What is the difference between a reproduction and a fake? To distinguish one from the other we need to understand the intentions of the maker. During the 1920s and 1930s, as the Colonial Revival style gained popularity, demand for "old-time" furnishings exceeded supply. Manufacturers corrected this situation by offering a wide variety of relatively inexpensive decorative furnishings intended to evoke an antique "look." Actual antiques served as models for many of these furnishings, but no effort was made by the manufacturer to rigorously duplicate the originals in all their detail. Likewise, the manufacturer employed modern production techniques to achieve maximum cost efficiency. These techniques generated telltale and obvious physical characteristics in the manufactured products. The manufacturer was not trying to deceive the buyer, and the new products, which we classify as "reproductions," generally differed significantly from the originals in proportion, size, design detail and method of production.
Fakes, on the other hand, were meant to deceive. The two lamps shown to the left, designed with tulip pattern fonts on columnar bases after the c.1860 Sandwich, Massachusetts originals, provide a classic example. By the 1930s when these fakes were commissioned from a Czechoslovakian manufacturer, apparently by an unscrupulous American wholesaler, glassmakers had advanced pressing technology to the point where they could easily mold an entire lamp in a single operation. In the case of the tulip lamp fakes, however, the manufacturer was instructed to imitate the construction techniques of the mid-19th century. During that earlier period fonts and bases had to be pressed separately and then joined together while still hot with a small wafer of molten glass. Collectors in the 1930s typically looked for the distinctive connecting wafer to authenticate lamps as antique. The Czechoslovakian fakes, with apparent wafer construction, were almost certain to fool them (see detail to left).
In 1938 author Ruth Webb Lee alerted readers to the presence of the spurious tulip lamps in her book Antique Fakes & Reproductions (Framingham Centre, MA: by the author, plate 38). She reported that they had appeared on the American market as early as the 1920s and might be circulating in color as well as colorless glass. In the 1950 edition of her book she illustrated an example with a green font joined to a white base. "Not long ago," she wrote, "an early Sandwich lamp was sent to me for examination. It was obviously new, and bore marks of carefully applied 'age' designed to deceive the purchaser. The lamp is pictured on Plate 117. It is eleven and three-quarters inches high and is merely an improved model of the [colorless] one shown in the lower left-hand corner of Plate 113. The opaque-white in the square base and pedestal is an unpleasant yellowish-white. The opaque green in the bowl is also off-shade. The collar has been dented in, presumably by a hammer, in such a manner as to give a realistic appearance of age. The collar also has been treated with a solution which contributes to its antique appearance. The same copy has appeared in several different color combinations."
Like many fakes of antique glass, the NBMOG pair exhibits induced wear. A stone wheel was used, probably by the manufacturer, to grind a relatively rough plane of "wear" into the underside of the foot (see illustrations to lower left). Later the "wear" was enhanced by scraping the lamp base against sandpaper, concrete or some other rough surface to produce scratch marks. Although stone flatting wheels sometimes were used by 19th-century glass manufacturers to straighten lopsided articles, this additional production expense was avoided or kept to a minimum whenever possible. The NBMOG lamps certainly did not need flatting, to judge from the uniformity of the ground plane around the underside of the foot. Additionally, the parallel orientation and uniform depth of the "applied" scratch marks are typical of induced wear. Normal wear, accumulated slowly over decades as the article is moved across table and shelf surfaces, has a soft overall character, heavier in some places than others, perhaps, but random in the orientation and depth of the scratch marks.
Authors Raymond Barlow and Joan Kaiser make a careful comparison of original and fake tulip lamps in The Glass Industry in Sandwich, vol. 2 (West Chester, PA: Schiffer Publishing Ltd, 1989, p. 103). In addition to color discrepancies they note that the fakes have only 16 beads along each side of the base, while the originals have 17. The dome of the font is slightly flattened on the fake, and the petals around the font are horizontal rather then slightly tipped up as on the originals. Furthermore, they note the apparent presence of the wafer on the fake, but suggest that it was simply a thick thread of glass wrapped around the upper edge of the column. Careful examination of the NBMOG pair shows no evidence of this wrapped thread overlapping itself, which it must have done at some point. In support of the authors' conjecture, however, is the appearance in certain places directly below the wafer of what might be a surface designed to receive the wrapped thread. Also, the underside of the wafer does show evidence of pulling to the left across the column flutes. Additional study could help resolve the question. Relative to any future study, it is important to note that the pair of fakes examined by Barlow and Kaiser in the formulation of their conclusions is the same pair from the Levine Collection now donated to NBMOG.
