The Eternal Renewal of Hand Weaving
(Published in The Journal of Modern Craft, Vol. 11, Issue 2, 2018: https://doi.org/10.1080/17496772.2018.1493794)
In late 19th and early 20th century, hand weaving was thought to become obsolete. Power looms – machines that had become increasingly automated, and ubiquitous in factories across the United States – would now, it seemed, replace all human labor. As a result, the creativity and knowledge of weaving would disappear. The worker, as the tapestry weaver and Arts and Crafts leader William Morris observed, would be no more than a “tender of machines.”[1] Over the last century, however, knowledge of weaving has not waned, nor has creativity declined. Instead, both have thrived, and even surged, as a result of advancements in technology. Three case studies evince this point.
After graduating from the Chicago Art Institute School of Design and studying in Paris, the American weaver Mary Meigs Atwater began to teach and write about weaving in the late 1910s. She was particularly fond of coverlets – bedspreads or table covers with repeating visual motifs or patterns. Two of these patterns are found in Virginia Beauty (1825) (Fig. 1), one in the center (a variation of a motif called “bird’s nest”) and another along the border (known as “pine tree”). Both stylized patterns are characteristic of a geometric style in coverlet weaving that typically involves two colors of contrasting value that help maximize the figure/ground relationship and enhance the visibility of the distinct motifs.
Coverlets were brought to the United States by colonists in the 17th and 18th centuries, and the exchange of their patterns – through instructions called drafts – was a means of passing along traditions in weaving.[2] In journal entries, Atwater writes of the difficulty of finding coverlet drafts. She describes research trips to the Philadelphia Museum of Art where she would dig through the museum’s archives, as if on a hunt for old treasure maps (instructions for coverlet making indeed resemble cryptic charts). One senses that Atwater feared these patterns had met the same fate as dinosaurs and that instructions to their design were like fossils – visual “footprints” that must be preserved.[3]
While the study of the past is one thing, reviving a bygone practice so as to maintain a “real American national art” is quite another.[4] In her efforts to teach coverlet weaving, from the late 1910s until the 1940s, Atwater could be construed as conservative. It was as if she rejected all innovation, such as that arriving in power loom technology at the time, which alleviated much of the strenuous physical labor of hand weaving. Yet in a close analysis of Atwater’s writing, one discovers that she was also after something else – something she understood as vital to a weaver’s knowledge.
Around the time that Atwater began to teach weaving, she noted, “handweavers were handicapped in their work…because few understood the art of draft writing.”[5] In the late 19th and early 20th century, knowledge of draft writing had dwindled due to the growth of factories, the spread of power looms, and the limitations in pattern design resulting from the use of new technology.[6] (Mechanically more complex looms, such as Jacquard looms, were still used for extensive pattern or figurative work in weaving; however, they were largely the province of textile factories.) Weaving on a shaft or harness loom was the most practical and affordable option for those who wove at home or in a small cottage-industry setting.
In one of the three main sections of her Shuttle-Craft correspondence course, Atwater focused on teaching weavers how to read and write drafts. A harness loom draft consists of four parts, one of which functions similar to a map or blueprint in its schematic representation of the basic visual form to be designed. The other three parts of the draft act as channels of communication, which give weavers information about treadling, threading, and the tie-up of threads. Figures 2 and 3 are examples of drafts made by contemporary weaver-artists Ted Hallman and Samantha Bittman. Much like the language of HTML, which translates the visual layout of images and text in webpage design, the “input” in a weaving draft acts as code for the creation of any pattern or pictorial design in woven fabric.[7] If a weaver can read a draft, s/he can recreate patterns others have designed (and drafted). In learning how to write a draft from scratch, a weaver’s design vocabulary becomes immediately personalized and potentially infinite.
It’s clear that Atwater perceived the draft as a locus of agency – a way to help weavers maximize their artistic expression. In the correspondence course, she advised students to understand draft writing as “the most important part of the study of weaving…to those who aspire to do original work in weaving.” In the final lesson of the course, she required students to complete a thesis, designing a bedroom set “either in an original pattern or in an adaption of an old one,” pushing them to not just follow someone else’s pattern but devise their own.[8] Thus, for Atwater, teaching coverlet weaving was far more than a preservationist reflex. Though she indeed passed along historic patterns through these lessons, she also shared critical knowledge of how to read and write in weaving, thus, breaking weavers’ rote copying habits and empowering them to create anew.
While Atwater had mostly taught weaving as a practice separate from large-scale manufacturing, the German émigré weaver Anni Albers encouraged hand weavers to see their designs as prototypes, intended for mechanical reproduction on a power loom. Albers, who had studied at the Bauhaus and taught at Black Mountain College in North Carolina between 1933 and 1949, was very much part of a new generation of weavers trained between the late 1920s and 40s, who sought to define standards of quality in mass production through careful, well-planned development of prototypes. During this time, many textile factories acquired power looms and began to use new synthetic fibers, allowing them to produce items rapidly at a more affordable price.[9] However, most manufacturers were ill-equipped to experiment with synthetic fibers, and unwilling to invest in tests that would improve a material’s overall function.[10] While some companies recognized the potential of hand weavers to intervene in the industrial process, others proceeded with what Albers characterized as a “neurotic aimlessness.”[11]
Throughout her writing in the 1940s, Albers criticized the quality and production of American textiles. Though she remained diplomatic, never naming the United States in particular, her immediate surroundings and direct engagement with American textile manufacturers suggested that she had her adoptive country’s conditions very much in mind. In “Designing,” an article Albers wrote for Craft Horizons magazine in 1943, she lamented the textile industry’s tendency to focus on decoration while ignoring structure and function. She saw the “trick of hiding a poor material under a rich pattern” as a facile means of designing, which lowered the quality of objects (similar to what happens today in “fast fashion”).[12]
Outside the factory setting, Albers also took aim at hand weavers, whom she viewed as equally lethargic in their thinking because they relied on “recipes” and “traditional formulas.” In what is perhaps her most oft-cited essay, “Handweaving Today,” she boldly stated that weaving “has degenerated.”[13] This concisely worded article was intended to provoke, and serve as a wake-up call to the textile community. Albers argued that through thoughtful planning on a hand loom, a weaver could ensure a more structurally and texturally sound material for mass production, thereby putting into place a check-and-balance system for the textile industry, which was susceptible to unrestrained automation.
Other weavers, such as Marianne Strengell and Dorothy Liebes, also advocated for (and taught) hand weaving as a means of designing prototypes around the mid-20th century. Like Albers, they encouraged the use of simple structures such as plain, twill, and basket weave, which could be made easily on a four-harness loom. Whereas more complex weave structures and more complicated looms allowed a weaver to create intricate visual motifs (as described above in the discussion of coverlets), a simple weave structure on a harness loom permitted a more direct study of a single fiber’s structural integrity and its textural relationship to other fibers. Certainly, while an understanding of texture and structure is inherent to all types of weaving, a specific focus on these elements was particularly relevant around midcentury when the development and production of fully synthetic fibers took off.
In a drapery fabric, made for Philip Johnson’s Rockefeller Guest House in 1944, Albers demonstrated how weaving is a process of sculpting with texture. To create the fabric (shown in a detail in Fig. 4), Albers used a plain weave, which allowed her to test the strength of fibers she used and exploit their materiality.[14] What makes the Rockefeller Guest House drapery striking are the shiny copper metallic wefts, which seem to undulate like waves as they reflect light moving across the fabric’s surface. To “foreground” them, Albers off-set their radiance with cream-colored fibers – one of plastic and another of chenille. The soft chenille is optimal for touch (as the drapery likely had a utilitarian function). However, Albers didn’t over use it; rather, she varied it with a tape-like fiber, which helped her to create a flatter, less dense fabric that would not cover up or block the copper metallic fiber’s extraordinary shine. In this way, she built a textile that exploits the materiality of its structural form (unlike what happens in traditional tapestry weaving and embroidery).
But Albers didn’t stop there. Though the interlocked, glinting stripes of metallic move rhythmically through the drapery, there is no sequential logic as to how they were inserted. In other words, to design this fabric, Albers improvised, making decisions about weft insertion spontaneously and to her own liking. Thus, while defining her approach to hand weaving as a means of prototyping, Albers also continued to exert her own creative agency. In place of a pictorial image, she emphasized textural sculpting; a deep knowledge of materiality offered as many opportunities for expression as coverlet weaving ever had, only this time, in the midst of a second-wave industrial revolution.
At midcentury, Albers and other weaver-designers in the United States primarily taught on harness looms. This was due in part to their interest and training in designing for machine reproduction (which most often took place on automated harness looms). Another reason pertained to the types of objects they designed – utilitarian fabrics such as upholstery, drapery, and carpets. At the time, the vast majority of utilitarian fabrics were conducive to harness loom weaving (sometimes simply called “cloth weaving”). However, plenty of fabrics with intricate pictorial or figurative imagery continued to exist: most of them were/are known as “Jacquard” – the type of loom used to make them.
Jacquard looms are typically associated with high-end, luxury production – not because they weave better materials, but because they are more complicated to set up and run. Unlike a harness loom, which lifts a predetermined group of warp threads at the same time, a Jacquard loom is capable of lifting a single warp thread at once or groups of threads in different sequences, unique to the particular image or pattern desired in the woven fabric. Over the late 19th and 20th century, Jacquard looms, like harness and dobby looms, became fully automatic.[15] Most recently, they have been digitally operated, allowing the weaver to transfer an image from a software program such as Adobe Photoshop. (This is not the same as printing color on pre-woven fabrics, which has also become more popular today.) The power loom and digital revolution that has affected Jacquard weaving has allowed the weaver to easily create intricate details and convincing illusionism without ever having to vary materials. Today, many design companies do not work with fibers at all, opting instead to send images to other countries, like China, where the fabrics are woven and then exported.
In the early 1990s, the Norwegian weaver and inventor, Vibeke Vestby looked to depart from such practices by adapting the digital, automated Jacquard for hand weaving. In doing so, she developed the Thread Controller loom (referred to by its acronyms TC1 or TC2, indicating the two different models available). Like a Jacquard power loom, the TC loom picks up either individual or multiple threads in a sequence, which can be reprogrammed anytime in the process of weaving. The fundamental difference is that the weaver of the TC loom has complete manual control of weft insertion and packing, the building of the textile. Because the TC loom functions like the digital Jacquard, it stores the weaver’s desired image or pattern, so that s/he is free from having to operate the highly technical process of opening the shed (i.e., pulling up certain warp threads every single time weft is inserted). The advantage is this: the weaver of the TC loom can concentrate entirely on sculpting the textile, while allowing the loom to control the “drawing” of the predetermined form.
In 2014, the Brooklyn-based weaver Robin Kang wove Chipset Artifact on a TC2 (Fig. 5). The textile portrays a graphic image of a microchip or circuit board – a system of intercrossing welded wires moving information from one place to another. A microchip resembles the technology of the loom, which manipulates the movement of threads, translating warp-weft codes into pictures. The uneven, knotty fringes on the left and right sides of Chipset Artifact are like bundles of unbound wire, rendering vivid the metaphor of weaving as a practice in trafficking data.
Last summer, I spoke to Kang about the advantages of weaving with the TC loom.[16] Having control of the weft, she can experiment more freely with fibers or other material, inserting elements as she sees fit into the skeletal layers of the warp and making organic decisions about how to build the textile. Kang’s ability to manipulate the sculptural and structural aspects of the textile through decisions related to weft placement is apparent in Chipset Artifact (Fig. 6). In several areas, she has left weft threads tangled within the gridlocked structure of the weave; in others, she has pulled them entirely free so that they hang loose, freed from their structural obligation. These “punctuated” marks on the screen-like surface function similarly to an actor who looks in the camera: reflexively, they point to the constructedness of the medium, the human hand at work on an artifact in an otherwise highly automated process of object-illusion making.
In her description of the TC loom, Vestby writes, “What I love about the TC2 is that it is a simple tool to operate. It turns the loom into a sketchbook, where ideas may be tested out one after the other.”[17] Vestby’s metaphor conjures up notions of weaving as unbound, an activity without exact beginning or end; the result is not predetermined, and indeed, not fixed until the artist stops. Sketching tends to be associated with the preparatory stage of idea-making in the design of an object, as in Albers’ creation of prototypes. However, the TC loom functions both inside and outside of this logic: textiles produced on it are not necessarily intended for machine reproduction, though certainly could be translated to a digital, automated Jacquard loom.
In the past century, as my three case studies make clear in their different ways, while automation has replaced certain tasks in weaving, it has strengthened others. Early American power looms wove cloth faster, but the products made on them were often deficient in visual complexity – which Atwater sought to remedy by teaching the skill of drafting. Midcentury power looms produced in mass quantity, but they lacked haptic intelligence, or sculpting with texture, which Albers sought to revive. Today, weaving has again been reinvented, on a continuum of processes between the automated loom and hand weaver, each complementing the other. As a result, the weaver’s potential for creative agency has arguably expanded. Interestingly, it is also with the most cutting-edge technology and automation of the TC loom that weaving’s oldest traditions – form and image making – have re-emerged, in a process that in many ways has also redefined the craft practice of weaving. Ultimately, a textile remains only as “good” as the tool with which it is made and the weaver who knows how to use it, or reinvent it – a cycle which eternally renews the practice of hand weaving.
[1] William Morris, “The Revival of Handicraft,” in The Craft Reader, ed. Glenn Adamson (New York: Berg, 2010), 150.
[2] Mary Meigs Atwater, The Shuttle-Craft Book on American Hand-Weaving (Worcestershire, UK: Read Books Ltd, 2013), 6-8.
[3] Atwater cited in Weaving a Life: The Story of Mary Meigs Atwater, ed. Veronica Patterson, comp. Mary Jo Reiter (Loveland: Interweave Press, 1992), 141.
[4] Mary Meigs Atwater, Shuttle-Craft Courses in Weaving (Anaconda: Standard Publishing Co. 1922), 4.
[5] Atwater quoted in Weaving a Life: The Story of Mary Meigs Atwater, 141.
[6] See for example Mary Meigs Atwater’s discussion in The Shuttle-Craft Book on American Hand-Weaving (Worcestershire, UK: Read Books Ltd, 2013), 26-27; Carroll Pursell’s, The Machine in America: A Social History of Technology (Baltimore: John Hopkins University Press, 2007), 39-45; Alice K. Waagen, “An Historical Survey and Analysis of American Handweaving” (Ph.D. diss, Pennsylvania State University, 1982), 32-33.
[7] Weaving drafts have also been compared to music notation or a musical score, most recently in T’ai Smith’s Bauhaus Weaving Theory: From Feminine Craft to Mode of Design (Minneapolis: Regents of the University of Minnesota, 2014), 149.
[8] Atwater, Shuttle-Craft Courses in Weaving, 9-11.
[9] William Mass’s Technological Change and Industrial Relations: The Diffusion of Automatic Weaving in the United States and Britain, The Journal of Economic History, Vol. 45, No. 2 (June., 1985); 458-460, Cambridge University Press. Synthetic fibers were first developed by Wallace Carothers in the early 1930s at DuPont.
[10] See letter to Dorothy Liebes from Ennis P. Whitley of The Doebeckmun Company, Dorothy Liebes papers, Dobeckmun Co. files, Archives of American Art, Smithsonian Institution.
[11] Anni Albers, “Designing” (May 1943) in Anni Albers: Selected Writings on Design, ed. Brenda Danilowitz (Middletown: Wesleyan University Press, 2000), 21.
[12] Ibid.
[13] Anni Albers, “Handweaving Today: Textile Work at Black Mountain College,” The Weaver 6 no. 1 ed. (January/February 1940), 3.
[14] A microscopic examination of the fabric shows that some of the cream-colored plastic fibers have fractured, leaving tiny “split ends” throughout the drapery; yet, they remain surprisingly intact because of the plain weave Albers used. It’s impossible to say when and how these fibers began to split (Albers could not have predicted their longevity at the time), but this is exactly the kind of knowledge Albers saw as vital and would have justified her approach to using hand weaving as a means of research.
[15] James Essinger, Jacquard’s Web: How a Hand-Loom Led to the Birth of the Information Age (Oxford, UK: Oxford University Press, 2007).
[16] Robin Kang in an interview with the author, summer 2017.
[17] Vibeke Vestby in an email to the author, 1 March 2017.