By Richard Howe*
On December 31, 1761, “the Frequent Instances of the Extensive Destruction made by Fire in many populous City’s [sic]” prompted the legislature of the Province of New York to pass an act “for the more Effectual Prevention of Fires and for Regulating of Buildings in the City of New York.” The act’s principal concern was the number of the houses in the city that were still roofed with wooden shingles; it decreed that four years hence, i.e., starting January 1, 1766, “every Dwelling House or Building Whatsoever whether public or private that shall be Erected… within [the city of New York] to the Southward of Fresh water [pond] shall be made of Stone or Brick and Roofed with tile or slate….” Existing buildings were grandfathered, though the act provided that any future reroofing of them had to be done with tile or slate. Builders and property owners protested that there was not enough slate or tile available to roof all the new buildings going up each year in the city — by my estimate already perhaps 100 or so — and on December 13, 1765, the legislature postponed the effective date of the act to January 1, 1768.
The protests continued on past the new deadline, and on December 31, 1768, the act was suspended until January 1, 1774. True to form, on May 1, 1776, four months after the act had become effective again, some 1,600 affected residents of the city petitioned Lieutenant Governor Cadwallader Colden, protesting the act and urging its suspension, this time citing not only the scarcity of materials and their cost but also the hardship to workers in the building trades that would result from the chilling effect of the act on the city’s construction industry. The petition was referred to committee, where it languished and died. Then on April 1, 1775 — less than three weeks before the Battles of Lexington and Concord signalled the start of the War of Independence — the slate and tiles act was radically amended to make it into just the opposite of what it had been originally: it would now be lawful “to erect any Building in [New York] with Wood or other Materials and cover the same with Shingles or Boards” south of a line running irregularly but more or less along the line of today’s Fulton Street from the Hudson River to the East River, so long as the building was no more than 14 feet high. The amended act also made it lawful south of the line “to cover the Flat of any roof with Boards or Shingles, provided such Flat do not exceed two equal fifth Parts of the Span of such Roof and there be erected around the same Flat a substantial Balcony or Balustrade and a Platform and Steps to the top of every Chimney.” Eighteen months later a large part of the city was burnt to the ground in a single night.
Whether roofing the buildings in New York with slate or tile starting in 1762 as the act of 1761 required would have been enough to prevent the fire of 1776 from reaching such disastrous proportions is a question that could scarcely be answered, then or now, though it is hard not to believe that it would have limited the damage to some extent. Whether the fire was the result of arson or was just an accident that had been waiting to happen, is another question that will no doubt also remain forever in doubt, but General Washington had orders from the Continental Congress not to burn the city, and neither the British occupiers nor the resident Loyalists had any reason to burn it, so an argument can be made that it was an accidental fire, possibly even a very minor one to begin with, minor enough that under ordinary circumstances a team of fire fighters working one or two of the city’s hand-pumped fire engines might have brought it under control before even one building was totally destroyed.
But by the night of the fire, New York was very nearly a ghost town: as much as 80% of the city’s population had fled, leaving only perhaps 5,000 in a city that had built accommodations for 25,000. The fire fighters who might have been able to contain the fire were probably not available, or not nearly in the numbers previously on hand, and it is likely that even much of the equipment remained unused. Whether or not there was anyone left in the city who knew how to operate a fire engine is another unknown, but it is a virtually certainty that many of the houses and other buildings in the path of the fire were vacant, as were many of the others in the city, so that there was no one present to fetch the “Leather Bucketts” that every building was supposed to have, which in turn would mean not only that there were not enough men available but not even enough buckets to form a proper bucket brigade, and many such bucket brigades were surely have been needed to bring the fire under control, if indeed it could have been controlled at all, once it really got going. Exceptionally high winds and an abundant supply of fuel — the hazard of building with wood — were enough to do the rest, and of course the boards and shingles didn’t help.
New York’s great fire of September 20–21 — just five days after the British captured the city — burnt out a large part of the city overnight, though it is difficult to say just how large a part: eyewitness estimates ranged from as little as the 493 houses estimated — probably many years later — by the proprietor of Hessian’s Coffee House, David Grim, to the 1,000 spoken of the next day by the chaplain of the British fleet, Thomas O’Beirne in a sermon at St. Paul’s — which had survived the fire — to a possible high of as many as 1,500 claimed by a Loyalist in a letter from New York dated September 23, 1776. Other, less specific, estimates ranged from prominent New York journalist and publisher Hugh Gaine’s one sixth of the city to the one-third also mentioned in the Loyalist’s letter. A second “great” fire — though not nearly as great as the first one — destroyed another 54 houses and “several” — meaning at least three? — warehouses on the night of August 3, 1778.
These estimates indicate a probable combined loss of at least 550 houses and other buildings in the two fires, and possibly as many as two to three times that number. But people living in 1776 were surely as given to exaggeration in the immediate aftermath of a disaster as they are today, which makes Grim’s figure of 493 houses lost to the 1776 fire and Hugh Gaine’s report of only one sixth of the city destroyed somewhat more credible than the many other — and much higher — estimates.
It’s possible that Grim arrived at his figure using the same method he used in 1813 to estimate the number of houses standing in New York in 1744, which, according to his own memorandum, was to sum the lengths in feet of the street fronts in the built up part of the city, divide this sum by an assumed average lot width of 25 feet, and subtract from the result the number of lots he remembered as having been vacant at the time. Grim made his 1744 estimate using one of Ratzer’s maps of the city; using the same approach, I come up with a round-number estimate of about 600 buildings in the area burnt out by the 1776 fire as delineated by Grim on his own map. This is about 100 more than Grim’s estimate — the difference might be due to the fact that unlike Grim, I find myself unable to remember which lots were vacant at the time…. Be that as it may, this result lends support to Grim’s estimate: 600 is only 22% more than 493, not 100% or 200% more. Moreover, the exercise shows fairly conclusively that it would have been next to impossible to fit 1,000 houses into the burnt out area, to say nothing of 1,500.
So there is reason to accept Grim’s figure of 493 houses lost in the 1776 fire, and the loss count for the 1778 fire is straightforward: the detailed list exists, and there is no compelling reason to doubt it. The remaining question is what fraction of the city’s houses and other buildings were destroyed in the two fires, which is what matters for understanding the magnitude of the losses for the city as a whole. Taken together cross-wise, the low and high extremes of the numbers and fractions supplied by the witnesses to the fire would imply that the number of houses standing in the city prior to the 1776 fire was somewhere between 1,650 (550 × 3) and 9,000 (1,500 × 6), although it is not certain whether these fractions were meant to refer to the number of houses burnt or to the burnt part of the city. In any case neither of these extremes makes sense as an estimate of the number of houses standing in a New York with a population thought to have been around 25,000: the low figure would imply just one building for every fifteen people — a ratio not reached in the city for another hundred years — and the high figure would imply that there were not even three people for every building in New York; moreover, such a number would imply that the areal extent of the built up part of the city was on the order of twice what it is known to have been from the excellent 1782 “British Headquarters” map.
Fortunately, there are two contemporary and probably independent sources that can be used to estimate more directly the number of houses standing in the city prior to the 1776 fire. In March of 1788, Noah Webster of later dictionary fame reported in his “General Description of New York in 1786” — published in his own New York-based American Magazine of Useful and Entertaining Knowledge — that according to the enumerations made by him personally in 1785–1786 the ratio of people to houses in twenty-two of the leading cities and towns of the United States was generally not more than seven to one. Webster was both obsessive enough and well-travelled enough in the new Republic to know what he was talking about. The ratio of the 1786 New York State census’s reported population of 23,614 to Webster’s count of 3,340 houses standing in the city in that year is almost exactly seven to one, and would imply some 3,570 houses for a New York population of 25,000. Hugh Gaine in his Universal Register for the Year 1787 published round-number estimates of population and houses in New York prior to the 1776 fire; his pre-fire population estimate of 30,000 was probably too high by about 20% — though it might have been at least that high at times during the occupation — but it is also probable that he arrived at his estimate of 4,200 houses using the same seven to one ratio of people to houses. Reducing Gaine’s number of houses by one-sixth to match a population of only 25,000 brings the estimate down to 3,500; given the inherent noise in all these estimates, the difference between 3,500 and 3,570 is de minimis: it’s just not a difference that makes a difference. Though of uncertain provenance, figures in this range are also if somewhat loosely confirmed by the History of North America and The North American and West Indian Gazetteer, both published 1776: the History claims 2,500 buildings, the Gazetteer 3,000. The sources are not independent: the History’s figure may have been taken from the Abbé Raynal’s great history of European commerce in “the two Indies,” published in Amsterdam (in French) in 1770; except for the bump from 2,500 to 3,000, the Gazetteer’s entry on “York, New” is virtually identical, word for word, with the History’s paragraphs on the city. If the History’s figure of 2,500 buildings did come from Raynal’s 1770 work, then the 20% bump to 3,000 by the Gazetteer is more or less in line with the city’s roughly 20% population increase 1770–1776.
The multiple meanings of the word “house” in 18th century usage, however, may make for differences that do make a difference: “house” might mean any kind of building, whether residential or not, or it might mean a residential building, as it generally does today, though residential buildings were often explicitly called “dwelling houses.” But not always: writers just as often relied on their readers’ sense of context to know which of these two meanings of the word “house” was the one intended. Based on Grim’s map of the 1776 fire, it seems likely that his count of 493 houses burnt included both residential and non-residential buildings while Gaine’s implicit ratio of about one house for every seven people suggests that his “houses” were dwelling houses, as were, explicitly, Noah Webster’s.
It’s probably impossible to come up with the exact or even an approximate ratio of non-residential to residential buildings in later 18th century New York, but estimates based on a ratio of one to five — to use, for simplicity’s sake, a very round number — at least don’t lead to absurd results, and although plausibility is surely the weakest of all arguments, the results obtained with this ratio accord fairly well with such numbers as we do have. Based on a one to five ratio and the reduced figure of 3,500 dwelling houses derived from Gaine’s estimate, the total number of buildings of all kinds, both residential and non-residential, standing in the city prior to 1776 fire would have been about 4,200 — that this figure agrees with Gaine’s unmodified estimate is purely coincidental — which would in turn imply that no more than about 3,650 were left standing after losing 550 or so to the 1776 and 1778 fires; the precision is of course spurious: an artifact of any such calculation. The same one to five ratio applied to Noah Webster’s report of 3,340 dwelling houses in New York in 1785–1786 yields an estimate of about 4,000 buildings of all kinds in the city at that time, which would suggest an increase of some 350 over post-fire estimate for 1778. With some allowance for rebuilding in the two to three years since the Patriots’ return — figures of 100 or more new buildings a year are not at all out of line with the two decades before the occupation — the fires of 1776 and 1778 would then account for most if not all of the losses incurred during the occupation.
Of course the difference of 350 between these estimates of New York’s standing stock of buildings in 1778 (post-fire) and in 1786 is less than 10% of either figure, and on that account could simply be due to the unknown amount of noise in the data — and demographic data of any kind prior to the 19th century is notoriously noisy indeed. More data, perhaps from comparable 18th century American towns and cities if such were available — Webster’s numbers would be a start, but only a start, as they do not include non-residential buildings — might yield different results, and different assumptions about the ratios of people to houses and non-residential to residential buildings certainly would, as would a more thorough-going analysis of such numbers as we do have. But as a physicist of my acquaintance says, if you torture the data long enough, it will confess, which is not my purpose here. Time dulls our response to loss, much as distance dulls our response to the suffering of others. New York’s losses to the fires of 1776 and 1778 were great enough to need no exaggeration, but they were also too consequential to pass by without pausing to pay our respects. Still vividly present in living memory for many early 19th century New Yorkers, the fires of 1776 and 1778 cast a long shadow on the city that would be built on the island of Manhattan: the street plan of 1811 that New Yorkers have been living with, happily or not, for over two hundred years was — among other things — a plan for a wooden city.
For Further Reading
I. N. Phelps Stokes’ index to his Iconography of Manhattan Island (1915–1928) lists nearly 50 “disastrous” fires in New York between 1628 and 1913.
A.E. Costello’s Our Firemen (1887 ) is a richly illustrated compendium of information abut fires, fire fighters, fire fighting equipment, and the New York Fire Department in the 17th through 19th centuries.
The great 19th century New York City Clerk David Valentine’s Manuals of the Corporation of the City of New York (1842–1866 & 1868–1870) provide statistics and accounts of fires in New York.
The Union History Company’s long out-of-print History of Architecture and the Building Trades of Greater New York (1899) has a chapter on fireproof construction that incidentally provides an overview of the history of fires in the city.
* Richard Howe is a frequent contributor to the Blotter. He is currently working on a history of how New York was shaped by the principal materials used in its construction: wood, stone and brick, iron and steel, concrete, and glass.