There are several challenges in assessing iconographic evidence of utilitarian implements and the contexts of their use. One is recognizing the difference between a representation of an object or process that may be stylized but can otherwise be corroborated, and an imagined depiction that coincidentally appears to be plausible. This difficulty is compounded when an image includes details that can be identified with a fair degree of confidence, in proximity to others that are more likely to be misrepresented. There is also a contextual aspect to this. If graphic evidence of a tool used for handicraft appears in seemingly realistic detail at a completely unexpected time or place, particular care is needed before basing revolutionary conclusions on it.
A range of such considerations attaches to early illustrations of the production of looped fabric. Current reviews of the history of knitting are commonly illustrated with portraits of so-called Knitting Madonnas. Perhaps the best known, and certainly the most clearly detailed, is in a scene on the Buxtehude Altar painted by Master Bertram ca. 1400. It unequivocally depicts four double-pointed knitting needles used for working a garment with what is now termed a seamless yoke construction.
This design matches that of the robe worn by Jesus on the cross described in John 19:23: “This garment was seamless, woven in one piece from top to bottom.” The precise wording varies among translations and I have cited the one at the top of the listing here. The accompanying vocabulary provides this definition:
woven
ὑφαντὸς (hyphantos)
Adjective – Nominative Masculine Singular
Strong’s Greek 5307: Woven. From huphaino to weave; woven, i.e. knitted.
The first translation of the New Testament directly from Greek into German was published by Martin Luther in 1522. He translated ὑφαντὸς as gewirkt (the adjectival form of wirken — a cognate of the English “to work”). This word has figured time and again in posts here dealing with German texts where it designates looped fabric in explicit contrast to woven. Luther states equally clearly that the robe (Rock) was made “unseamed, looped from the top through and through” (…vngenehet von oben an gewirckt durch vnd durch…). In contemporary text, wirken is frequently translated as warp knitting but was used historically to designate tapestry and a number of production techniques for non-woven fabric.
Earlier bibles in vernacular German were translated from the Latin Vulgate Bible, in turn translated from the Greek in the 4th century. I haven’t seen any of the intervening versions but Luther’s translation could as easily have been made directly from the Vulgate, “…inconsutilis, desuper contexta per totum….” The pivotal “contexta” is commonly translated as “woven.” However, it does not specifically designate work on a loom but rather joining in the general sense of intertwining. There is no question otherwise about the description of the garment as seamless and worked in one piece starting at the top.
Religious commentaries that began to appear in Germany at the end of the 12th century are consistent with Luther’s wording and add that Mary made a robe of the same design for Jesus as a child. At least one account states that he wore it for his entire life and that it grew with him to match his body size. This leads to a reasonable expectation of Master Bertram having depicted the child-sized version of that garment. However, he shows it being worked from bottom to top, rather than top down as explicitly noted in the gospel verse.
This typifies the questions attaching to iconographic analysis that introduce this post. The array of needles is represented with enough accuracy to make it unlikely that they were painted from memory or imagination. Among the things the artist might have determined in preparatory consultation with a knitter is that, if shown being worked downward from the top, the garment would also need to be portrayed upside down. If it is safe to assume that this was artistically (and perhaps exegetically) unacceptable, the obvious next question would be if it could be knitted upward from the bottom.
One answer to it may conceivably have been, “There is, but you won’t find a knitter who knows how to do it that way anywhere nearby.” Mary’s hands are not depicted in a plausible working position and the yarn is wrapped around the needles beyond where they intersect. It therefore seems unlikely that anybody directly familiar with knitting technique assisted in arranging the scene. The grand question then becomes whether a set of knitting needles had been lashed to the top of a finished garment by someone uncertain how to do so realistically, or the actual explanation involves more extensive artistic license.
Although not normally noted in discussions of the Knitting Madonnas, such images are only one facet of a broader genre showing Mary engaged in a range of crafts. Embroidery, spinning, and tablet weaving are the most common, and knitting is not the only represented type of loopcraft. Regina Flury-Bültzingslöwen, whom we met in an earlier post about the distinction between true knitting and other forms of looping, discusses the broader scope of such illustrations in an article titled Maria mit Handarbeiten (Mary with Handwork), published in 1957.
In it, she compares the preceding garment with another of the same design but unquestionably worked top down, noting that these are the only known portrayals of Mary making the ostensible Holy Robe in a complete enough state to support its identification. The second of them is seen in an engraving by Veit Stoss, from ca. 1480.
If the tool in Mary’s right hand were an eyed needle rather than a bobbin, this would be taken as an illustration of medieval nalbinding. That method of fabric production was used for liturgical gloves at the time and would have been known to Mary if she indeed did practice the looped arts. An epic poem in Middle High German about the life of Mary, freely translated and adapted by Phillip von Seitz ca. 1340 from an earlier Latin text relates, “…the robe was made as one loops (wirket) gloves and caps that one wears on the head, not seamed (geneit) and yet made from yarn…”.
If a garment of this design is knitted bottom up with twisted stitches and compared side-by-side with one made top down as nalbound cross-knit looping, the structure of the fabric and the orientation of the stitches in it will be identical in both. In principle, the same would be true if the looping were worked with the bobbin seen in the engraving. The problem is that when judged in proportion to the hand, its aggregate diameter with the wrapped thread is not amenable to producing fabric with closely worked small stitches. Such implements (albeit wrapped differently) are more commonly associated with netting. This has a far more open structure than the illustrated garment likely would have had, and might otherwise be expected to have been captured in the engraving.
This concern would be reduced by seeing the bobbin as a stylized representation of the tool in the following two illustrations. This is a specialized form of bobbin with a thin cylindrical shaft and a small fork at each end, called a “netting needle.” The thread is wrapped lengthwise around the shaft and held in place by the forks. Such tools have been recovered from archeological sites that far predate the entire period under discussion, as this exemplar from Egypt in the collections of the Metropolitan Museum of Art, dated to 1550–1458 BCE.
On some needles, the forks are positioned at a right angle to each other. When looking toward the open face of one, the other is seen from its side. This form appears to be illustrated in a painting from ca. 1460 where Mary is embroidering in the company of three women engaged in other activities. The reproduction in the Flury article is of limited quality but this detail taken from it clearly shows the woman at Mary’s right side to be netting with the full range of additional implements used for the decorative form of that craft (just as they were to appear in countless 19th-century fancywork instructions).
The needle is in her left hand and the implement in her right hand is the “mesh” used to set the gauge of the netting. As with the image on the Buxtehude Altar, there are flaws in the posing and finer detail, of which the most relevant is the thread being wrapped similarly incorrectly around the needle. Such tools are still produced, as shown here in the same 20 cm length as the one made over three millennia earlier, and correlating to a knitting needle of ≈4 mm diameter.
Smaller netting needles were used as alternatives to tatting shuttles during the 1850s. Although viable from the technical perspective, they were ultimately abandoned because they could not be tucked into a pocket as conveniently and safely as shuttles could. The possibility of the Stoss engraving reflecting the use of a netting needle for plain or cross-knit looping might be worth similar empirical assessment before concluding that he had conflated tools and crafts. (There is also a potential can of nomenclatural worms here.)
The thread does appear to be wrapped around the illustrated bobbin clearly enough to contraindicate the notion of it being a stylized representation of a netting needle. There is no such question about the tool shown in the earlier painting, despite the incorrect wrap. If it in any way served as a model for Stoss (as might also be indicated by the floor stands in both) this flaw could easily have been carried into it.
lovely read cary!
Thankee kindly!
Nice topic and writing. It would be interesting to dive deeper into the Greek verb used instead of the subsequent translations. I do not have my library at hand to do that research, but I estimate that Greek etimology would give more insight into the nature of the textile. The verb “hufainoo” is also used for the making of a spider’s web or the invention of a trap or trick. the “huf”-part suggests that something goes down under something else in the process, which could suggest loom movements.
Thank you for the appreciative words. It’s quite interesting that the Greek hufainoo designates the making of a spider’s web. Luther also uses wirken to designate what a spider does with its hands, in his translation of Proverbs 30:28. (The most common English translation of the original Hebrew is grasp.)
There are several competing claims of the robe Jesus wore on the cross having survived as a relic. To the best of my knowledge all are woven. I similarly believe that the descriptions of Mary making this garment for the child Jesus are all German, as are the only two known illustrations of her crafting it. Both unequivocally show it being looped. This does suggest that Luther’s reading of the Greek designates that mode of fabric production, whatever additional senses and connotations the original term may have had.
Yes, woven, and one such is understood to be buried. It is also understood that it was woven to shape on a warp-weighted loom (Chrysostom and Theophylact of Ochrid are two authors from different periods); also, such tunics have been found, primarily in Egypt. H. Granger-Taylor and O. Shamir discuss examples, if I remember correctly. I have been wondering about possible depictions of the Virgin doing loop-crafts (most Byzantine iconography depicts spinning, but perhaps some examples do not; descriptions are vague at times). It seems that you have uncovered a literary description of needlebinding (I have been combing everything for plausible examples). Thank you for this illuminating post!
Philipp von Seitz’s Latin source was the thirteenth-century Vita Beate Virginis Marie et Salvvatoris Rhythmica. Here is a literal translation of what appear to be the lines that he had in mind: The Mother of Jesus had made, of textile work, A tunic, which she wove by means of an art so fine, That in the whole tunic, not any seam Was there, and moreover, in its mending not even a rent was made. While this passage may strike some readers as a description of something miraculous, it stands within a long tradition of making a contrast between patchwork and integral fabric constructions, whether woven, knotted, looped, or felted. It seems that von Seitz, who may not have been aware of the process of weaving to shape, or even aware that a woven cloth would require patching, glossed this passage with the explanation of the looping technique. Looped garments would allow darning or restoration if in need of repair, dispensing with a need for patches and doing away with torn edges by employment of the initial, or a complementary, construction method, rather than an intruding seam. Some of the earliest material that may have relied on the audience’s knowledge of… Read more »
Various sources describe the tunic as “opere reticulato” (“network”). The earliest I’ve found is from the 12th century (Peter Comestor’s Historia Scholastica). It looks as if at least two describe or qualify the technique somehow, but I’m working out OCR errors in order to read the text. One of these is a passage in Denis the Carthusian, another in Cornelius Lapide’s commentary on John (but not in the English translation of the Great Commentary). It seems that such an idea was interpreted with what we would call netting today, but it may have referred to needlebinding. The art historian Bettina Bildhauer considers knitting to be out of the picture, as discussed here. I have been able to get a few more fragments of text cleaned up. Several sources, the earliest from Nicholas of Lyra, describe the reticulated fabric as “ad modum ciliciorum.” Given the late Medieval provenance, it seems that “hairshirts” is meant, not cloths of goat’s hair in general. I haven’t come across details on extant hairshirts from anywhere at any time, except for that of Birgitta of Sweden, which is looped and is within the same time period. I surmise that more than one technique may have been… Read more »
The Medieval descriptions of non-woven fabric structures and the tools used to produce them certainly deserve all the attention we can muster. The sources you cite suggest quite clearly that I was too quick in dismissing the possibility of the garment illustrated by Stoss having been netted. They also suggest that there may have been less of a hard boundary — conceptually if not in actual practice — between such work and nalbinding than is currently believed. (Perhaps it’s time to start a discussion about replacing the poorly formulated label “knotless netting” with the more appropriate coinage “loose-knot netting.”) The next blog post is scheduled to go online two days from now and is very much a continuation of the present one. Alongside other evidence of early knitting tools, it examines a Knitting Madonna painted in Italy that is a candidate for being the first in the genre. Although it has been described as illustrating four double-pointed knitting needles as all other such paintings do, that description appears to be based more on expectation than on visible evidence. The post discusses the possibility of the tool instead being a single flexible needle. This is reinforced by, and reinforces, the reading… Read more »