«by Robert C. Allen Professor of Economic History Department of Economics and Nuffield College Oxford University Email: ...»
The spinning jenny was an engineering challenge. It did not require a scientific breakthrough or a great leap of imagination.
Arkwright’s water frame was another spinning technique that was more portentous in its consequences and arguably more clever in its design. But, again, it was based neither on a scientific breakthrough nor on an original idea.
Figure 8 shows a water frame, and Figure 9 is a close-up of the ‘clockwork’. The rovings entered at the top. They then passed through three pairs of rollers. The rollers operated like mangels, pulling the cotton between them. The second pair spun at twice the speed of the first, and the third doubled the speed again. For this reason, the first pair of rollers simultaneously pulled the roving into the mechanism and at the same time held it back with respect to the second pair, which was spinning faster and tugging it forward. The cotton was, thus, stretched and thinned out as it went between the two pairs of rollers. The stretching was repeated Figure 9 between the second and third pairs of rollers since the third pair spun faster than the second. In this way, the water frame accomplished the first task in spinning–drawing out the fibre.
The second task was accomplished by the flyers, which spun around at the bottom of the frame, simultaneously twisting the fibre and coiling it on the bobbin.
Not much of this was original with Arkwright.
The flyer, indeed, was an old device and none of the cotton inventors could take credit for it (another example of copying). The novelty of the water frame lay in the trains of rollers that drew out the cotton. This idea, however, was not Arkwright’s either: Wyatt and Paul took out patents on the idea in 1738 and 1758. Much effort was put into perfecting the machine, licenses were sold, and they erected their own factory in Birmingham.
It was not successful, although Matthew Boulton thought it might have been had it been well managed. The Wyatt and Paul R&D program was a failure.
If there were any macro inventors, they were Wyatt and Paul. But were they? The test of a macro-invention is whether it was conceived ab nihilo or whether it had a pedigree that shows that it involved only a small variation in practice. By that test, roller spinning was a micro-invention. Rollers were a general purpose Figure 10 technology whose use was spreading in the early eighteenth century.8 Rollers had a long history in metallurgy where bars, ingots, plates, and nails were shaped (Figure 10). Coin faces were pressed into gold and silver with engraved rollers.
Indeed, the similarities between a metal rolling mill and roller spinning were so great that Rees (1819-20, II, p. 173) reports that Arkwright conceived of roller spinning when looking at a rolling mill. There are sixteenth and seventeenth century designs for corn mills using rollers. In the late seventeenth century, cast glass was rolled at Saint-Gobain and polished with a roller. Cloth was pressed by rollers under enormous weight in the calendering process. In 1696, the Paris mint was using rollers. In the late seventeenth century, ‘milled’ sheet formed by rolling lead replaced cast lead sheet. In 1670, the Dutch developed a roller device with spikes to tear up rags for paper making and in 1720 applied rollers to pressing paper. Rollers were also used to crush rock. Applying rollers to stretching cotton was no doubt clever, but the idea had a history. When he discussed Cort’s invention of puddling and rolling, Mokyr (1993, p. 22) discounted it as a macro invention since rolling had a long history in metallurgy. The same argument applies to cotton. Rollers were in the air in the first half of the eighteenth century.
Wyatt and Paul did not think them up from nowhere. Roller spinning was not a macro invention.
Hargreaves’ and Arkwright’s R&D projects
The challenge with roller spinning was making the idea work. Hargreaves faced the easier challenge. His first jenny was reportedly made with a pocket knife, but getting a design that could be operated satisfactorily took from 1764 to 1767 (Aspin and Chapman 1964, p. 13). Hargreaves began trying to realize money from his invention almost immediately by selling jennies. He moved to Nottingham. As he continued to improve the jenny he needed a financial backer. He first went into partnership with a man named Shipley and later with Thomas James (Aspin and Chapman 1964, 19, 22-3, 34-5). They established a spinning factory. In 1770, Hargreaves patented the jenny, but it was too late. His patent was challenged in court and eventually voided on the grounds that he had sold jennies before it was issued. Despite the widespread use of the jenny in the late eighteenth century, Hargreaves realized very little money from the invention.
Arkwright’s challenge was far greater. Figure 11 shows Wyatt and Paul’s diagram from their second patent, and it can be compared to the Arkwright machine to see the engineering problems involved. Both devices used a flyer to twist and wind the finished thread. Wyatt and Paul’s diagram shows one pair of rollers, whereas Arkwright’s frame had Singer, et al. (1957, Vol. III, pp. 16-7, 32, 45, 47, 177, 238-9, 340-4, 414-5), Raistrick 1972, p. 91), Rowe (1983, pp. 8-10), Beveridge (1939, pp. 191-2, 287-9, 485-9, 652-6) Mokyr (1990, p. 60), Hunter (1930, pp. 170-1).
Figure 11 three. It was essential to have several in a series so that they could pull against each other. Wyatt and Paul did mentioned two pairs in the description of the machine in their first patent: Deciding the number of rollers was a development challenge, and it looks as though Wyatt and Paul went down a wrong alley in their R&D program by trying to develop a machine with only one set of rollers.
They never confronted, therefore, the other development challenges that Arkwright
overcame in the 1760s. These included:
• The increase in speed from one set of rollers to the next. In the early water frame displayed in Strutt’s North Mill, Belper rotation speed doubles from one train of rolls to the next.
• How to arrange the gears to connect the main power shaft to the rollers and coordinate their movements. The rollers and gears were produced as a module known as the ‘clock work’ in recognition of the apparatus that inspired it.
• The spacing between the rollers. The distance had to be slightly less than the length of a cotton fibre. That allowed stretching and thinning of the thread since a fibre that was past the grip of the first rollers and caught by the second pair could be pulled ahead of an adjacent fibre that was held by the first rollers but not yet in the grasp of the second. If the rollers were too close, all of the fibres would be gripped by both pairs, so there would be no stretching. If the rollers were too distant, the thread would be pulled apart: Proper operation required some fibres to be gripped by both rollers to prevent breakage, while others were held by one or the other pair for thinning.
Thought and experimentation were required to work this out.
• The materials with which to make the rollers. One was grooved metal and the other wood covered with leather. They had to pull the fibre without catching.
• The pressure with which the top roller pressed down on the bottom one. This was regulated by hanging weights from the top ones, as shown in Figure 9. The optimal weight could only be determined by repeated trials.
The point of this discussion is to emphasize the real issues involved in ‘inventing’ mechanical spinning. The originality was not in thinking up the roller; rather, the challenges were the practical issues of making the roller work in the application. Wyatt and Paul spent some years on this, but did not succeed. Arkwright employed clockmakers over a five year period to perfect the design. We have no record of exactly what they did, but the comparison of the Wyatt and Paul design with Arkwright’s frame highlights the problems they faced.
These challenges could only be met by constructing models or experimental prototypes.
‘Inventing’ the water frame involved a significant R&D program.
The R&D program had very modern financial implications that are worth noting.
First, the object was to make money for Arkwright, and patenting the invention was the essential step in securing that income. This was done in 1769. Second, there was the formidable problem of financing the R&D. Arkwright did what modern inventors do: he found venture capitalists–‘projectors’ in the language of the eighteenth century. His patent was jointly held with John Smalley and David Thornley, and each partner was committed to finance one third of the development costs. Quickly they ran out of money, and Samuel Need and Jedediah Strutt were brought in as partners. Strutt was an established ‘projector,’ who had already made a fortune financing improvements in frame knitting. Development work continued. Strutt himself suggested dusting the rollers with chalk to prevent the cotton from sticking to them. Several cam operated devices were added to wind the thread, raise and lower the bobbins and move the thread back and forth along the rollers to prevent a groove’s being worn in the surface. In 1774, Jedediah Strutt claimed that £13,000 had been spent on developing Arkwright’s device. This included the construction of buildings, which posed problems of layout and power transmission, and it indicates the scale of the finance required to turn the idea of roller spinning into the reality of a working cotton mill (Hills 1970, pp. 60Roller spinning was not unusual. If we examine the revolutionary inventions of the eighteenth century, we see that they were not based on revolutionary ideas. They were based on little ideas and often on copying products and practices from other places or industries.
Success depended on solving the engineering problems in making the simple idea work.
Edison famously remarked that ‘invention was 1% inspiration and 99% perspiration.’ Sweat was at least as important in the eighteenth century as it was in the late nineteenth. Mokyr (1993, p. 33) correctly observed that Britain ‘had a comparative advantage in microinventions.’ The questions are where that advantage came from, and why it was activated.
What was the motive for mechanizing spinning?
Mechanical spinning was a child of globalization. India was the world’s greatest cotton textile producer, and the East India company imported vast amounts of printed cotton cloth. This was important for later developments, for it showed that there was a large British market. So much was imported, that wool and linen manufacturers succeeded in 1701 in having printed cotton fabrics excluded from Britain. The import of white cottons was still permitted, and printing was done in England. A small British production of cotton cloth ensued. In 1721, the ban was extended to all cotton fabrics: the domestic production and consumption of purely cotton fabrics was made illegal. “The Lancashire cotton industry...secured in 1736 a relaxation for goods of flax warp and cotton weft [called fustians], a relaxation which by custom (or subterfuge) came to cover the great bulk of the industry’s production and even, it is probable, the growing part of it that used hand-spun cotton twist for warps,” i.e. all cotton cloth (Fitton and Wadsworth 1958, p. 68). English cotton producers, thus, received ambiguous protection from Indian imports. Similar restrictions were imposed in other European countries. While offering domestic protection, the laws did permit the importation of Indian cottons for re-export, and that market boomed with the growth of the slave trade in the mid-eighteenth century, for cotton cloth was bartered with African chiefs for slaves. This was another market which British producers could hope to supply–if their costs were competitive.
Britain’s high wage economy affected the cotton industry in two respects. First, the high incomes of British workers underpinned the mass market in cloth that was revealed during the period of unrestricted imports (Lemire 1991, p. 55). Second, at the exchange rate, British wages were considerably higher than Indian wages. While distance provided some protection, English spinners could only compete in producing the coarsest yarn, which was the least labour intensive.
Lowering labour costs was the key to competitiveness. There was a large potential domestic market, and a vast foreign market supplied by India and other producers. Cost reductions promised a large increase in market share and immense fortunes for the successful innovators–both of which were realized through mechanization.
Why not France?
Globalization affected other European countries as it affected England. For much of the pre-industrial period, France had possessions in India and was flooded with Indian calicoes in the late seventeenth century. Their importation was banned in 1686. France also had new world colonies and was active in the slave trade where French ships carried about 40% of the volume of English ships (Curtin 1966, pp. 211-2). French producers had an African market, albeit a smaller one than the English. In 1786, when English production was already soaring as mechanized spinning spread, Britain imported 18 million pounds of raw cotton, while France imported 11 million (Crouzet 1985, p. 32). The French cotton market was substantial, and French manufacturers had opportunities to compete against Indian textiles in Africa like their British counterparts, a feature emphasized by Inikori (2002, pp.