The Forgotten History of the Knitting Machine: How One 1589 Invention Industrialized Cloth

The cotton gin and the spinning jenny appear in every school account of the Industrial Revolution. The stocking frame, which preceded both by two centuries and arguably did more to shape the textile economy of early modern Europe, appears in almost none. The frame was the first machine to mechanically reproduce a complex craft operation that had been performed by hand for centuries, and the framework of issues it raised (patent monopolies, automation displacement, geographic concentration, vertical integration) became the recurring vocabulary of industrial economics for the next four centuries.

The 1589 invention

William Lee was a Cambridge-educated clergyman in Calverton, Nottinghamshire, with a wife who supported the household income through hand-knitting stockings. The story (probably embroidered but at least plausible in outline) is that Lee was annoyed by how much his wife's knitting absorbed her attention and set out to mechanize the process. The result, completed around 1589, was the stocking frame: a wooden machine with a row of hooked needles arranged in a horizontal bar, each needle individually controllable through a system of levers and treadles operated by a seated knitter.

The mechanism reproduced the fundamental knitting operation (forming a loop, drawing the next loop through it, advancing) but performed it across an entire row of needles simultaneously rather than one stitch at a time. A skilled framework knitter could produce a stocking in a few hours where hand-knitting required several days. The machine was elaborate (around 2,000 parts) and required substantial skill to operate, but the productivity gain was a factor of 10 to 50 over hand-knitting.

Lee took the machine to Elizabeth I in 1589 and asked for a patent monopoly. Elizabeth refused. The traditional explanation is that she objected to the displacement of hand-knitters (a substantial cottage industry in late Tudor England), but Joan Thirsk's economic history suggests the refusal was more political: granting a monopoly to a clergyman with no court connections was unusual, and the machine's first products were silk stockings competing with imports that the crown was already trying to discourage through other channels. Lee went to France in 1612 hoping for French royal support, did not get it, and died in Paris in 1614 apparently in poverty.

The slow diffusion

The frame survived Lee. His brother James returned to Nottinghamshire with the design and a few prototypes, and the village of Calverton became the seed of a new industry. By 1660 there were several hundred stocking frames operating in the East Midlands. By 1720 there were 8,000. By 1812 there were 25,000, almost all of them owned by entrepreneurs who rented them to journeymen knitters working in their own cottages on a piece-rate basis.

The diffusion was slow by modern standards (a century to scale from prototype to thousands of units) but the pattern is what is interesting: the machine spread without patent protection because no one could effectively enforce the protection across a decentralized industry of cottage workers, and the lack of protection meant that incremental improvements (wider frames, finer gauges, mechanisms for patterned knitting) appeared from many sources and propagated quickly.

The economic structure that emerged was the framework-knitting trade: London-based merchants supplied the yarn, contracted with frame owners (often called bag hosiers because they delivered raw materials and collected finished goods), who in turn rented frames to journeymen. The journeyman paid frame rent to the owner, paid for the yarn at a wholesale price, and was paid for the finished hosiery at a piece rate that the merchant set. The journeyman bore the risk of bad weather (which damaged the wool), bad eyesight (which ended a career), and bad markets (which left finished goods unsold).

The Luddite period

The economic crisis of 1810-1815 (caused by Napoleonic War-driven trade disruption combined with falling demand and tightening wages) produced the most famous chapter in the frame's history. Journeyman knitters across the East Midlands began breaking frames in coordinated attacks, supposedly under the leadership of a probably-fictional "Ned Ludd". The attacks targeted specific frames owned by merchants who had cut piece rates or who had begun producing inferior cloth (cheap cut-up hosiery rather than the more skilled fully-fashioned hosiery that traditional knitters made).

The Luddite framing in popular memory is anti-technology: workers smashing machines because they did not want progress. The historical reality, well-documented by E.P. Thompson's The Making of the English Working Class, was narrower: the frame had been the workers' technology for two centuries, and the protest was against specific market practices and specific employers, not against the frame itself. The traditional knitters who broke frames in 1812 owned frames themselves; they were targeting the frames that produced the cheap goods undercutting their market, not the frames as a category.

The government response was severe: machine-breaking was made a capital offense in 1812, several leaders were hanged, and the immediate movement was suppressed. The longer-term consequence was that the frame industry consolidated under merchants and the cottage industry slowly disappeared over the following 50 years as the technology centralized into factories.

The Cotton's Patch industrialization

The factory transition was gradual. The first power-driven stocking frame was built by William Cotton in 1864, using the steam-driven shaft drives that the rest of textile manufacturing had adopted decades earlier. The Cotton's Patent machines could knit fully-fashioned hosiery (with shaped panels) at substantially higher rates than hand-operated frames, and they required less skilled operators because the shaping logic was encoded in the cam patterns rather than in operator judgment.

The Cotton's Patent design was the dominant industrial knitting platform from 1864 until about 1965. The post-1965 transition was to circular knitting machines (which produce tubular fabric continuously rather than flat panels that have to be seamed) and to electronically-controlled flat machines (whose patterning is software-controlled rather than cam-controlled). Both transitions were enabled by the same trends: better motors, better sensors, better control systems. The fundamental knitting mechanism (hooked needles forming loops in sequence) has not changed since 1589, but the elaboration around it has gone through three major reorganizations.

The persistence of the technology

The knitting industry in 2026 produces about 50% of all cloth manufactured globally, the other 50% being woven cloth. The split has been roughly stable for a century: knitting suits stretchy and form-fitting garments (socks, sweaters, t-shirts, athletic wear), weaving suits structured and shape-stable garments (shirts, suits, denim, formal wear). The knitting half is dominated by machines whose lineage descends from the Cotton's Patent design and ultimately from Lee's 1589 frame.

The geographic concentration is interesting. The East Midlands of England (where Lee's machine originated) lost the industry through the 20th century to lower-cost producers; the dominant 2026 producers are China, Vietnam, Bangladesh, Turkey, and India, with smaller specialist producers in Italy and Germany supplying high-end markets. The labor economics that drove the consolidation (low piece-rate labor in cottage industries, then in factories, then in international supply chains) is exactly the pattern that the original framework knitters were protesting in 1812.

Three observations

First: foundational manufacturing technologies often appear two centuries before the conditions for industrial scaling are in place. The stocking frame existed in 1589 but the steam engine and the factory system that allowed it to be deployed at scale did not arrive until 1820-1860. The intervening 250 years were spent in the cottage-industry phase, which produced significant economic activity but did not look like industrialization. The general pattern (long latency between invention and industrial deployment) recurs in printing, in lockmaking, in many areas of chemistry.

Second: the patent question matters less than usually assumed. Lee's failure to secure a patent in 1589 did not prevent the technology from spreading; it just changed who captured the economic value. The frame industry that grew up in the absence of patent protection was substantially more decentralized and slower-moving than a patent-protected industry would have been, but it was also more resilient to single-firm failures and more responsive to incremental improvements from many directions. The current debate about software patents has structural parallels.

Third: the geographic concentration of industries tracks the institutional infrastructure (skilled labor, supplier networks, knowledge transmission, finance) more than the underlying technology. The East Midlands made knitting machines for 350 years because the infrastructure was there, not because the technology was particularly suited to that location. When the infrastructure shifted to East Asia in the late 20th century, the technology followed within a generation. The pattern recurs in steel, in shipbuilding, in semiconductor fabrication.

The deeper observation is that the Industrial Revolution is conventionally dated to 1760-1840 in the schoolroom account, but the technologies it deployed were often 200-400 years older than that. The stocking frame, the printing press, the heavy plow, the windmill, the watermill, the lock and the pin, the screw and the lathe: each was invented during the medieval or early-modern period and spent centuries in pre-industrial deployment before being scaled to factory operations. The Revolution was less about new technologies than about new institutional arrangements for deploying technologies that had been waiting for the institutional architecture to catch up. The framework knitters of Calverton in 1620 were operating one of the most sophisticated machines on Earth; they just did not know that yet.