The art and science of furniture connoisseurship

Magazine Antiques, July, 1997 by Philip D. Zimmerman

The history of early American furniture is built on documented examples. Family histories of ownership allow furniture historians to identify specific times and places of furniture manufacture and to reconstruct broader trends in furniture history. However, the cautious collector will look closely at the object as well as its history. A case in point is the mahogany side chair shown in Plate I, one of four given to the New-York Historical Society in New York City by a member of the de Peyster family of that city. According to the donor, a descendant by marriage, the set was part of a larger body of furniture that belonged to John Watts (1749-1836), a prosperous and influential speaker in the New York State Assembly in 1792 and 1793 and a member of the United States House of Representatives from 1793 to 1795.

A comparison of this chair to the late eighteenth-century New York side chair shown in Plate II yields several differences in over-all design. Proportionately, the de Peyster chair is slightly broader and shorter. The dramatic taper of the front legs from top to bottom - measuring an inch rather than the more usual three quarters of an inch - emphasizes the wide stance and broad seat. It communicates a preference for comfort over stateliness, a slight but important departure from design principles typical of the Federal period. Another subtle but equally revealing difference is the quality of the carving in the splat, which, in the de Peyster chair, is uneven [ILLUSTRATION FOR PLATE IC, IIA OMITTED]. Determining whether carving is not of the period or is simply inferior can be problematic, but in the case of the de Peyster chair the carving clearly falls outside the limits of acceptable practice. The mahogany of which the chair is made is very pale and has been stained to resemble the typically dark mahogany of New York Federal furniture. Mahogany does, of course, bleach with exposure to bright light, but in this case it is uniformly pale under the stain.

Repairs to antique furniture may include the replacement of seat rails with new parts cut with modern saws, or the use of dowels to repair failed mortise-and-tenon joints. In both cases there will be evidence of scarring or cracks, indicating that joints have been disassembled. In the case of the de Peyster chair, there are band-saw marks on the inside of the ash seat rails [ILLUSTRATION FOR PLATE IA OMITTED] and the joints are doweled rather than mortised and tenoned [ILLUSTRATION FOR PLATE IB OMITTED]. There is no evidence that the chair has ever been taken apart. Any one of these observations raises suspicions about the age of the chair. Together they make a compelling argument that the chair and the other three in the set were made in the late nineteenth or early twentieth century. Despite the alleged history of ownership, Watts could never have seen these chairs.

The discovery of such unequivocal evidence of modern workmanship as the use of a band saw or dowels might suggest that the efficient and knowledgeable furniture connoisseur should look for this kind of thing first and then look no further. However, as the next two case studies show, it is essential to conduct an orderly and progressive review of all the evidence before drawing conclusions.

An examination of the Philadelphia walnut dressing table shown in Plate III reveals several glue lines where pieces of wood have been joined to create the necessary width. The top, which is original, is made of two boards glued edge to edge - a common practice in all kinds of eighteenth-century furniture - although collectors tend to prefer single wide boards. The bottom half of the pierced heart in the center of the skirt is glued to the upper section, which could be the result of a break but was almost certainly a feature of the original heart to create the necessary width without sacrificing a wide board.

More glue lines appear on the case immediately above each leg bracket, where strips of wood one eighth of an inch high have been laminated to the bottom of the case sides and to the skirt [ILLUSTRATION FOR PLATE IIIA OMITTED]. These strips extend the edge of the boards so that they align with the beginning of the curve of the leg. At first glance these strips appear to be repairs - inserts to compensate for the shrinkage of the boards across the grain as they dried.

Thin laminations such as these are common repairs, but there are three mitigating circumstances suggesting that the laminations may be original. First, the twenty-inch-high boards used for the sides of the case could easily have shrunk one eighth of an inch. However, the front skirt board is only three inches high and would have shrunk only one seventh as much, yet here too there are one-eighth-inch laminations [ILLUSTRATION FOR PLATE IIIA OMITTED]. Second, the underside of the laminations and the surrounding wood have oxidized to a uniform color, and there are no extraneous toolmarks, suggesting that nothing has been altered since the piece was made [ILLUSTRATION FOR PLATE IIIB OMITTED]. Third, the glue line on the heart aligns with these laminations indicating that no smoothing of damaged edges has occurred.