A Short History of Hand-Joined Wallpaper
by Robert M. Kelly
(presented at the "Early American Wallpaper" workshop, Eastfield Village, Nassau, N.Y., July 17-20, 1995)
The Value of Raboute
Before addressing the history of raboute (joined paper) it would be well to discuss the value of it. After all, from a distance a reasonable well-joined strip of wallpaper is almost indistinguishable from a continuous strip. So why study it at all?
Joined paper is important as a dating tool. There have been many cases of joined paper showing tax stamps and watermarks; there are even instances of handmade lining papers used under wallpaper carrying a date[1]. This type of evidence becomes much less common after the introduction of continuous paper.
Wallpaper manufacturers have always been promiscuous and unrepentant borrowers. The "quicksands of style and fashion", as Fowler and Cornforth put it, are unreliable indicators of the time and place of manufacture. In contrast, the evidence of a joined wallpaper is very concrete. There is only a twenty year changeover (1830-1850) between the era of joined paper and the era of continuous paper. And there are distinctions within the technique of joining, as we shall see.
Another value is that the time frame of joined paper (1700-1850) represents the solid middle course of wallpaper history, a time when wallpaper became much more than a "domino" but not quite an everyday item. This tumultuous period saw wallpaper breaking away from stationer's shops and the furniture trade, and carving out it's own place in the stylish "Paperhanging Warehouses" in the years leading up to the Civil War. It is remarkable that many of the highpoints of wallpaper production (English flocks and fancies, French arabesques, scenics and draperies) were achieved within this period; joined paper was no impediment to invention, nor was the continuous roll a guarantee of further progress.
Perhaps the greatest value of joined paper is that it stands for early wallpaper production itself, a simple but profound human achievement. Joined paper succeeds in spite of itself, and confirms the human dimension in a product that is artificial by nature. Like the slight squeaks and pauses that accompany the masterful guitar playing of Segovia, joins and seams in wallpaper are a constant reminder that it is the work of human hands.
In the present age the typical wallpaper consumer seems to believe that a completely smooth and seamless wall of design is the ideal. And yet, the sight of joined paper confronts us with a different reality. The joined sheet is a mute reminder that wallpaper is nothing more or less than the working up of paint and paper into an object of delight.
Nancy McClelland said that "....wallpaper has succeeded in being always a reproduction and yet keeping always a definite character of its own, due to its texture and its processes." The joining of paper is an important part of the "definite character" of wallpaper.
English and American Beginnings
In Chapter Two of "The Papered Wall", Anthony Wells-Cole[2] documents the development of joined paper in the years leading up to 1700. Of particular importance is the testimony of John Houghton, who described both types in 1699: "......a great deal of Paper is now a-days so printed to be pasted on Walls, to serve instead of Hangings; and truly if all Parts of the Sheet be well and close pasted on, it is very pritty, clean, and will last with tolerable Care a great while; but there are some other done by Rolls in long Sheets of a thick Paper made for the Purpose, whose Sheets are pasted together to be so long as the Height of a Room.......". The "thick Paper made for the purpose" was (and occasionally still is) called "cartridge paper". The words "cartridge", "cartouche" and "cartoon" are all related to trade practices.
Notwithstanding Houghton's reference to the height of a room, which obviously varied, it is surprising how quickly a "piece" (standard roll) came into existence. There is a c. 1700 advertisement of the Blue-Paper Warehouse that refers to a roll 12 yards long and a half-ell broad. This was apparently the length already in use for fabric hangings. An English "ell" was based on the dimensions of an outstretched arm and had become standardized at 45", so the "half-ell" was about 22" or so.
The surviving examples of the single-sheet type of paper described by Houghton, with their wide black borders, dark stenciled outlines and simple washes, look much like the French domino papers[3]. It must have been types like this that made up the single sheets of painted papers in the 1700 estate inventory of stationer Michael Perry of Boston, which is the first reference we have in the U.S. of paper intended for walls.
It is curious that no trace of a domino paper has come to light on an American wall. It is only my hunch, but it may be that the hanging of the newly joined paper may have been so much easier for unpracticed hands in the new world that "roll paper" quickly superceded single sheet papers. Whatever the truth of this, by 1730 or so advertising references to "Stampt Paper in Rolls" and "Roll Paper for Rooms" begin to outpace references to painted paper sold in quires (a bundle of 24 or 25 individual sheets).[4]
Breaking the Bounds
It is well not to be doctrinaire about the sizes of the early sheets, for there were many variations. Suffice it to say that a common size was about 22" x 19", with the 22" being the horizontal join.
As pointed out by Rosoman, "......while this is certainly not as large as can be made by hand, it is as large as can be made with a view to high volume production (i.e., between 1,000 and 2,000 sheets per day), assuming a vatman who is not only highly skilled, but who also has great strength and stamina.[5] The vatman was assisted in this work by the coucherman and the layer.
The whole point of joining the sheets prior to printing, of course, was to allow for flexibility in design and printing. Whatever the technique of printing, joined paper allowed a paperstainer to break the bounds of the single sheet. Once "roll paper" was in use, the door was open to imitate an ever widening variety of designs once reserved for fabric hangings.
The word "dozens" became associated with wallpaper, and it is sometimes said that this term was related to the "double-demy" sheet size of 36" x 22", twelve of which would have made up a roll. However, there are many references to rolls of 24 sheets, not twelve, and much more documentation that limits the term "dozens" to borders[6]. A more likely explanation for the term is that it refers to linear yards, which were the usual selling unit for borders. A narrow border of 1 or 2 inches would make up about a "dozen" courses across the width of the paper. A person ordering "12 Dozens Oak Leaf Border", for example, as at Prestwould, may have been ordering 12 yards of border, which could easily contain as many as 300 linear feet of 2" border, when they were all trimmed out.
Later in the century some English production featured a 23" x 28" size for joined wallpaper, with the 23" continuing as the horizontal join. The 23" x 28" size is a more economic size, in that fewer sheets and hand-joining operations were required to produce a "piece". It is not surprising, then, that in 1836 the writer of an English engineers' and mechanics' encyclopedia stated: "Elephant size handmade paper (28 inches by 23 inches) is used almost exclusively for the manufacture of paper hangings, being joined together and printed on".[7]
A Digression: Some Paperhanging Practices
It is this larger size, known as "elephant" from the tradition that it was marked with a crude watermark of an elephant, that was also used as a lining paper under flocks and other delicate papers. Another common practice throughout the century was for a paperhanger to paste the individual pieces of elephant all over a wall, then paint the whole and border the perimeter. The usual method was to stretch canvas on battens, (or apply it to the wall), followed by the elephant, several coats of paint, usually a "drab" color, and finish with gilt or paper mache border. There is documentation in England for this practice being performed by Chippendale's outworkers.
In the U.S., Lord Botetourt of Williamsburg, ordering in 1768, asks for prussian blue and verditer pigments and ".......2 reams of fine large elephant paper 2.10.0", apparently for the decoration of the ballroom at the Governors Palace. It is possible that Joseph Kidd, an experienced upholsterer indentured to Botetourt in England who came with him to the U.S., was responsible for the work.[8]
A smaller sheet size is specified in an interesting notation by Thomas Jefferson, writing in the following year: "Send for cartridge paper to color on for room. It comes in quires each sheet 18 I. Sq. costs 1/ a quire. 25 quire will give a dble coat to my dining room so send for 3 reams. Verditer Blue. Prussian blue. Spanish white. Cuttins of white leather to make a size to prevent its rubbing off...Rolls to hang paper on [possibly linen], yard wide, 10d sterlg. p.yd."[9]
During the construction of the Riddell-Carroll House in Baltimore, in a bill of July 13, 1811, paper hanger Robert Elliot charges for "1 Rimm Paper for lining walls...........3.50" and "Lining walls & rubbing down w Pumice..........8.00". In this case the paper is apparently used as a liner, because the next several entries are for "Pannell Paper". It is interesting that the liner was sanded after installation, probably to remove ridges and deckles to make a smooth surface for the extensive panel decorations.[10]
In London, the practice of lining with elephant and painting continued until at least 1815.[11]
It is interesting that the two surviving American installations of elaborate handpainted "landscape papers", both products of London tradesmen, and both dated to the 1760's, feature a sheet size of 23" x 28". The Van Rensselaer wallpapers, now at the Metropolitan Museum, seem to have an approximately 1/4 inch overlap at the horizontal join, which uses the 28" dimension of the sheet.
The handpainted wallpapers at the Jeremiah Lee Mansion in Marblehead, Massachusetts, were observed by this writer to have an approximately 1/4 inch horizontal overlap, although the paper conservator who worked on the papers, T.K. McClintock, recollects larger overlaps and some untrimmed overlaps as well; in other words, a variety of shop practices. No doubt the unusual placement of the 28" on the horizontal is due to the custom handpainted nature of the large panels that make up the decoration. Plate 28 of Rosoman's "London Wallpapers" shows another painted paper, c. 1755, made up of sheets about this size.
The Van Rensselaer paper was examined by this writer in November of 1994, and the following is from my report: "The most important finding was that, like the wallpaper at the Lee Mansion, the wallpaper is definitely put up in vertical strips hand-joined and hand-painted in the workshop. When first encountered the wallpaper presents a puzzle. Although the underlapping vertical seams are clearly visible and are fairly uniform, the overlapping individual sheets making up the panels seem haphazard, almost randomly applied. This leaves the impression that they may have been applied domino style, either in the workshop or on the job. There are also areas of greater regularity, where the vertical seam lines up uniformly."
The answer to the puzzle is that at every one of the randomly applied areas, where the vertical seams did not line up, there was a repair. It took some time to confirm this, but the pattern of repairs left no doubt that when a repair was carried out, the paper conservators simply closed up a tear, resulting in a shortened displacement at the vertical seam, or added paper, resulting in a wider seam than was the case originally. See diagram for a sketch of this process. (reproduced here as Appendix A.) It is not known why they did not observe the original method of production and installation".
The best account of the constituent sheet sizes of early American joined paper is in "Wallpaper in New England". Abbott Cummings describes the search undertaken by John Welsh, Jr., for appropriate sheets for paperstaining. Interestingly, he merely echoes European standards, asking specifically for sheets that are at least 22" to 23" in one direction; he mentions that he has recently bought (in 1785) two different types of paper: one shipment was 23" x 28", and the other was 23" x 17 1/2".[12]
The engraving by Samuel Hill of early paperstaining techniques shown in Appendix B[not shown here] is from the billhead of Ebenezer Clough of Boston. Dated 1798, it very likely shows the gluing up stage; single sheets on the left are glued and joined together into rolls, using the simple press on the table.
The extensive redecoration of two well-documented American houses help us to visualize the transition from joined to continuous paper: At Prestwould the papers supplied and installed by Francis Regnault in 1831 were joined sheets, but by 1841 the paper newly installed at Lindenwald, as Martin Van Buren prepared to retire from the Presidency, were on continuous paper. In her study of the Lindenwald papers Andrea Gilmore found that they were block-printed, and bought from Pares & Fay in New York City. Although Pares and Fay were both "importers and manufacturers" it is likely, based on style, that at least some of the papers were American made[13].
French Standards
There were several customs in France at variance with English practice. There seems to be no counterpart in England for the counter-proofed symmetrical single-sheet "Papiers de Tapisseries" of Papillon's workshop that rather miraculously broke the bounds of the single sheet while always remaining a single sheet.[14]
The elaborate use of single sheets in the early 18th century, so well documented in the drawings of Jean Michel Papillon, far from dying out, continued right through the 1760's, and even beyond. But it is a mistake to call all of these types "domino papers". According to an authoritative definition, the term "domino" can be applied to any type of decorated paper (sierpapier, imagery, lining papers), whatever the technique (marbled, printed, patterned) as long as it consists in a sheet measuring about 12" x 16".[15] On the other hand, the term "Papiers de Tapisseries", used by writers of the period, is more descriptive of the papers depicted by Papillon; they were intended specifically for walls.
Moving on to joined paper, which was not usual in France until 1765 or so, the sheet and roll sizes used in France seem always to have been a bit smaller than their English counterparts. Horizontal overlaps of 17" and even 16" are not uncommon in surviving French wallpaper found in America, and the pattern widths, too, often edge closer to 18 or 19 inches rather than the English standards of about 21 inches. As always, there are variations in the widths of the papers, especially with scenics. As for length, in general a "piece" of only about 9 meters in length seems to be the French norm, and at times it could be even shorter.
A special case are the arabesque panels, which often feature cross-joining as well as vertical joining in order to provide enough width for the design, as well as an appropriate amount of free space between the design and rectangular borders; an example is shown in Appendix C. It is fitting to note here that the Musee du Papier Peint in Rixheim continues to lead the way in wallpaper study.[16] Their appreciation for raboute in their studies is shown by the inclusion of abstracted diagrams (releve) showing patterns of joined paper. They do this for sidewall papers as well as for the more elaborate arabesques and papiers de tapisseries.
In Appendix D and E[not shown here] of this paper are images of the Arthur pere et fils & Grenard factory, Paris, c. 1788.[17] The colleuses et rogneurs (pasters and trimmers) section (just below the penthouse) shows long tables and a trimming apparatus.
The working methods of some 40 years later are described in Appendix G, an extract from Sebastien Le Normand's "Nouvel manuel complet du fabricant d'etoffes imprimees et de papiers peints" of 1832, pages 190-193. Le Normand is describing working methods of Dufour's factory.[18] Some trade activity can be seen in Appendix F[not shown here], which shows a Sevres plate of about 1828.
The joiners were paid as well as the printers and grounders in ReveillonÆs factory of 1789. They were all part of the 2nd class of workers, earning 30-50 sous per day.[19] They may have been toward the 30 sous end of the pay scale, as we can see in the c. 1788 illustration that this work is being done by women, not to mention that in the Le Normand it is said that a little girl performs the pasting operation.
The Le Normand extract is cross-checked by consulting a slightly later source, Dr. Andrew Ure, who, in his "Dictionary of Arts, Manufactures, and Mines" published in London in 1839 described the joining methods of a French wallpaper factory. The Zuber factory is cited as the source for his descriptions. It will be noted that there is only one difference that stands out from the description of Le Normand 7 years earlier, but that one difference is a critical one.
He states that the following is standard practice(18): "1. The paper should be well sized. 2. The edges should be evenly cut by an apparatus like the bookbinder's press. 3. The ends of each of the 24 sheets which form a piece, should be nicely pasted together; or a Fourdrinier web of paper should be taken. 4. Laying the grounds, is done with earthy colours or coloured lakes thickened with size, and applied with brushes."
The nature of joining paper
Handmade paper, because its fibers are omnidirectional, will baffle prospective present-day joiners by its tendency to "waffle" rather than expand at an even rate, across the grain (width), as all machine-made paper does. Handmade paper will often cockle at the join, and some small wrinkles are almost inevitable. But because of the long fibers and rough texture of the paper, it is remarkable how these minor flaws recede into the background, along with the join, when grounded and printed upon.
To return to John Houghton's account of 1699, another important clue to successful joining is retold by Treve Rosoman in "London Wallpapers": "Houghton described how the sheets were glued together along the deckle edge, where the paper was thinner and therefore the joints would show less clearly."
Although this may have been true enough in 1699, the trade practice evolved, so that by mid-century it was probably more common to find either a cut edge on top, and deckle on back, or a cut edge on both overlap and underlap. This allowed for more precise alignment and explains how the overlaps at the Lee Mansion and the Van Renssalaer paper could be so slight; they measure approximately 1/4 inch.
Here in the U.S. the same progression can be confirmed at the Cooper-Hewitt study collection of early American wallpapers, where most of the early overlaps are long, with deckle edges underneath, and most of the later overlaps feature two cut edges, and are on the short side, approximately 1/2". William Flynt, architectural conservator at Historic Deerfield, has also found cut edges on front and deckle edges on the back, with an overlap ranging from 3/4 to 1 inch, on early American papers at Deerfield.[20]
Postscript: "A superior article without joints"
Correspondence on the subject of the transition to continuous paper was received recently from Bernard Jacque, curator at the Musee du Papier Peint in Rixheim. Regarding the size of the overlap, he states that: "Le Normand in the extract of 1832 speaks about 'six lignes", i.e., during this period about 1,8 cm.........When I look at our documents, it depends on the trimming of the sheets. In the middle of the 18th century, the sheets were not trimmed and the raboutage is long, about 2 cm. The Reveillon's are trimmed and in the 19th century the raboutage is always short, about 0,8 cm. Your date of 1835 (for introduction of continuous paper) is similar here: the first continuous paper roll was in the collection of 1832 (printed in 1831) and in the following years, Rixheim also sold paper rolls to the other mills. Louis Andre, curator of the Musee national des techniques in Paris has written a good theses of the industrialization of papermaking".[21]
As a survey of published photos will readily show, the horizontal joins in joined paper only rarely coincide from strip to strip. Examination of joined paper will also show that the overlap is toward the top just as often as toward the bottom; there seems to be no particular rule. Obviously, the only reason for the variation is the direction of the paper on the table just prior to printing.
By 1834, an advertisement in New Bedford, Massachusetts offered a wallpaper described as "a superior article without joints." But the continuous roll did not immediately replace the joined roll. As Catherine Lynn notes, "It is all but impossible to establish a cutoff date for the use of joined sheets for American made wallpaper. There was doubtless a transition period of mixed use of handmade and machine-made paper in the wallpaper industry between 1820 and 1840 - and perhaps later."[22] On the other side of the ocean, Rosoman too, says that there is evidence of joined paper in rolls later than one might think, into the 1850's.[23] It is appropriate that we leave our account of hand-joined paper with the date of 1850, which is the also the cut-off date for the Eastfield Village workshop on Early American Wallpaper.
Excerpt from "Manuel du fabricant de papiers peints", by Sebastien Le Normand, Paris, 1832, p. 190-193. Translated by Jeanne Randorf for WRN Associates and edited by Robert M. Kelly.
# II. Trimming of the Paper
It is important that the cut paper be square in order that the gluing, which should immediately follow this operation, can be done evenly, and that the finished roll which results from assembling several sheets of paper has two side edges that are decidedly straight and parallel.
The tools that one uses are a press and a bookbinder's knife. The worker takes two reams of paper (1,000 sheets) laid in a perfectly even stack on a board larger than the size of the sheets of paper; then, on top of the paper he places another board which is the same size as the sheets minus the trimming allowances which are to be removed and which are as narrow as possible so as to waste the least amount of paper.
This board is made with great care to ensure that the corners are straight; that is to say in shop terms, that the piece must be squared. One places this assembly between the "twins"[24] of the press in such a say that the front board be touching the surface of these "twins", and one tightens the vise; then one removes all the excess paper with the trimming knife. One loosens the vise just enough to allow one to turn the entire assembly without disturbing the paper alignment; one trims this side after having tightened the vise; and one continues in this manner until one has trimmed the four sides; then the worker delivers the paper to the gluer.
# III. Gluing of the paper
Each roll of wallpaper is generally composed of 24 sheets that one glues end to end, the gluing edge being the widest edge, and this roll forms one piece. The gluing operation is extremely curious, because of its exactitude and the facility with which it is done.
The gluer, who is ordinarily a little girl[25], carries the paper flat onto the end of a table which is much longer than the piece (will be); she takes twelve sheets which she spreads out in one movement of her hand, and with the help of a small, flat piece of wood which is rounded on one side, she places the twelve sheets so that the edges overlap each other. [literally, "places in stair steps"]; each overlaps the one underneath by about 1/2 inch[26]. She puts a fairly heavy stone on these 12 sheets which are placed to her left; on her right she spreads another 12 sheets in the same way; but she spreads them only 1/4 inch apart from each other[27].
This operation is done with great ease. She places, flat on the table, an evenly stacked pile of sheets, no edges out of line: with the flat piece of wood she pushes them lightly, the sheets slide in a parallel fashion, and with a simple turn of the hand, they are moved to the desired distance. With a wide brush, from 4 to 5 inches in diameter, she spreads flour paste on the overlapped parts of the 12 sheets that she has on her right; then she places them one after the other on those which are to her left and which are spread out at 3/4 inches of distance[28], being careful not to cover more of one edge than of the others, in order that the edges of the piece may be straight. She follows the edge of the table which is perfectly straight. This operation can virtually be done without looking, if the sheets have been properly spread.
Having glued the first twelve sheets, she lays a thick plank on top that is weighted down with a heavy stone, in order to give the glue time to set. The second set of twelve sheets being spread out, she continues to glue in the same manner, until she has placed 24 sheets end to end; these then constitute the piece or roll.
1. see T. Rosoman, “London Wallpapers”, 1992, English Heritage, Plate 15.
2. Wells-Cole has done the pioneering work in describing early English wallpaper production and use. His study centering on Temple Newsam in 1983 (Historic Paper Hangings) paid careful attention to the nuances of joined paper, use of lining papers and fabrics, cost of the wallpaper, and relative size of sheet and repeat. This high level of interest in the material culture of wallpaper has been taken up by several subsequent researchers.
3. see for example "The Papered Wall", 1994, Abrams, ill. 26.
4. C. Lynn, “Wallpaper in America”, Norton, 1980, pg. 32.
6. see for example the bills for Schuyler Mansion (1761) and Prestwould (1799), both shown in “Wallpaper In America”.
8. Wallpaper file of Colonial Williamsburg.
10. Winterthur Portfolio, Vol. 28, Nos. 2/3, pg. 128.
11. Letterbooks of James Duppa, Victoria and Albert Museum, 86.AA.14.
12. A. Cummings, in “Wallpaper in New England”, 1986, SPNEA, pg. 10.
13. for Prestwould, see Antiques Magazine, Jan. 1995. for Lindenwald, see Historic Structure Report, National Park Service, and letters from Van Buren to Harriet Butler in the Joseph Downs Collection at Winterthur.
14. P. de Fabray, “Technique et Papier Peint”, 1992, Rixheim, pg. 69. see also G. Wisse, “The Papered Wall”, pg. 13.
16. see, for example, “Petit Vademecum Technique A L’Usage de l’ Amateur de Papier Peint Ancien”, by Bernard Jacque, a chapter in “Technique et Papier Peint”.
17. from the collections of the UCAD, Paris, as shown on pg. 59 of “Les Papier Peints en Arabesques”, 1995, Paris.
18. I am indebted to Bernard Jacque for sending me the extract and to Jeanne Randorf for her translation of it.
19. in the English version of Reveillon’s comments on pg. 58 of “The Papered Wall” the word for “joiner” is mistakenly translated “carpenter”. For another version, see N. McClelland, “Historic Wall-Papers” pg. 119. Alas, McClelland also mistranslates; the workers who earned up to 3 francs a day are not the “gluers, who fastened the sheets of paper together”, as she says, but the paper hangers, yet another part of the Reveillon empire, who installed the wallpapers throughout Paris.
20. correspondence, Sept. 1, 1994.
21. see “Technique et Papier Peint”, pg. 87. for a superior study of the mechanization of the paper industry, see “Most Wonderful Machine: Mechanization and Social Change in Berkshire Paper Making, 1801-1885” by Judith McGaw, Princeton University Press, 1987. McGaw includes an excellent summary of handmade paper techniques and mentions the Laflin wallpaper factory several times.
25. “....une petite fille......”
26. “......d’un demi-pouce environ.”
27. “......qu’a deux lignes l’une de l’autre.”
28. “.......et qui sont echelonnees a six lignes de distance,.....”
Back to the top of the page.
