Rug Hooking and Tufting: Materials and Tools, 1840 to Present
On August 4, 2012, Michael Heilman
gave a program at The Textile Museum in Washington, D. C. on “American Home Rug Making.”
Tom Goehner, the TM Education Curator, introduced Michael, saying:
“With us today is Weaver Michael Heilman, who will discuss and demonstrate tools developed and invented since 1840 in the United States for use in making rugs in the weaver’s home
Michael is a self-taught rug maker and he has been making various types of rugs for about 13 years. Michael became interested in the rug making process through the discovery of an old rug making tool that he found in an antiques shop; he still uses that very tool to this day. After his interest continued to grow Michael signed up to take a weaving class and since then he has built a number of looms.
When creating a rug Michael combines weaving with tufting and hooking. Michael uses many of his own hand dyed yarns to make his rugs. His rugs have been displayed in a number of venues and he he also demonstrated rug making techniques in places such as the Maryland Sheep and Wool Festival, the Art League of Alexandria and the Virginia Fall Fiber Festival. His recent work has centered on experimentation with various materials and techniques and design. Michael has spent the last 5 years teaching nontraditional rug making at the Art League of Alexandria.”
by saying that while there were a variety of ways in which rugs have been (and still are) being made in American homes, his particular focus would be on the kinds of rug making that consisted of pressing some sort of material through a backing to create loops. The loops could be left or subsequently cut.
This kind of rug making, he said, is usually described as “hooking” or “tufting.”
There is a debate among authorities about the origins of hooked/tufted rugs. Some say it emerged in England as a way of using scrap yarns from the textile mills. Others have suggested that it was brought to Scotland by the Vikings.
Anthony Landreau, once a Director of the Textile Museum, published, in his “America Underfoot,” the hooked rug below,
with this caption:
Landreau indicates that the Shelbourne Museum hooked rugs were made in Amish communities.
Whatever the date of its first beginning, here, the sort of hooking/tufting Michael was treating seems to have developed more broadly in the northeastern U.S. and some parts of eastern Canada, including Labrador and Newfoundland, during the last half of the 19th century. It was primarily a way for poorer folks to make rugs for their own use.
One hears “hooking” and “tufting” used almost interchangeably. Technically, “hooking” should be reserved to describe the process above in which a hooked tool is used.
But working with a simple hook has some disadvantages.
Each time one must identify the hole in the backing material to be filled, push the hook through the backing and grasp, with the needle, the strip of material, or yarn, to be hooked through.
In addition, the rug maker must also control the length of the loop when pulling the cloth strip or yarn through the backing. This can be particularly challenging if the material is dark, as it is difficult to discern the height of the loop against a dark background.
But it was discovered that things could be sped up considerably if the working end of the tool was more like a needle with a hole in it than like a hook.
Once the working strand was threaded through the needle, the needle (now it was advantageous to work from the back of the backing material) could be punched through the backing material as rapidly as one could find the next hole to be filled.
There was no longer any need to search for and hook the working material strand because it was threaded through the hole in the needle.
Michael said that he would describe the materials and tools used over the years in this sort of rug making and would demonstrate a variety of “tufting” tools.
He had a demonstration setup in the front of the room.
It was a frame with a backing cloth stretched taut on it. The frame was attached to a weighted pedestal that provided a firm basis for holding the frame stable in the face of back and forth movement of even the fastest and most vigorous tufting needle, as it pierced the backing cloth and then withdrew.
He said the backing cloth would usually have some sort of design on it (he had drawn some simple ones on this backing). The drawn design would often indicate what colors were to be used in particular areas.
(Note: Technically, when working with a simple hook, one does not have to stretch the backing on a frame. But it is advantageous to do so, and necessary, when the tool one uses entails real acceleration of the back and forth movement of the tufting.)
Michael said that a variety of cloth has been used for backing.
He passed some of these labeled samples into the audience for a closer look.
Although they vary in fineness, you can see that these backings are all woven in a balanced plain weave (that is both the warps and the wefts are equally visible). There are clear “holes” framed by the warps and wefts.
Often the choice of backing is made for the rug maker, since designs are available commercially, and come on a particular backing.
Michael had some examples of hooking/tufting designs available to rug makers.
The design above is for a single open rose and a nearby bud. Notice that color usage is specified with a kind of code.
The commercial design below seems rooted in a device often seen on “penny” rugs.
Michael, personally, believes that something important, on the creative side of rug-making, is lost if one does not fashion one’s own designs.
Below is one of his own designs, not yet realized in actual tufting.
Here is a closer detail of it.
There is something else it might be useful to note, while we are talking about tools and materials, and that has to do with usual practices of hookers and (non-mechanized) tufters, as they worked.
In her book “Floor Coverings in New England Before 1850,” Nina Fletcher Little indicates that the usual practice was to “fill every hole” in the background material, and it is true that some hookers, like those at the Grenfell Mission in Labrador, did adopt this standard.
But Michael says that Ms. Little’s statement is too general, and the question of the optimum number of holes to be filled (that is the interval between loops that will be adopted) depends on a number of factors, such as the width of the cloth or yarn to be hooked or tufted, and the weave structure.
For example, he says, “…if one tried to stuff a wider strip, say that 1/4 inch or wider, into every hole, one would end up with, what hookers call, a “highly packed” rug, that would be susceptible to puckering and would, probably, curl up around the edges.”
Michael adds: “Among the rugs I have, I do not believe that any of them meets such a standard, nor do I consider this fact to detract from their quality.
“In hooking, the important issue is whether the loops (ed. on the front side) cover the backing so that the backing is not visible. (ed. Remember this sentence when we look at a miniature, tufted piece, below, which shows that this can be achieved, sometimes, with quite large areas on the back seemingly empty).
“I am tempted to say,” Michael continues, “that if one sees a back that is as uniform as Ms. Little proposes, it would be a good indication that the rug is a machine-made manufactured rug.”
In her, “Silk Stocking Mats: Hooked Mats of the Grenfell Mission,” Paula Laverty, reports that the Grenfell supervisors insisted on a “fill every hole standard,” and also, in another kind of uniformity: “straight-line” hooking. ” A few Grenfell designs required departure from straight-line hooking, but Grenfells are famous for their nearly universal, visible use of it
A “Grenfell” Silk Stocking Hooked Mat
About straight line hooking, Michael says, “As to Canadian hooking, especially from the Maritime Province, it is true that many rugs show a one directional hooking, usually from bottom to top. This approach, in my view, can create a very static effect on the surface.”
The Grenfell supervisors would agree about the “static effect” on the surface, but it was part of the standard they wanted to achieve. They wanted to create a very smooth surface of even loops, resembling needlepoint.
Michael said that the “present day successor of the Grenfell enterprise is Red Spruce Rugs (www.redspruce.ca). He was of the opinion that they did not practice straight line hooking.
(ed. I visited their site and found that design seems to determine whether they use straight line hooking or not. If the design is rectilinear straight line hooking seems to be used, but not if the design has curves.)
I kidded Michael that his dislike of “static effects,” potentially, saves him a lot of money, because Grenfells are usually expensive nowadays.
Now that Michael had treated backing materials and designs (and in a very preliminary way), the hooking/tufting tools,
he moved on to a survey of how these tools (they are now all “tufting” tools) developed.
The move from the hook to a needle with a hole in it, was only the beginning of a considerable articulation of a variety of tufting tools.
Next, it was discovered that the use of the needle could be mechanized slightly to make things go even faster. One version looked like this.
This tool is made of two separate pieces of wood (vertical in the image). There is a needle with a hole in it attached to the front of the top part (the right side in the photo above). The two pieces of wood are attached to one another toward the bottom of this photo by the rectangular metal device that slides in a slot in the upper part, permitting the top part to move forward and backward. Another metal strap holds the two wooden pieces together at the front while this movement is occurring.
This move to mechanization is not without ideological potential. Any move away from the simple hook is seen by some to be an impoverishment of the tradition (although Michael said this sort of thing does not worry him, personally). Some sites on the internet, that identify themselves as “hooking” sites, and that sell supplies, offer plain hooks, but not any tools that are even slightly mechanized.
Michael demonstrated the use of this kind of tufting punch.
Tool 2 demo
It was remarkable to watch. He put the needle through the backing and could put loops in as fast as he could move the top half back and forth.
The result looked like this.
Needles can be changed in such a simple machine. And they can be of different sizes (as can the fineness of the backing and working material).
This move to mechanizing the tufting process also makes automatic, the selection of the interval between one tuft and the next. The tufter no longer has to search for and direct the punch to the desired hole. The machine does that as an artifact of the movement of its parts back and forth. And this interval will always be uniform. Michael said that it is usual to have things set so that about every other hole in the backing material is being filled.
Two additional points need to be made with regard to the image above.
First, Michael said, folks new to tufting of the sort that uses a needle with a hole, and that is at least somewhat mechanized, often try to “move ” the tool to make it track in a desired direction. A tufting tool cannot be “moved” to control the path of its tufting. Instead the rug maker must “rotate” the tool around its long axis.
The tool is already “moving” in a path, as it forms one loop after another, but its tracking path is determined by rotation of the tool, not by attempting to physically press it in one direction or another.
Second, a row of loops put in by either hooking or tufting can be pulled right out by pulling on the end of the strand of material forming the loops. (Michael did it repeatedly during his demonstrations.)
To secure a section of tufting one needs to take the end of the strand of material used to form the loops and press it to the back, through the backing material, this time without looping. Now do the same with the other end of this same working strand. It seems odd, but this move, taken on both working ends of a strand of loops, will make the row of hooked/tufted loops in between them secure.
As you have seen, above, the most basic hooking tool looks like this.
It is a small, cylindrical metal rod with a hook on one end and secured in a wooden handle on the other. Michael said it likely evolved from a bent nail.
The basic hooking process is to take a length of the material to be hooked (a narrow strip or an item of yarn) behind the backing material, then push the hook through the backing material from the front and catch it on the strand of material being hooked, and pull it back through the backing material so that a small loop appears on the front side.
In general, the loops should have the same height and be the same distance apart. The top loop in the image above will need to be pulled from the back to reduce it height before the next loop is hooked.
Here is a drawing you saw above, taken from Marla Mallett’s web site, that shows this process from another perspective using a needle with a hole in its end rather than a hook.
Simple tufting tool (no hook; hole in needle), worked from the back
Because they are digital (that is, each loop in a backing hole can be a discrete unit) hooking and tufting are both very unrestrictive rug making processes.
Take, for example, the hooked rug below.
One could make this rug in any sequence of colored loops one wanted. There would be, sometimes, a time advantage in doing colors one at a time, but that is not necessary.
In hooking and tufting, there is no restriction parallel to that faced by a weaver of a pile rug, who must tie all the knots in a given row before moving to the next row above it. The primary reason for this pile weaving restriction is that
one or two strands of weft must be inserted the full width of the rug before the next row of knots can be begun.
Kilim weavers, weaving tapestry, have more flexibility than do pile weavers, since they can build up tapestry unevenly in areas in which the patterning wefts do not continue across the entire piece being woven.
For example, in the drawing above of slit tapestry, the weaver could build up, independently, sections where there is a turning back of patterning weft. Both the top areas of light and dark can be continued independently (although a long vertical slit would be a structural weakness).
Note: Some flat weaves, noticeably sumak, are, in principle, completely digital, and therefore, as flexible as hooking/tufting. It is not the case that all flat weaves are more restrictive than are hooking/tufting.
The hooker has no limitation at all, in this respect, and can, theoretically, fill any hole in the backing material with any color and in any order.
You can get a concrete sense of this flexibility by examining the front and the back of a hooked/tufted rug.
Here, below, are some images of a miniature tufted rug that fits nicely on my scanner, being about 8 inches by 10 inches.
First here is an overall photo of its front, showing its Art Deco design.
Here is a closer view of a quarter detail of this front.
Notice in this close-up that you can see that the “loops,” that form something like “pile,” are not arrayed in terms of the strict grid-like rectilinearity of the backing cloth.
Instead, “rows of tufting seem to move in various directions, even the large “white” area on the left, in the front detail image above, shows no grid-like arrangement of the tufts.
This flexibility becomes even clearer when one examines the back of this piece.
Here, again, is an overall view.
And here is a closer quarter detail of it.
Notice that not only do the tufts move in different directions, but that, despite the front looking entirely “filled in” in all of its areas, there are, in fact, some gaps and blank areas where no tufts were pushed through.
But the central point of this example is to show how flexible tufting can be. Look at the black areas with the triangular shapes in the image above, and see how “messy” the tufter’s progression seems to have been as he/she moved the tufting around to form the triangular designs on the that are neat and look completely filled in on the front.
Here is another, much larger, example, about 4 feet by 6 feet.
This is a rug that has been made using a variety of materials, likely including some from old clothing. I don’t know its actual age, but this is a classic “poor man’s” rug. Notice, even in this overall image how, except for the black bands, its various colors swirl without any real pattern of design devices. The hooking/tufting tool tracked all over the backing in various directions.
The way the maker moved in nearly any direction with any color at any time is even more evident in these closer detail images.
Only the use of black, to make it seem, perhaps, a bit like an oriental rug, is more controlled and regular.
The balance of Michael’s talk was largely devoted to showing, describing, demonstrating (sometimes passing into the audience) the various kinds of tufting tools that have developed since 1840.
You’ve already seen the first two types of tool and Michael showed.
Tool 3 is an all-metal version of the two-piece, largely wooden, Tool 2.
Comment on Tool 3: A popular shuttle hook sold in stores and by catalog, perhaps around 1900. A predecessor version appeared first in 1870, sold by Ebenezer Ross, under the name “Novelty Rug Machine.” These shuttle hooks were a significant step forward in rug making tools. They allowed the rug maker to set the height of the loop and the tool design allowed the tool to advance at a uniform paces across the rug surface, unlike the original rug hooking tool which required the maker to control both loop height and spacing of the loop. Hundreds of version of this tool eventually entered the market.
The next tool was the one below:
Comment on Tool 4: A shuttle hook apparently meant to appeal to a rug maker by the addition of extraneous stenciled motifs. It is a heavy tool and its weight detracts from its function. Perhaps from about 1890.
The fifth tool looked like this:
Comment on Tool 5: A curious version of the shuttle hook, probably around 1890. In use, it is an awkward tool to control, as it tends to wobble from side to side.
The next tool was not a hook or punch, but rather a cloth cutter, used to create strips of hook/tufting material.
You can see that it clamps onto a table or bench. Here is the other side of the tool above.
Tool 6, other side
Comment on Tool 6: This tool appeared around the turn of the 20th century, and greatly increased the speed at which hooked and tufted rugs could be made, . It can cut 3 or 4 uniform narrow cloth strips basically as quickly as the user can turn the handle. It replaced the tedious process of tearing or cutting cloth pieces with a scissors.
The following tool looked like this:
Comment on Tool 7: Yet another version of the shuttle hook sold by mail order catalog. Probably 1890 and later.
The eighth type tool, with a kind of crank, to accelerate its back and forth movement, was provided in the two versions below:
Comment on Tool 8: The one on the left is a Perfection “Tufting and Embroidery Machine.” The yarn used in the tool is threaded through the hollow brass tube, which has an eye at the tip. The Star rug machine next to it has the same set up. These tools appeared in 1880 and after.
The next tools, below, are variations on Tool 8.
Comment on Tool 9: The manufacturer of the tool on the right offered the option of several needle sizes.
Here is the next tool Michael had:
Comment on Tool 10: This is the rug needle manufactured by the Columbia Minerva company and also sold as a Lees and presently as a Craftsman needle. It appeared sometime after 1900. It consists of a hollow tube with notched flanges that allow the user to set a prescribed height of loop. The tube fits into the Bakelite handle. Up to 4 needle sizes were available. It is necessary for the user to control the spacing of the loops. This tool is still made and sold.
Here is the next tool shown:
Comments on Tool 11: An example of a Rug Crafters “Speed Tufting Tool ” from about the early 1970′s, when crafts enjoyed something of a resurgence.
It is in the shuttle family as the user holds the wooden handle in one hand and pushes the metal flange back and forth with the other hand.
This tool was sold in kits and the manufacturer also sold yarn and backings. For a brief period, Rug Crafters also had small stores in shopping malls. This tool was promoted for use as a family activity.
This next tool Michael had brought was this one:
Comment on Tool 12: This Danish tool was sold in fairly large numbers in the 1960′s and 1970′s through Better Homes and Gardens Magazine. A crank type or “egg beater” type, as it was commonly known, it had the advantage of a reversing cam that allowed the user to reverse direction by simply changing the direction in which the handle was turned. A drawback was the narrow diameter of the needle which limited the size yarn that could be used.
The thirteenth tool Michael had was this one:
Comment on Tool 13: The Rumpelstiltskin Hand Needle from about 1973. It contains features that drew on prior tools.
It is a crank type tool with a hollow needle and flat “tongue” that passes back and forth inside the needle. The tool has 5 settings for loop heights. The “foot” on the end of the tool which travels across the surface of the backing material allows for the loops to advance at a uniform rate, as well.
Like other crank type tools it can be operated very quickly.
The next tool featured a move to electrical power and looked like this:
Other side of tool above.
Tool 14, other side
Comments on Tool 14: This is the electric version of the Tool 13. It operates at a high speed, and was often used by artisan rug makers to do commission work on a fairly large scale. It is no longer manufactured.
The last tool Michael had brought was a dramatically large, commercial tufting punch. It looked like a large electric drill or a small jackhammer.
You can see that Michael enjoys a tool he admires.
This heavy tufting gun was originally made for use in commercial carpet factories, to repair defects or “skips” in carpet runs. Sometime in the 1980′s it crossed into use by artisan rug makers who valued its speed and maneuverability.
It comes in different versions, electric, electric/pneumatic and completely pneumatic. Most are manufactured in China and Thailand.
When I first became interested in this tool, I could only find versions sold through a limited number of companies in the U.S., and the asking prices were about $2,000 and up.
One day, browsing the internet, I came across a Chinese site that displayed the same photo of a tufting gun as used by a U.S. seller. I contacted the Chinese factory and discovered that they would sell it to me for $140.
Michael had also brought in several hooked/tufted rugs. They were arrayed in two levels on the front board. Here is the first level.
These are not rugs that Michael has made himself. They are types that he brought to illustrate particular variations in hooked/tufted rugs.
Let’s look at these rugs, more closely, each in turn.
The first rug treated was the one below.
Notice that there are subtle designs in the gray field area that look a bit like calligraphy. The woman who made this rug was very skilled, as she used very narrow wool strips that tend to break easily as the rug is hooked. Given the uniformity of the width of the strips, it seems all but certain that the maker used a cloth cutting machine. I would venture to say that the floral design came from a commercial source, or was copied from a design,
but the rug is elevated from the ordinary by the artistic use of the “calligraphic” motif in the field and the free form hooking in different colors in the border. Probably made in the 1920′s or 1930′s, and apparently kept stored away most of its existence.
The second rug Michael had brought in was this one:
Comment on Rug 2: A courser and less accomplished rug than the previous example, but in that sense, more representative of the majority of hooked rugs. The cloth strips are fairly wide and as a result, this rug could have been hooked quite rapidly.
The design closely resembles a “log cabin” pattern commonly found in U.S. quilts.
It is hooked on a burlap backing that exhibits a fair amount of fraying along the edges. There are also loose strips. These defects indicate that this rug was actually used on a floor for some time. Perhaps made in the 1930′ or 1940′s.
The next rug shown was this one:
Comment on Rug 3: Though at first glance, this appears to be a commonly used commercial pattern, a closer examination shows that the design was developed with a great deal of freedom. For instance, the flowers in the field are all different, and the scallop or shell motifs on the ends are of different colors and slightly different pattern.
This rug also shows signs of use, as its burlap backing is unraveled at the ends.
Probably from the 1920′s.
Rug 4 was the one below:
Comments on Rug 4: This rug was found in a box wrapped in a newspaper dated 1905, and apparently was packed away for most of its life, as its colors are still quite vibrant.
This rug was probably made on a commercially stenciled burlap backing, given the overall uniformity of the design.
It is the colors that make this rug attractive.
The last rug on the first level of the front board was this one.
Comment on Rug 5: Just a fun rug. Its abstract, “modernist” design suggests that it was probably made on a commercially stenciled burlap backing, sold in the 1950′s.
Mostly wool strips, with some cotton strips
The rugs on the second level of the front board were pieces that Michael had made himself.
Here is the first of these.
A little closer look at it.
Comments on Rug 6: This is a rug made with hand dyed, wool yarns, on a cotton monk’s cloth backing, the ends unraveled and knotted to form a fringe.
The “tree of life” design is drawn from a computer circuit board.
This rug was made with a Rumpelstiltskin Hand Needle, and a Wilson Brothers shuttle hook.
Rug 7 was this one:
Comment on Rug 7: A simple geometric pattern of the type that succeeds or fails with the color choice.
Made with hand dyed yarn on cotton monk’s cloth backing, using a Wilson Brothers shuttle hook.
The next rug was this one:
Comment on Rug 8: A fairly large rug, about 3′ x 5′.
It is made with plied weaving yarns, wool and cotton and synthetic blends on a commercial polypropylene rug backing.
The color blocks are wool strips added with a shuttle hook.
Because it is a simple design, I was able to make it with a Rumpelstiltskin Hand Needle.
The last of the rugs Michael had brought in was this one:
Comment: This is a rug made with a variety of tools and materials. Most of it is hand dyed wool yarns. Interspersed in the stripes are shuttle-hooked cloth strips, looped wool and cotton yarns added with a Rumpelstiltskin Hand Needle, and tufted wool yarn added with an electric tufting gun.
This rug was sheared at the end, with an electric sheep shears, to give a more uniform pile.
It was made on a German rug backing of unknown content.
The discussion and questions had been vigorous as Michael went along, but he answered the remaining ones
and adjourned his session.
Folks in the audience moved to the front.
My thanks to Michael for this interesting program, and for working with me after to fashion this virtual version of it. Thanks, too, to Wendel Swan who took the useful close-up photos of the tools afterward.
I want to testify, here at the end, that I like hooked/tufted rugs, myself, and can demonstrate that with one that just came in the mail to me as I was typing this sentence.
I hope you have enjoyed this virtual version of Michael Heilman’s exploration of hooking, tufting and the development of the associated tools over the years.
R. John Howe