|
Dutch Barn Preservation Society Dedicated
to the Study and Preservation NEWSLETTER,
SPRING 1997 by Neil Larson A short time ago, I was invited to co-author a chapter for a book on the architectural information contained in the assessment schedules from the 1798 Direct Tax. Tom Ryan, a doctoral candidate from the University of Delaware, where the project had developed, and I were assigned to review and evaluate the lists that survived for New York. The Direct Tax was the first levy made by the U.S. Government on the states. It was instituted to raise money, approximately two million dollars, to improve the nation's coastal defenses. Unlike the income tax with which we are all- too-familiar today, this tax was based on property value. To this end, every property in the United States was inventoried and given a value. What a remarkable record! Every house, farm building, mill, shop, every building was itemized and described, in some cases, down to their window size (it was this assessment that generated the myth of the window tax) along with their lands and together with the slaves that inhabited the property. This was truly a benchmark view of the extent of cultural landscape in the United States at the end of the eighteenth century.
But, this story does not have a happy ending. As it turns out, precious little of this information survives, especially in New York. Because the Direct Tax was computed on aggregate property values compiled by each state from summaries made by locally-appointed assessors, state and federal records do not contain the actual property-by-property inventories. These documents remained in the hands of town assessors and have largely disappeared from view. For New York, assessment lists for only eight towns have been located. All but one are in the Hudson Valley: four lists survive from Ulster County, two from Orange and one from Dutchess. Only one town has a complete set of three assessment schedules: Schedule A, which itemized houses valued at $100 or more (it is only this schedule that survives in all cases); Schedule B, where houses valued at less than $100 were recorded along with all other building types-including barns-and improved land; and Schedule C, which was a census of slaves residing in the town. The town fortunate enough to have all three schedules is New Paltz in Ulster County. There has been no secret to its existence. It is a town document and a copy of it has been used for years by local historians and Realtors at the Huguenot Historical Society Library. However, to my knowledge, it was not until the current project to recover and analyze the research value of the information contained in the assessment lists that this extraordinary data was looked at in a systematic way. Many interesting things were discovered about the architecture of these towns and the region, but you will have to wait for the upcoming publication of the book to read about it. I am going to use the space here to reveal what the assessments made in the Direct Tax have to tell us about Dutch barns. The immediate thing to recognize on the New Paltz B Schedule is that it actually refers to certain barns as "Dutch barns." Since plan dimensions are included in the entries (due to the fact that the 88 barns so classified all have the square footprint on which we now rely to distinguish Dutch barns), we can be confident that historic and current usage refer to the same buildings. This early occurrence is significant on at least two levels. First, it documents the use of the term to the late eighteenth century. This is a wonderful thing to see in print, to discover that the term goes back before John Fitchen's The New World Dutch Barn or Helen Wilkinson Reynolds's Dutch Houses in the Hudson Valley before 1776, or any other secondary source we have used as an authority. Second, this choice of words indicates that the word "Dutch" was a critical modifier two hundred years ago. It is unarguable from this perspective that the term was actively used not only to characterize a barn form, but to associate it to a cultural group. There are names linked to all the barns on the list, and this allows us to actually see who had "Dutch" barns and who did not and determine the role of ethnicity, kinship and class (for values are a key factor in this assessment) in the choice of barn types. This process is helped in that all barns were placed in some kind of descriptive and dimensional category. Other barns listed in Schedule Bare classified as simply "barns," "frame barns," "log barns," and there is one "crotch barn." With Dutch barns clearly identified in this way, we can have the
opportunity to interpret them in the architectural, geographical
and cultural contexts that the assessment lists create. For example,
the dimensional data for Dutch barns can be organized to document
the range of barn size and compare the scale of the building to
the relative wealth or the extent of arable land of farms. In New
Paltz, land is described and assessed on Schedule B and given a
geographical reference, so it would be possible to map farms and
estimate their productivity. (This still needs to be accomplished.)
We can immediately compare the ethnic, wealth and class distinctions
between Dutch barns and the others. Just a cursory review of names
confirms what is generally understood about Dutch barns: they were
owned overwhelmingly by the local "Dutch" families in
identifiably "Dutch"
communities. There is another source of information in the New Paltz assessment lists that involves the Dutch barn in another significant cultural issue: evidence for the African-American presence on the eighteenth-century cultural landscape. As noted above, a third schedule used for assessing property value for the 1798 Direct Tax was a census of slaves. Unlike the other two schedules, which gave names to the owners and occupants of the inventoried properties, slaves were anonymously enumerated by age and gender (the basis by which they were comparatively valued). However, they were identified with their owners, so we can identify who in New Paltz owned slaves and how many they owned broken down by their age and gender. As with Dutch barns, this data allows us to see the relationships of ethnicity, kinship and wealth in slave-ownership more clearly. But, more importantly in this instance, it allows us to examine the coincidence of slavery and Dutch barns. By using the three assessment schedules together, we are able to reconstruct the slaves' milieu and locate them in the landscape of the town and on the farms where they lived. We can see that they were employed in agriculture, commerce and trades. Where were they housed? This remains a nagging question in slave and architectural history, but the lists provide a certain insight in the fact that they do not specifically link slaves to any dwelling type while they list a variety of accessory buildings: kitchens, ells, linters (leantos), outlets (outshots), shops and barns where slaves would have worked and possibly resided. If nothing else, it is significant to realize that over 83% of the 280 slaves that were counted in the Town of New Paltz in 1798 lived on farms that also had Dutch barns. Clearly the Dutch barn was a conspicuous landmark in the African-American experience. Using the data from the lists, let's now turn to some of the actual information about Dutch barns and the cultural landscape in which they existed. Dutch barns and other barns The floor plan of the smallest Dutch barn in the Town of New Paltz in 1798 measured 30 feet by 19 feet, with the larger dimension presumably being the facade. Two of these would practically fit in the largest barn, which was listed as 60 by 50 feet. In between, the barns ranged widely in size, although most (55) were from 45 to 50 feet wide. They were typically slightly rectangular; only fifteen of the eighty-eight Dutch barns were exactly square in plan. Ten Dutch barns measured 50 by 40 feet and eleven were 50 by 45. These were clearly the most frequently repeated dimensions in the entire tax list. The other barns cited on the list were significantly smaller. The ninety-five barns entered simply as "barn" or "frame barn" ranged from 22 to 48 feet along their principal facade. These were what are commonly called English barns today, and would have had their threshing floors and large wagon doors on an axis perpendicular to the roof ridge. Still, they were frequently as square in plan as the Dutch barns suggesting that by 1798 the Dutch barn could have been evolving into the hybrid type seen in surviving examples where the Dutch bent frame was preserved while the facade and central aisle were rotated ninety degrees to the side elevation. The median sizes recorded for "barns" and "frame barns" concentrate in the range of 30 to 45 feet, significantly smaller than the median "Dutch" barns. The thirty additional barns described as "log barns" were even smaller. The smallest log barn measured only 16 by 15 feet. The median size for log barns was at the higher end, however, although the largest was only 30 by 20 feet. Therefore, in terms of scale, there was a hierarchy of materials and form with the Dutch barn clearly the largest and most substantial of all the barn types recorded in the Town of New Paltz in 1798. Dutch barns and their farms When the data on assessment schedules A and B are combined and organized by name of occupant, a fuller view of the farms in New Paltz begins to emerge. This gold mine of comparative data has yet to be examined in any detailed or systematic way, but even in its relatively raw state we can identify the milieu of the Dutch barn on a farm by-farm basis. Just in terms of buildings (the extent of land and land value relative to Dutch barns still awaits even cursory analysis), we can make some interesting observations concerning common assumptions. For example, sixty five New Paltz farms had both Dutch barns and stone houses. This combination represented approximately three quarters of Dutch barns (73.9%) and stone houses (75.6%). Thus, the image of stone house and Dutch barn together is an accurate one. Still, twenty-three Dutch barns were paired with wood houses, and at least two of those were log houses! More than half of the Dutch barns (46) were the only other building inventoried with the house. The other entries included a wide range of outbuildings, such as hay houses, corn houses, smoke houses, stables, and various shops and sheds. This indicates that for many farms, even at the highest property values, the multipurpose Dutch barn was the sole building directly linked to farm activity.
After barns, the farm building most commonly listed is the "hay house," which appears in 45 different entries. It is worth mentioning in this context because it is paired with Dutch barns in 31 of those instances (it is cited with other barns only five times and without barns the rest). The dimensions of these hay houses. range from 20 feet by 18 feet to 60 feet by 20 feet with the median around 30 x 18 feet. These were not barracks. The measurements describe a rectangular building of a standard width and of varying lengths that would be similar to the adjunct frequently seen attached to Dutch barns that are dedicated to hay storage, sometimes with animal stalls on the ground or basement level. In 1798, the presence of hay houses in this number and in combination with barns would document the increasing importance of hay and, by extension, numbers of cows (likely milk cows) in the farm economy. With only thirty of them inventoried on the most successful farms, one supposes that this is the harbinger of the Dutch barns' transition from wheat to dairy facilities in this part of the region during the early nineteenth century. Dutch barns and class, wealth and ethnicity Considering stone houses were the most highly assessed object in the town, the correlation between them and Dutch barns firmly associates the agricultural building with the wealthiest properties in New Paltz. In fact, all the Dutch barns except four are associated with houses valued $100 or above on Schedule A, stone or otherwise (one of them belongs to the church and was not listed with a house). In 1798, using just Schedule A, the average value of farms that had stone houses on them was assessed 33% higher than the average value of farms with frame houses ($2942 vs. $2205). If the most highly valued farm in the town is removed from the list of farms with frame houses (Cornelius Dubois's 1668acre property was valued at $12,800), this value percentage of stone house farms over frame house farms grows to fifty-eight percent. Much of this value difference is represented in the houses themselves. Taken alone, stone houses on Schedule A were valued, on the average, 216% higher than frame houses on the same list ($429.19 vs. 198.85). Land values probably playa role in this comparison, but the complex manipulation of the 1798 assessment data has yet to be done. The class distinction between stone houses and frame houses is further demonstrated when properties itemized on Schedule B are considered. There are only two stone houses on Schedule B while forty-four frame houses with a value less than 100 dollars were tabulated there, ten more than are recorded on Schedule A. Thus the frame house overlapped value and class distinctions whereas the stone house did not. Log houses were valued consistently below $30, some as low as one dollar, and represented the lowest end of the socioeconomic scale in the Town of New Paltz. Dutch barns were not typical on leased farms. Of the 151 properties listed as rented in the 1798 New Paltz assessments, Dutch barns were identified with only fifteen of them. Among these, only one was posted on Schedule B, the farms of lesser value. Of the remaining fourteen found on Schedule A, six Dutch barns were on farms leased to individuals who shared the same names with the owners. This indicates that these farms were held within families (some owners were identified as estates) and tenanted by kin. Eight Dutch barns, less than 10% of the 88 inventoried in the town, were leased with farms valued at over $500, some well over $1000, and one assessed at $3500. So, Dutch barns were not characteristic of the leaseholds of the lower socioeconomic classes. Another way to associate Dutch barns in a class structure is through their overt identification with the ethnic patriarchy of New Paltz. The Huguenot Duzine and its kin were the dominant economic, social and political class in New Paltz.The 1798 assessment lists document the extent of this control and position the Dutch barn as a landmark in this cultural landscape. Scanning the owners' names for New Paltz's 88 Dutch barns, Huguenot names are quite prominent, though not exclusive. Families with the names of Deyo (12 Dutch barns), Dubois (10), Eltinge (3), Freer (12), Hasbrouck (10), and Lefever (11) claimed two-thirds of the Dutch barns in the town. The remaining thirty were distributed among twenty-two names, only three of which had more than one but less than five. Many shared names with a broader Ulster County "Dutch" genealogy, like Dewitt, Jansen, Louw, Vanvliet and Vandermark. Others were clearly German: Ein, Hardenburgh, Swart, Terwilliger and Wirtz. Then, some were British in origin, like Broadhead, Birdsall, Donaldson, Merrit, Wood and Waldron. The social and kin relationships of these non- "Dutch" persons warrant greater exploration. Dutch barns and slaves Among the 280 slaves in New Paltz, 145 were male and 135 were female. Eighty-four males (57.9%) were between the ages of twelve and fifty and subject to taxation; sixty-seven females (49.6%) were placed in the same category. Only one female was exempted from taxation because of disability. These slaves were divided among ninety owners. Exactly one-third of these owners (30)' possessed only one slave. This group was evenly divided among three value categories: taxable females (11), taxable males (10), and those too young or old to tax (9). Fourteen of those listed owned two slaves, and fifteen owned three. The numbers of owners decreased steadily as the number of slaves per owner increased. The highest number of slaves recorded for one owner was thirteen (seven males and six females), five of which were taxable, although only ten slave holders owned more than six slaves. Overall the ninety slave owners represented a relatively small proportion of the census. They constitute twenty-two percent of the 406 different names identified as "occupants" for the 1798 Direct Tax. It can be inferred from these statistics that slave owners were, to a very large extent, land owners and were, other than in rare instances, in the wealthiest class established in the assessments. Looking at information about barns and other property provides another way to address the African-American presence. Like the stone house, according to the assessment lists, slave owners were prone to have Dutch barns more than twice as often than English and log barns put together. Slaves were also owned by people who had shops (numerous blacksmithies and one each for weaving, carpet and hat shops), mills (grist, saw, fulling) and a dye house. In both cases, agricultural and industrial, the resource of slave labor was a valuable commodity. Only seven slave owners had no entries for agricultural or industrial buildings in the assessment. Some were widows or estates; all likely had an association with another unidentified person or work location. Clearly, slaves were a commodity available only to a wealthy, elite class. The emerging image of the slave's milieu in New Paltz is one of a large, prosperous farm-often a network of farms and shared labor linked by kinship-with a substantial, multipurpose dwelling reflecting the status and diversity of the household and a well-developed complex of agricultural support buildings. The Dutch barn was a frequent component of this environment. And, consistent with the cultural associations of the Dutch barn, a significant proportion of slave-owners (72%) had surnames which immediately identified them with New Paltz's leading Huguenot families. The New Paltz assessment schedules for the 1798 Direct Tax are priceless documents for exploring the geographical and cultural milieus of the Dutch barn, simply because the assessors there made the distinction in their inventories. It may be that this distinction was made in no other place (assessment schedules did not follow uniform standards, state to-state or town-to-town), but nonetheless, there is a wealth of data with which to work, even if a single barn no longer exists! This article only touches on what is most obvious in and easily retrievable from the data. I have tried to suggest some of the more compelling directions to go with the list. Future work with the schedules, whether it pertains to Dutch barns or other aspects of the cultural landscape, awaits completing the difficult task of mapping the data. I hope in the next year or so, the Hudson Valley Study Center will be able to organize some courses and workshops to work with this document and, using computer data base and mapping tools, recreate the landscape that existed in 1798. This will greatly expand our understanding of architecture, landscape and culture and allow us to visualize what people actually saw two hundred years ago.
A 1796 map shows several farms with Dutch barns along Route 32 near Route 23A in Greene County, NY. (Detail shown.) Although this survey is not from New Paltz in Ulster County, it clearly illustrates the former density of Dutch barns in areas settled by Dutch or Palatine families. Photo by Shirley Dunn. Source: Vedder Library, Greene County Historical Society, Coxsackie, NY. Neil Larson is Executive Director of the Hudson Valley Study Center; State University of New York at New Paltz. |
