'Learys and from Daniel Sullivan, the neighbour who saw the fire first; the interviews A. T. Andreas took in 1884 and 1885 for his History of Chicago; the city directories; an 1891 property atlas; a 1906 fire-insurance map; the present-day lot lines; and one photograph of the O'Leary cottage. The inquiry transcript itself was not read, and the 2002 book of Richard Bales, the lawyer who accused Sullivan, was read only in fragments.
From the maps, the lot lines and the photograph the block was rebuilt to the foot, and a three-dimensional model of it was run for 7,200 draws, runs with every uncertain dimension set at random within its range, to ask, for each of the four places the record puts Sullivan, whether the barn or a flame above it could be seen. The 46 statements witnesses made about the hour were tabulated and sorted by the clock each rested on. Every theory of the fire's origin was then set against the record and given a share, as a judgment, and the shares were recomputed under the three readings that change them most.
Figure 1 shows the north side of DeKoven Street, the barn, and the places Sullivan may have sat according to the different statements of his position, with lines of sight from three of them. Figure 2 follows the line of sight from each of those seats: from the seat Bales chose, the midpoint of the west parcel of William White, a neighbour on Sullivan's side, the houses between rise above the line to the barn's ridge, and from the half-lot two parcels further west it reaches the barn's wall three feet above the ground. Table 1 shows how the ranking moves under three alternative readings, one reading changed at a time.
Two findings tipped the balance. The yard and barn were in complete darkness, so whoever was inside carried a light or struck one, which is why a flame carried into the barn by a person heads the ranking, at about 85 percent, with the fire beginning at about 8.40 by the neighbourhood's clocks. Bales's blocked-view argument succeeds for the one seat he chose for Sullivan and fails for the three other places the record puts him; the seats differ by one 25-foot parcel and the outcome changes with about three feet in building height, which is within the uncertainty of every dimension in the reconstruction.
So the person cannot be identified: in the central assessment, the ranking with each piece of evidence at the weight the report gives it, no candidate reaches 30 percent. Sullivan and a member of the O'Leary household stand at about 22 percent each, someone from the alley at 20, a party guest at 15. The percentages are judgments that depend on the list of scenarios being complete and on the weight given to each piece of evidence; each of the three alternative readings lifts one candidate to between 38 and 45 percent, and none of the three rests on evidence that can be checked, so the central figures stand.
| Reading | Sullivan | O'Leary household | Party guest | Alley | Chimney | Other |
|---|---|---|---|---|---|---|
| Central | 25 | 22 | 15 | 20 | 8 | 10 |
| A. Accept Bales's blocked view as proof that Sullivan lied about his seat, and read the lie as covering his presence in the barn | 45 | 15 | 10 | 15 | 6 | 9 |
| B. Accept the Dalton scream as reliable and the 1871 neighbours' affirmations as independent of the newspapers | 17 | 38 | 15 | 15 | 6 | 9 |
| C. Accept Cohn's bequest story or the Journal's "strange man" as evidence rather than rumour | 17 | 15 | 10 | 42 | 6 | 10 |
Sullivan's column includes Regan as accomplice; "Other" is arson, spontaneous heating and unclassified accident. Reading A corresponds to Bales's position, reading B to Musham's, and reading C to the Cohn family's. None of the three is supported by evidence that can be verified, and the central row is therefore retained.
What is new here
- Five documents not previously used in the published discussion, as far as could be checked against a book read only in fragments: the entries in Edwards' city directory for 1871 (the annual list of residents with their trades and addresses) showing Sullivan and White as teamsters (drivers) for the same firm, Chase, Hanford & Co., an oil merchant on South Water Street; the Tribune's reports of the inquiry sittings of 29 November and 4 December 1871, on pages missing from the Library of Congress's online newspaper collection (30 November p. 4 and 5 December p. 2) and recovered from another scan; the silence of the Illinois Staats-Zeitung, the city's German-language daily, on the inquiry and the Board's report; the dating of the Evening Journal's O'Leary interview to 21 October 1871; and a second Chicago extra of 9 October 1871 carrying the cow story.
- The sightline test: a 3-D model built from the 1891 atlas, the 1906 Sanborn map, the present-day parcel, one photograph and the deeds Bales found in the tract books, run with uncertainty ranges over all four placements of Sullivan already in the record (his affidavit's "a few lots east", the Board's "nearly opposite", Andreas's "in front of his own house", and Bales's midpoint of White's west parcel), 7,200 draws. From the seat Bales chose nothing under about 16 ft above the barn's ridge is visible (median 15.6 ft; 14 to 16 ft across the three eye heights tried); from the parcel his own deeds put Sullivan's house on, a flame within 5 ft of the ridge is visible in 61 percent of the model's draws and the barn itself in 17 percent. The model adds no document to the record.
- The three-cluster reading of the 46 time statements, grouped by the clock each rests on, and the 10 to 15 minute offset between the neighbourhood's clocks and the fire department's, a re-reading of the Tribune's and Andreas's reports. By Patrick O'Leary's account he had been in bed about ten minutes (none to twenty) and his wife about forty (thirty to fifty) when the barn caught, on times for which the O'Learys cited no clock. Sullivan's sequence of events is consistent once the clock times he gave are discarded: an hour at the O'Learys' from about a quarter past seven, twenty to twenty-five minutes on the sidewalk, the fire seen at about 8.40.
- The inference from computed sunrise that the 5 a.m. milking was in darkness, so a lamp was kept in the barn every day and its remains prove nothing; whether Bales makes the point could not be checked, since his chapter 2 was read only in snippets.
- The re-working of the arithmetic in Bales's Appendix B, where he adds up the distance Sullivan ran to the barn: his 193 feet depends on the seat he chose, and a man who had driven a dray (a delivery wagon) for fifteen years on a wooden leg covers it in thirty to fifty seconds.
- The finding of Anthony DeBartolo, a Chicago Tribune writer (1997, 1998), that Cohn's recorded will contains no confession is his and is not claimed as new; it is repeated because the contrary is often printed.
Already established by others. The Board of Police found in December 1871 that the cause could not be determined from the evidence; Andreas restated that in 1885 and Carl Smith in 2011 and 2020. Bales (1997, 2002) showed that the case against Mrs O'Leary was never more than hearsay, and Fred Fedler (1985), a journalism scholar, that the cow story was in print before any reporter could have invented it as later claimed. The Chicago City Council adopted the same view by resolution on 28 October 1997, a legislative judgment on Bales's research that adds nothing to the evidence about the night. DeBartolo found in 1997 and 1998 that Cohn's recorded will contains no confession.
Where this verdict stands among historians
What historians have said. The Board of Police (December 1871), Andreas (1885) and Carl Smith (2011, 2020) found the cause undetermined, and Musham (1940) affirmed the cow story. Bales argued in 1997 and 2002 from the Chicago Title land records and the houses on the block that Sullivan could not have seen the barn from where he said he sat, that his story covers his own presence in the barn, and that he, perhaps with Regan, started the fire; his case has been repeated in print for nearly thirty years. The Chicago City Council, by resolution of 28 October 1997 introduced by Alderman Edward M. Burke and approved by its Committee on Police and Fire on 6 October 1997 after a hearing at the fire academy that stands on the site of the barn, resolved to "forever exonerate Mrs. O'Leary and her cow from all blame in regard to the Great Chicago Fire of 1871": a legislative judgment on Bales's research, which adds nothing to the evidence about the night.
It agrees with the official finding that the cause cannot be determined and adds no name. It differs from Bales (2002) on Sullivan: his sightline argument holds for the one seat he chose, the midpoint of White's west parcel, and fails for the three other placements the record gives, and his time argument rests on an error about the hour that the official record as a whole makes. Nothing here overturns what historians already think. Bales's refutation of the cow story stands and the analysis here depends on it, and any model of the block depends on his tract-book research.
What would overturn it
- The full inquiry transcript, read in the original, holding a statement by Sullivan or Regan about what they saw from the street that fits no placement at all, or an admission by anyone that they were in the barn.
- A record of Dalton's house (a builder's contract, an insurance survey, a photograph) fixing its width and height: with the house two feet lower or five feet narrower, the ridge is visible from White's parcel in most draws; with it two feet higher, nothing under ten feet of flame is.
- The 1870 or 1872 directory, or a deed, settling whether No. 134, Sullivan's house, was the east half of Lot 18 or the west half of Lot 19.
- A text from October to December 1871 in which Mrs O'Leary, or anyone in the house, admits being in the barn; none was found in about 120 issues of 1871 papers, but the Chicago Times of 18 October and the Evening Journal extra of 9 October were read only through later quotation.
- Documentation of Cohn's story from before 1942, which would do the same for the alley scenario.
Limits
- The inquiry transcript was not read in the original: the Chicago History Museum's transcription could not be obtained, and Bales's Appendix D, his transcription of the witnesses, could be recovered only in fragments of about 100 characters.
- Bales's book could not be read in full: it was read through two search services that return fragments of a book rather than its pages, 128 Google Books queries returning 534 snippets, each tagged with its page number, from 187 pages, and 435 Internet Archive search-inside queries returning 1,439 fragments of about 100 characters joined into 76 passages; his Figures 54, 64 and 78 and his 1997 article were not seen.
- Carl Smith's 2020 book was read through its excerpt in American Heritage magazine; the later literature on the fire, Sawislak (1995), Pauly (1984), Miller (1990), Cromie (1958) and Kogan and Cromie (1971), was not obtained; Ahern's and Cohn's originals are known through later quotation. The 1997 resolution is cited from its own text as reproduced by the Glessner House Museum, a Chicago house museum, and from the press of October 1997; the Journal of the Proceedings, the City Council's own record, for 28 October 1997 was not reached.
- The barn's orientation is unsettled between Andreas's plan and Musham's, and the heights of Dalton's house, of the O'Learys' two cottages and of the cottage at No. 133 rest on one photograph and typical construction; the model carries these as ranges. It leaves out trees, smoke, the coal shed in the passage beside the cottage and the light of the fire itself. Its result is a range of heights, and it cannot settle a claim that falls inside those ranges.
- The 1870 and 1872 directories could not be obtained, nor could several archive holdings; no paywall or bot wall (a block on automated visitors) was bypassed.