A Short History of the British Industrial Revolution (Palgrave, 2011)

6. ‘The Mechanical Age: technology, innovation and industrialisation. Were we required to characterise this age of ours by any single epithet, we should be tempted to call it, not an Heroical, Devotional, Philosophical, or Moral Age, but, above all others, the Mechanical Age. It is the Age of Machinery (Carlyle, ‘Signs of the Times’, 1829). Thomas Carlyle, ‘Signs of the Times’, Edinburgh Review, 49 (June 1829), 438-59. It is upon the excellency of machinery that the superiority of British manufactures chiefly depends. In other countries labour may be cheaper, and in some the raw material may be more easily obtained, but as yet no country can equal Great Britain in the speed and perfection of machinery (Lawson, Geography of the British Empire, 1861). William Lawson, Geography of the British Empire (London, 1861; repr. 1866, p.192. For many Victorians rapid advances in technology, in particular the use of machines to perform work that had previously been done by hand, was one of the most striking developments of the age. The mechanisation of the cotton industry, the invention of the steam engine and a myriad other ‘contrivances’ and innovations in many branches of industry were been taken as emblematic of nineteenth-century economic progress. This emphasis on technology and machines has also continued throughout much of the twentieth century. In the late 1940s, for example, the economic historian T. S. Ashton spoke of a cadre of ‘Inventors, contrivers, industrialists, and entrepreneurs … from every social class and from all parts of the country’ busy at work fashioning the inventions that were to drive the industrial revolution. He continued: ‘It was not only gadgets, however, but innovations of various kinds – in agriculture, transport, manufacture, trade, and finance – that surged up with a suddenness for which it is difficult to find a parallel at any other time or place. T. S. Ashton, The Industrial Revolution, 1760-1830 (Oxford, 1948; repr. 1996), p. 13, 48. Whilst not everybody shared Ashton’s generally rather rosy account of the industrial revolution, his emphasis on the transformative role of new technologies continued to resonate throughout the second half of the twentieth century. In the Cambridge Economic History of Europe, published in 1965, David Landes’ offered a broad definition of the industrial revolution based upon a number of technological advances, encompassing the substitution of human strength with machines; new sources of power (fossil fuels and the steam engine); and new materials (iron and minerals). David Landes, ‘Technological change and development in western Europe, 1750-1914’, in H.J. Habbakuk and M. Postan, eds., The Cambridge Economic History of Europe, vi. (Cambridge, 1965); Idem, The Unbound Prometheus. Technological Change and Development in Western Europe from 1750 to the Present (Cambridge, 1969), p.41. More recently still, Joel Mokyr has argued that if ‘European technology had stopped dead in its tracks – as Islam’s had by about 1200, China’s had by 1450, and Japan’s had by 1600’, then Europe would not have continued down its path of industrialisation in the two centuries following 1750. Britain’s industries, he adds, ‘displayed an unprecedented technological creativity that lay at the foundation of the British Industrial Revolution’. Joel Mokyr, The Lever of Riches. Technological Creativity and Economic Progress (Oxford, 1990), p.81; Idem, ‘Editor’s introduction: the New Economic History and the Industrial Revolution’, in idem. (ed.), The British Industrial Revolution: an Economic Perspective (Boulder, Colorado, 1993), p.18. For many, both contemporaries and historians, the rapid pace of technological change after 1750 is not simply a colourful historical curiosity; it is the key to understanding the world’s first industrial revolution. This amounts to a large claim for the significance of technology and in this chapter we shall consider whether inventions and technology do indeed deserve a place at the centre of our definitions of the industrial revolution. It should immediately be clear that this account of pervasive and transformative technological change has been seriously challenged by the recent, and hugely influential, macroeconomic analyses of British industrialisation. Crafts and Harley’s estimates of national economic growth suggested that productivity growth was heavily localised in two ‘modern’ industries – cotton, and, to a lesser extent, iron – with only meagre productivity gains elsewhere. See especially Harley, ‘Reassessing the industrial revolution: a macro view’, in Mokyr, ed., The British Industrial Revolution: an Economic Perspective (Boulder, Colorado, 1993), pp.171-226, pp.197-200; Crafts, British Economic Growth, pp. 84-6. The poor productivity gains in industries outside the two modern sectors, led them to infer that most ‘other industries remained largely unchanged’, and therefore to reject accounts stressing widespread technological change. Harley, ‘Reassessing the industrial revolution’, p. 200. We have already reviewed a number of criticisms of macroeconomic approaches to British industrialisation and cautioned that in the absence of reliable data to measure the various elements of economic growth the results must be viewed as subject to a sizeable margin of error. There are a number of further reasons why their description of a languishing manufacturing sector, if not incorrect, might be misleading. In the first instance, Crafts and Harley’s estimates for industrial productivity were never based upon a complete analysis of the manufacturing sector. Their indices of industrial production were initially based upon samples of around a dozen different industries, though the size of the sample was slightly extended in subsequent revisions. Harley, ‘British industrialisation before 1841: evidence of slower growth during the industrial revolution’, Journal of Economic History, 42/2 (1982), 267-89; Crafts, British Economic Growth, pp.17-25; N. F. R. Crafts & C. K. Harley, ‘Output Growth and the British Industrial Revolution: A Restatement of the Crafts-Harley View’, Economic History Review, 45/4 (1992), pp. 703-730. Nonetheless, a recent study of industrial output for twenty-six industries for the period from 1815-1850 acknowledges that these industries amount to only about 60 per cent of total industrial production – still leaving fully forty percent entirely out of the account. D. Greasley & L. Oxley, ‘British industrialization, 1815-1860: A disaggregate time-series perspective’, Explorations in Economic History, 37/1 (2000), 98-119, p.101. Most of those excluded were relatively small industries, and it has even been suggested that it was in precisely some of these small, dynamic industries that technological innovation was most pervasive – a claim that Crafts and Harley, unsurprisingly, dispute. See M Berg and P Hudson, ‘Rehabilitating the Industrial Revolution’, Economic History Review, 45/1 (1992), pp.24-50; Crafts & Harley, ‘Output Growth’, pp.711 At any rate, our current macroeconomic estimates do not measure the productivity of these smaller industries, and the absence of such measurement should force us to pause before deciding whether change in manufacturing was localised in the ‘modern’ industries or was in fact more widely spread throughout the sector. Peter Temin has turned to import and export data as an alternative way of gauging the extent and significance of technological change in some of the smaller industries not included in the Crafts-Harley analysis. Peter Temin, ‘Two views of the British industrial revolution’, Journal of Economic History, 57/1 (1997), 63-82. Between a quarter and a half of all manufacturing exports between the late eighteenth and mid nineteenth centuries were of goods other than textiles and iron: they included items such as cutlery, pottery, clothing, glassware, books, umbrellas, hats, and fishing tackles. If the manufacture of these goods was undergoing improvements in productivity, one should expect their export to expand: without these improvements, exports should stagnate or be replaced by imports. Temin’s analysis indicates that exports in such industries were keeping pace with exports in cotton, and he infers from this that technological progress must have occurred within them. This, it should be stressed, is not direct evidence for technological improvements within these smaller, older, industries, any more than Crafts or Harley ever provided direct evidence for its absence. Trade data reveals that Britain enjoyed a comparative advantage in these industries relative to her foreign neighbours, but does not indicate what underpinned this comparative advantage. Nonetheless, the existence of a buoyant export market of so many items outside the lead sectors of cotton and iron sits rather uneasily with the Crafts-Harley account of a stagnant traditional manufacturing sector. See, however, their response to Temin’s analysis. C. K. Harley & N. F. R. Crafts, ‘Simulating the two views of the British Industrial Revolution’, Journal of Economic History, 60/3 (2000), 819-841. The evidence from patenting also paints a rather different picture of manufacturing progress to that depicted by Crafts and Harley. A patent is a grant by the state of exclusive rights for the use of a new invention for a defined period of time (during this period it was set at fourteen years), and the granting of patents has been used by historians as a crude yardstick of inventive activity. The definitive history of patents remains Christine MacLeod, Inventing the Industrial Revolution: The English Patent System, 1660-1800 (Cambridge, 1988). A useful summary may also be found in Kristine Bruland, ‘Industrialisation and technological change‘, in Roderick Floud and Paul Johnson, eds., The Cambridge Economic History of Modern Britain: Industrialisation, 1700-1860, i. (Cambridge, 2004), pp.117-146. Care must be exercised in linking the patent series to technical and industrial change. A patent was expensive and difficult to obtain and some of the industrial revolution’s most significant inventions – James Hargreaves’ spinning jenny and Samuel Crompton’s mule, for example – were never successfully patented, though the omission of such major innovations from the patent series was in fact a rather unusual occurrence. Hargreaves sold a number of jennies before taking out a patent, and for this reason his claim was later rejected in the courts. Crompton was unable to patent his mule, probably because Arkwright’s patent on his ‘throstle’ or ‘frame’ barred the way. See Ashton, The Industrial Revolution, pp.58-60. At the same time as some major inventions slipped through the patenting system, not every invention that was patented signified a fundamental technological breakthrough: whilst a handful were obtained for a radical new invention, many others were obtained for relatively minor improvements to existing techniques, or even simply in an attempt to evade the restrictions of existing patents. Bruland, ‘Industrialisation’, pp.122-3. Furthermore, the mere existence of a patent does not provide evidence that the patented device was ever produced and marketed successfully. More than one patented idea has failed to make the transition from inventor’s workshop to commercially viable product. It is little wonder, therefore, that the historian of patents, Christine MacLeod, has cautioned that the patent series ‘related to technological change in an erratic and tangential manner’. MacLeod, Inventing the Industrial Revolution, p. 157. Despite these caveats, however, a number of clear and very interesting trends emerge from the patenting record. In the first instance, the scale and extent of patenting activity expanded considerably during this period, particularly in the years following 1750. The rate of change accelerated so sharply after 1762 that one historian has suggested that England “entered her ‘Age of Invention’” at this time, defined as a period of self-sustaining growth in technology. Richard J. Sullivan, "England's 'Age of Invention': The Acceleration of Patents and Patentable Invention during the Industrial Revolution," Explorations in Economic History, 26 (1989), pp. 424-52, p.447, p.445. Given the great variety of inventions that underlie the patent series, it is helpful to break it down further, and consider which sectors of the economy were patenting new ideas, and what kinds of inventions they were seeking to protect. An analysis of the series by Richard Sullivan indicates the particular importance of machinery and motive power inventions (steam engines and other devices for transmitting power). Between 1750 and 1850, about a third of all patents concerned machinery and machine parts, whilst motive power accounted for about 7 per cent in the first fifty years rising to 14 per cent in the fifty years thereafter. Sullivan, ‘England’s “Age of Invention”’, p.442. And whilst some of this machinery was for use in the textile industry, taken as a whole, the patent series is not dominated by the cotton industry in the way that productivity figures might suggest. Around fifteen per cent of patents for capital goods between 1750 and 1800 were for textile machines (106 patents in all); and textile machines made up no more than six per cent of all patents issued. MacLeod, Inventing the Industrial Revolution, p. 148. See also Trevor Griffiths, Philip A. Hunt, Patrick K. O'Brien, ‘Inventive Activity in the British Textile Industry, 1700-1800’, Journal of Economic History, 52/4 (1992), pp. 881-906. This left over one and a half thousand patents taken out on a very wide range of inventions and innovations spread across the economy – agriculture, shipbuilding, canals, chemical equipment to name a few. Once again, the evidence from patents sits rather uneasily with the claim that there was little technological change outside the cotton industry before about 1850. The error lies in assuming that measures of productivity are a good indicator of the industrial processes at work in the economy. New technology is expensive to purchase, often prone to failure, and requires new workers to be trained to its use. All these factors mean that manufacturers are often slow to purchase new equipment and likely to wait a considerable period before seeing much return on their outlay. To give one example, James Watt’s separate condenser undeniably improved the fuel efficiency of existing steam engines, yet many manufacturers preferred to continue with their fuel-hungry Newcomen engines, as the cost of purchasing a Watt engine and paying his annual premiums outweighed any fuel savings that could be made, at least before his patent expired in 1800. Michael W. Flinn, The History of the British Coal Industry: 1700-1830: the Industrial Revolution, ii. (Oxford, 1984), pp.124-6; Steven King and Geoffrey Timmins, Making Sense of the Industrial Revolution (Manchester, 2001), pp.84-5. Despite the fact that productivity gains were highly localised in two sectors of the manufacturing economy, the evidence from exports and patenting suggests that technological change was occurring on a much wider basis in the century following 1750. New inventions can be slow to diffuse and measurements of changes in the rate of productivity are unlikely to reflect the underlying changes in industrial techniques. It is one thing, however, to demonstrate that technological change was pervasive and quite another to evaluate its significance in powering the industrial revolution. As the history of earlier great civilisations demonstrates, it is possible to have extensive, and even revolutionary, technological change, without having an industrial revolution. In the case of Britain, industrialisation appeared to turn a switch, marking the end of a period of limited population growth and limited economic expansion and the beginning of an era in which both population and the economy appeared able to grow without limits. Our question concerns the role played by new technologies in turning that switch. Let us turn, then, from assessing the extent of inventive activity to evaluating its wider impact on the British economy. Ever since Toynbee’s popularisation of the term ‘industrial revolution’ in the late nineteenth century, the mechanisation of the cotton industry, and of cotton spinning in particular, has lain at the heart of historical accounts of British industrialisation. It is not difficult to understand why so much significance has been placed on this one industry. In the period 1772-74, England imported 4.2 million lb of raw cotton. By 1839-41, the annual average had risen by an astonishing one hundredfold to 452 million lb. Geoffrey Timmins, Made in Lancashire. A History of Regional Industrialisation (Manchester, 1998), p.85, table 6.1, p.159. Over roughly the same period, the price of cotton cloth dropped by 85 per cent. Mokyr, Lever of Riches, p. 111. See also M. J. Daunton, Progress and Poverty. An Economic and Social History of Britain (Oxford, 1995), pp.186-90 and ‘Statistical Appendix’, table 3.d.i.-ii, pp.586-7. In fact, as this period also witnessed the rapid growth of muslins and fine cotton cloths, requiring less raw cotton to produce, it is likely that cotton output actually increased yet more rapidly than the figures for raw cotton imports suggest. Certainly, by any measure, the manufacture of cotton textiles underwent an astonishing expansion in the century following 1750, experiencing an acceleration of growth that was unmatched in Britain’s earlier industrial history. It is clear that something exceptional was happening in the cotton industry in the late eighteenth and early nineteenth centuries, and there can be no better starting point for considering the importance of technological change to Britain’s industrial revolution. Whilst the myriad changes at work in the cotton industry escape simple classification, we must look to a series of major technological breakthroughs in the eighteenth century, causing the mechanisation of work that had previously been done by hand (and the subsequent movement of work out of the home into the factory), in order to understand the historical development of this one particular industry. Before cotton cloth can be woven, the yarn has to be spun into thread, and it was in this branch of the industry – the spinning industry – that some of the most significant advances were made. During most of the eighteenth century and earlier, cotton yarn had been spun by hand, between thumb and forefinger, at a small wheel turned by hand: it was usually women’s work, and it was generally performed at home. It was a labour intensive process, and so, despite the low wages generally paid to women, a relatively costly task. In the 1760s and 1770s, the process of spinning was revolutionised by a series of inventions: the spinning jenny, the water frame, and the mule, which together replaced the work performed by women’s hands with various mechanical devices. Good accounts of technological change in the spinning industry may be found in: Landes, Unbound Prometheus, pp. 82-86; Geoffrey Timmins, ‘Technological change’, in Mary B. Rose, ed., The Lancashire Cotton Industry. A History since 1700 (Preston, 1996), 29-62, p.44-5; Mokyr, Lever of Riches, pp.96-99; Bruland, ‘Industrialisation’, pp.135-6. James Hargreaves’ spinning jenny replaced the spinner’s one spindle with several (initially eight or sixteen), enabling the machine operator to spin the yarn onto several spindles at the one time, thereby considerably increasing the quantity of yarn that could be spun in a given period of time. Richard Arkwright’s ‘frame’ (or ‘throstle’) produced a stronger thread by using three sets of paired rollers to produce yarn and a set of spindles to twist the fibres together; it was too large to be operated by hand, and after some experimentation was powered by a water wheel instead, thereafter becoming known as the ‘water frame’. Samuel Crompton’s mule combined elements of both the jenny and the water frame to spin strong and good quality cotton thread, which in turn facilitated the weaving of fine cotton cloth on a large scale. The mule required a skilled operator, but Richard Roberts’ ‘self-acting’ mule, patented nearly fifty years later in 1825, made the operator unnecessary, and ushered in the first truly automatic machine. The mechanisation of elements of the process of cotton manufacture that had traditionally been performed by hand enabled industrialists to replace human skill and effort with machines and vastly increased the productivity of the industry within a matter of decades. Whereas a worker spinning cotton on a hand-operated wheel in the middle of the eighteenth century might take more than 50,000 hours to spin 100lb of cotton, by the 1790s the same quantity of cotton might be spun in just 300 hours by mule, and the self-acting mule reduced the figure to 135. Timmins, ‘Technological change’, pp.44-5; and Mokyr, Lever of Riches, p.99. Figure 6.1 provides an illustration of rising imports over the century 1750-1850, and clearly illustrates the great expansion of the industry that occurred during this period of rapid technological change. [insert Figure 6.1. Retained imports of raw cotton, 1750-1810, raw cotton consumption, 1810-1850. Source: Daunton, Progress and Poverty, ‘Statistical Appendix’, table 3.d.i.-ii, pp.586-7. ] [insert images 6.1, 6.2, & 6.3. 6.1 Woman spinning on the one thread wheel. Edward Baines, History of the Cotton Manufacture in Great Britain (London, 1835); 6.2. Arkwright, Hargreaves and Crompton’s spinning machines. John James, History of the Worsted Manufacture in England (London, 1857); 6.3. Mule spinning. Edward Baines, History of the Cotton Manufacture in Great Britain (London, 1835). Layout and text: the three images, 61.-6.3 to appear close together with the following text. 1. Woman spinning on the one thread wheel. A woman using a hand-operated spinning wheel could only spin one thread of yarn at a time. It was a labour intensive, and therefore relatively costly, procedure. 2. Arkwright, Hargreaves and Crompton’s spinning machines. These three inventions – Arkwright’s throstle or frame, Hargreaves’ spinning jenny and Crompton’s mule, which combined elements of the other two – enabled one operator to spin several threads at one time. 3. Mule spinning. This image shows the interior of a large, steam-powered spinning factory, using the self-acting mule, first patented by Richard Roberts in 1825. Here a handful of employers oversee the spinning of hundreds of threads at once. The advantages over the simple spinning wheel are clear.] Whilst the spinning industry witnessed the most rapid increases in productivity in the late eighteenth and early nineteenth centuries, a host of new innovations in the bleaching, dyeing and printing sections of the cotton industry helped to transform these sectors too. In the bleaching industry, for example, the traditional method of open-air bleaching using buttermilk took up to eight months to complete. In the eighteenth century, this method was superseded by new techniques developed in chemical industry, using at first dilute sulphuric acid, and then lime chloride to cut production times from months down to days. Timmins, Made in Lancashire, p.127. See, also, G. N. von Tunzelmann, ‘Time-saving technological change: the cotton industry in the English industrial revolution’, Explorations in Economic History, 32 (1995), 1-27, 11. New washing and drying machines introduced in the early nineteenth century speeded up the bleaching process yet further. In the printing branch, Joseph Bell’s mechanised copper rollers replaced the older method of block printing by hand. One contemporary estimate suggested that a cylinder printing machine operated by a man and boy could do the same work as a hundred block printers, each with a boy to assist. Timmins, Made in Lancashire, p.128. These innovations led to far-reaching changes in these industries, driving up both output and productivity, and changing the working patterns of the thousands of men, women and children employed in them. [Insert image 6.4 Calico printing. Cylinder printing using copper rollers. According to one contemporary: ‘The saving of labour … is immense: one of the cylinder printing machines, attended by a man and a boy, is actually capable of producing as much work as could be turned out by one hundred block printers and as many tear-boys! In consequence of the wonderful facility given to the operation, three-fourths of all the prints executed in this country are printed by the cylinder machine’ (Baines, History of Cotton Manufacture, p.266.) The cotton industry was in the van of technological progress, yet it is important to note even here the incomplete nature of technological change. Although productivity in the spinning and finishing branches of the cotton industry was quickly pushed up through the use of new powered machinery, change occurred much more slowly in the intermediate stage: weaving. With the advantages of powered machinery so evident in the spinning industry, cotton manufacturers held high hopes that machines might also replace human labour in the field of weaving and a series of power looms invented by Edmund Cartwright in the 1780s, William Horrocks (1803), and Sharp and Roberts (1822) appeared to herald the realisation of these hopes. In 1825, the Manchester Chamber of Commerce declared that the new power loom brought ‘the whole process of manufacture, from the raw material to the cloth, into one connected series of operations, by means of which, a cheaper, more uniform and better fabric has been produced’. Roger Lloyd-Jones and M. J. Lewis, British Industrial Capitalism since the Industrial Revolution (London, 1998), p.41 But the reality was rather different from this hopeful vision. Cartright’s and Horrocks’ power looms did not work properly, and although Sharp and Roberts power loom marked a noticeable improvement on its predecessors, even it could not weave fine or weak threads. Timmins, The Last Shift. The Decline of Handloom Weaving in Nineteenth-century Lancashire (Manchester, 1993), pp.157-9. Furthermore, early power looms required very close attention from the operator as they needed to be stopped as soon as a thread broke or the shuttle became empty. Most weavers could only operate one, or at best two, power looms at a time, so gains in productivity were less significant than their purchasers might have wished. Ibid., p.159 and Timmins, ‘Technological change’, pp. 45-7. Given the difficulties that manufacturers faced in developing an effective automated weaving machine, inventive activity continued to be focussed upon creating a more efficient hand-loom. William Radcliffe’s ‘dandy loom’, which enabled the woven cloth to be wound automatically onto a beam at the back of the loom marked a significant improvement on the existing handloom: raising the hand weaver’s productivity by as much as fifty per cent, it helped to sustain handloom weaving in the first half of the nineteenth century. Timmins, Made in Lancashire, p.130, p.170 In the mid 1830s, there were about twice as many weavers operating a variety of different handlooms as there were power-looms. Ibid., p.87 Determining the relative importance of the hand-powered branch of the industry on the one hand and of the water- or steam-powered branches on the other is difficult: the output of power looms was certainly greater than that of the handlooms, but the handloom weavers produced higher quality cloths with greater profit margins. Wherever the balance lies, it is certainly the case that the weaving trade was only partially mechanised before about 1830, and that even so late as 1850, the handloom weavers made up a sizeable minority of the total weaving workforce. Idem., The Last Shift, pp.108-118. Furthermore, despite the undoubted technological advances in some branches of the cotton industry, the endpoint of cloth manufacture – the turning of manufactured cloth into clothes, hats and accessories – was largely done by hand until the invention of the sewing machine in 1860. Mokyr, ‘Editor’s introduction’, p.12 Even in the cotton industry, then, where new technology undoubtedly ushered in some phenomenal advances, the technological revolution was not completed by the middle of the nineteenth century. Yet taken as a whole, the cotton industry, and the spinning industry in particular, demonstrate how powerful invention and innovation can be as a source of economic change and growth. Cotton manufacture industry combined a handful of path-breaking inventions with countless minor adjustments and modifications to existing processes to sharply cut the cost of producing cotton textiles. The mix of radically new inventions and small adaptations to existing machines underpinned explosive industrial growth of a kind that Britain had never seen before. Outside the cotton industry technical change proceeded far more slowly and with rather less spectacular results, but it is nonetheless possible to identify other industries that were revolutionised, at least to some degree, by the emergence of new technologies. The iron industry, for example, was the site of significant technological progress, and whilst this did not lead to growth rates as impressive as those of the cotton industry, it did help to revolutionise both this industry and other parts of the British economy. The production of wrought iron depends on two processes. First the raw iron ore is smelted in a blast furnace to produce pig iron. Owing to its high carbon content, pig iron is hard and brittle; it can be cast items such as into pots, ovens and cannon, but its uses are rather limited. The soft, malleable wrought or bar iron, which can be fashioned into nails, locks, tools, cutlery, horse shoes, machine parts, railway tracks, and countless other items has a far wider range of uses, so pig iron is therefore put through a second process, in order to refine it into wrought iron. Daunton, Progress and Poverty, pp.211-19; Mokyr, Lever of Riches, pp.93-5; J. R. Harris, The British Iron Industry, 1700-1850 (London, 1988), pp.30-40; Richard Hayman, Ironmaking: The History and Archaeology of the Iron Industry (Stroud, 2005), pp.34-63. New technologies significantly improved both of these elements of iron manufacture – smelting and refining – during the eighteenth century. Smelting was transformed by the replacement of charcoal with coke and by the development of the ‘hot blast’ furnace, which used the furnace’s own gases to heat the air inside. Refining was improved first by the Woods Brothers ‘potting and stamping’ process, and soon after by Henry Cort’s ‘puddling and rolling’ process, patented in 1783 and 1784. The work of refining pig iron had traditionally been performed by skilled ironworkers, who repeatedly heated and hammered the pig iron in small forges to beat out the impurities. Cort’s procedure used iron rods to stir and beat impurities out of the molten pig iron, and then passed what was left between iron rollers to press the final impurities away. The technique heralded the end of small forges and helped to produce a more uniform and cheaper product. These various improvements to iron manufacture helped to raise the output of wrought iron by an average of 4.5 per cent per year in the first half of the nineteenth century and led to steep reductions in its cost. Joel Mokyr, ‘Technological Change, 1700-1830’, in R Floud and D McCloskey, eds., The Economic History of Britain since 1700 (Cambridge, 2nd edn. 1994), pp.25-7; Roger Burt, ‘The extractive industries’ in Floud and Johnson, eds., Cambridge Economic History, p.448-9. Figure 6.2 provides an illustration of rising output over the century 1750-1850, and once again indicates that technological improvement during the period went hand in hand with a significant increase in output. [Insert figure 6.2 Pig-iron output, 1750-1850. Source: Daunton, Progress and Poverty, ‘Statistical Appendix’, table 3.c, p585.] In many respects, the changes that occurred in the iron industry were less impressive than those that occurred in cotton. In iron refining, for example, Cort’s puddling process was vital in increasing the production of wrought iron, but various rolling techniques had in fact existed for centuries, and as the foremost historian of technology, Joel Mokyr, has noted, ‘the conceptual novelty of the process was modest’. Mokyr, ‘Editor’s introduction’, p.22. Furthermore, improvements in smelting and refining iron ore, though they certainly led to a significant expansion of the industry, did not cause growth of the magnitude seen in the cotton industry. Whether measured in terms of its workforce, its output, or its growth rate, the achievements of the iron industry are overshadowed by those of cotton. Landes, Unbound Prometheus, p. 89; Nonetheless, the technical achievements of the iron industry are perceived by many as a cornerstone of the industrial revolution, as they made available a raw material – wrought iron – for which no substitute existed. Cotton textiles could be, and were, produced by hand. New spinning, weaving, bleaching and printing technologies all replicated, with greater or lesser success, processes that had previously been performed by hand: by improving production processes they radically lowered the cost of cotton cloth, but they changed the product in only minor ways. By contrast, there was no alternative to cheap, wrought iron, and as its price dropped it began gradually to replace the wood in bridges, ships, buildings, and machinery, and of course, it enabled the construction of new inventions, such as the steam engine and the railways. Indeed, so numerous were the uses of wrought iron that Cort’s puddling process has been singled out by one historian as ‘a crucial invention which made the industrial revolution possible’. P. Deane, The First Industrial Revolution (Cambridge, 1965), p. 130. In their different ways, the cotton and iron industries both lend support to the claim that technological change was the driving force behind the industrial revolution. Elsewhere, however, it is more difficult to demonstrate the primacy of new technologies in powering British industrialisation. The difficulty lies not in identifying inventive activity, which was present, to some degree, in almost every section of the manufacturing economy; but rather in evaluating the overall significance of this inventive activity. Nowhere are the difficulties of placing new technologies at the heart of the industrial revolution more apparent than with the example of the steam engine. Engines of various kinds had been available since the late seventeenth century, and their numbers had grown since the invention of Thomas Newcomen’s self-acting atmospheric engine in the early eighteenth century – the first engine properly to use steam to operate machinery. But Newcomen’s engine was large and noisy, operated in a jerky motion, and had a voracious appetite for fuel, all of which effectively limited its use to pumping water from mines, where the size and noise of the engine posed few difficulties and fuel was plentiful. As a result, most other industrial processes continued to be powered by other means: by wheels driven by water or by horses; or by small machines operated by hand or by foot. The value of steam power to manufacturing industry was greatly enhanced by James Watt’s realisation that the two phases of the engine’s cycle – the heating and cooling – could be separated. This enabled him to create an engine in the 1760s that was considerably more fuel efficient (four times so) and that ran with a smoother motion than its predecessors. Mokyr, Lever of Riches. Both factors contributed to its early adoption in the mining industry and helped to facilitate its spread to the textile industry in the early nineteenth century. Timmins, Made in Lancashire, p.131. By the 1830s steam had largely replaced the waterwheels that had powered the cotton industry through most of the eighteenth century; it was indeed an integral part of the technical revolution that occurred in that industry. Outside the mining and cotton industries, however, steam power penetrated far more slowly, and its contribution to the industrial revolution is far less clear. The mining and textile counties of Cornwall, Durham, Lancashire, Northumberland, Shropshire, Staffordshire and Yorkshire had between them well in excess of a 1,000 engines by the end of the eighteenth century, but several others – for example, Bedfordshire, Dorset, Hertfordshire, Suffolk, Sussex, and Wiltshire – had not a single one. See J. Kanefsky and J Robey, ‘Steam engines in eighteenth-century Britain: a quantitative assessment’, Technology and Culture, 21 (1980), pp.161-86, table 5, p.176. The historian of the west Midlands, where steam certainly did penetrate in the early nineteenth century, has nonetheless concluded that ‘many of the midland manufactures had no use for such a large measure of power which could not easily be turned on and off. For the majority of processes and in many works the traditional power of wind, water, man and animals continued to be used not only because they were cheaper but also because they were more appropriate and efficient in the particular context’. Marie Rowlands, The West Midlands from AD 1000 (Harlow, 1987), p.236. In the 1970s, von Tunzelmann set out to put some figures to the importance of steam engine by counterfactual analysis, or the ‘social savings method’, which seeks to measure the economic contribution of an innovation by calculating the saving in costs compared with the earlier alternative technology. His assessment of the number and use of steam engines based upon Watt’s ideas in the early nineteenth century reveals that had they never existed, national income around 1800 might have been lower by about 0.11%; without steam engines of any kind, it would have been 0.2 per cent lower – clearly very negligible quantities. G. N. Von Tunzelmann, Steam Power and British Industrialisation to 1860 (Oxford, 1978), p.286-7. These figures have received more recent confirmation from Nick Crafts. His calculations suggest that steam’s contribution to improvements in labour productivity was never more than 0.02 per cent per year prior to 1830, rising to 0.2 per cent per year over the next two decades; the contribution of steam power to labour productivity growth, he concludes, was trivial before 1830. Nicholas Crafts, ‘The first industrial revolution: Resolving the slow growth/rapid industrialization paradox’, Journal of the European Economic Association, vol. 3 no.2/3 (2005), pp.525–534 p. 528. Inevitably, measuring something so complex as the contribution of steam power to labour productivity in the early nineteenth century cannot be done with any great precision and the exact figures that von Tunzelmann and Crafts have provided should be taken with a pinch of salt. Nonetheless, the broad outlines of their estimates fit neatly with a wide range of qualitative studies, and we might readily concur with Patrick O’Brien that the ‘”age of steam” … remained imminent rather than dominant during the first stages of the industrial revolution’. P. K. O’Brien ‘The deconstruction of myths and reconstruction of metanarratives in global histories of material progress’, in Benedikt Stuchtey and Eckhardt Fuchs, Writing World History (Oxford, 2002). Yet the history of the steam engine also poses some difficulties for accounts that place technology at the base of the industrial revolution. The evidence concerning national economic growth rates and population growth and movement considered in the previous three chapters all points towards deep-seated change dating from as early as 1700; by 1800 Britain’s economy and population appear to be following a trajectory different to that of both the rest of Europe, and that of its own the past. Yet steam power had no real impact until at least the 1830s, and prior to then Britain’s economy was powered largely by its existing mix of waterwheels and hand power – indeed the period 1760-1830 has been labelled the ‘Age of Water Power’ by one historian. A. E. Musson, The Growth of British Industry (London, 1978), p.109. It is not simply, then, that the technology was slow to diffuse, as that so much was achieved without it. We will return to this problem in the next chapter where we consider coal – that vital fuel that made steam power possible. For the present, however, it must be emphasised, that the steam engine, one of the most significant technological inventions of the entire period, does not fit within the timeframe of our traditional narratives of industrialisation. Von Tunzelmann’s demonstration that Britain’s early industrialisation proceeded largely without the benefit of steam power has encouraged a generation of historians to turn attention away from the dramatic technological breakthrough and to emphasise in addition the significance of small scale invention and of successive improvements and modifications to existing techniques. These incremental innovations are sometimes termed ‘micro-inventions’ in contrast to ‘macro-inventions’ – the much rarer, groundbreaking inventions that open the possibility of performing tasks in an entirely new way. The distinction is helpfully described in Mokyr, ‘Editors introduction’, pp.17-24. By expanding our framework for technological advance, it is possible to develop a much broader view of technological progress during the industrial revolution. In this vein, hand power is not viewed simply as the poor cousin of the steam engine, but as a viable alternative equally capable of technological improvement. Early steam engines were not well adapted for many manufacturing processes and machines or tools powered by human muscles offered a vital degree of precision or dexterity with which steam powered machinery could not compete. We have already seen how this was the case with cotton weaving: the power looms could weave large quantities of coarse cloth but could not weave fine and fancy cloths as effectively as the handloom. Such examples can be multiplied endlessly. In ship-building, for example, timber needed to be sawed with an accuracy that machinery was as yet unable to impart, so hand-operated saws remained the mainstay of the industry. As one commentator explained: ‘it might at first thought be imagined that machine-worked saws would be used; but the curvatures and angles of the timber are so extremely varied, not only in different timbers, but also in different parts of the same timber, that the precision and regularity of machinery would here be thrown away, and indeed unavailable’. Quoted in King and Timmins, Making Sense, p. 75. And of course, hand tools and innovation are not exclusive. On the contrary, hand tools could be, and were, improved by incremental technological progress, in ways that were more or less significant. Raphael Samuel, ‘Workshop of the world: steam power and hand technology in mid Victorian Britain’, History Workshop Journal, 3 (1977), pp. 6-72. Once again, we have already described one such example from the cotton industry – the dandy loom – but countless other examples abound. Glassmaking, for example, was improved by the cylinder method adopted by Chance Brothers at their Birmingham works in the 1830s. The cylinder method enabled the production of larger panes of glass than existing methods, yet it still remained a hand technique performed by a skilled workman rather than a machine. King and Timmins, Making Sense, pp.71-2. Likewise, this more expansive approach to the role of technology has encouraged historians to look beyond machines and industrial processes and to focus attention on the finished product instead. This reminds us that part of the inventive process lay in providing attractive and desirable goods to consumers: new fabrics, coloured and patterned cloths, shiny buttons, cheap buckles, affordable tableware – the list of new products designed to appeal to the changing tastes of consumers is endless. One study of the cotton industry has suggested that inventive energies were equally divided between devices aimed at reducing the costs of labour and raw materials on the one hand and product innovation – improvements to the nature and appearance of the finished cloth – on the other. Trevor Griffiths, Philip A. Hunt, Patrick K. O'Brien, ‘Inventive Activity in the British Textile Industry, 1700-1800’, Journal of Economic History, 52/ 4 (1992), pp. 881-906, p.892. Similarly, Maxine Berg’s study of patents in the metal-wares, glass, ceramics, furniture, clocks and watches demonstrates that one quarter of the patents specified new products, improvements in ornamenting and finishing, or imitations of existing goods. M. Berg, ‘From imitation to invention: creating commodities in eighteenth-century Britain’, Economic History Review, 55/1 (2002), pp. 1-30. These improvements involved inventions such as Keen and Schmidt’s method of binding gold and silver to woollen cloth, John Peele’s method of printing images onto linen handkerchiefs, and Henry Clay’s manufacture of black, lacquered buttons. Griffiths, Hunt and O’Brien, ‘Inventive activity’, p.895; Berg, ‘Imitation to invention’, p.23. See also, more generally, Maxine Berg, Luxury and Pleasure in eighteenth-century Britain (Oxford, 2005). It is doubtless important to emphasise that hand powered technology was more than a simple precursor to steam and that it was often the site of significant improvement and innovation. It is also helpful to shift attention away from machines altogether and consider the range and extent of improvements to goods and products as well. Yet placing hand technology and product improvement at the heart of the industrial revolution would arguably be placing more weight on these small scale improvements than they can really bear. No matter how much hand technology was refined and improved, it still remained hand technology, an essentially older way of doing things. And whatever inventiveness was displayed in designing more attractive consumer products, it is also possible to remain unconvinced about the wider significance of new techniques to decorate cloth with gold, print images on handkerchiefs, and manufacture lacquered buttons. Enterprising producers had always sought to improve their existing methods of production and searched for new ways to make their goods more attractive and it is likely that during the eighteenth century they became more inventive in their quest to do so. For our purposes, however, the critical question remains what role this played in promoting British industrialisation. By any measure, the industrial revolution marked a great discontinuity with the past, and however numerous and pervasive micro-inventions and product improvements may have been, it is difficult to see how they could have played more than a small role in forcing this momentous transition. If extensive, yet ultimately small-scale, invention seems insufficient to account for the industrial revolution, it is also interesting to consider the example of towns and regions that managed to industrialise without a corresponding leap in inventive activity. Whilst the fit between technology and industrialisation is convincing in the case of Lancashire, the industrial history of many other towns and regions demonstrates the complex relationship that existed between technical innovations on the one hand and the process of industrialisation on the other. Consider, for example, the history of Birmingham. As we saw in the previous chapter, Birmingham had grown to become England’s fourth city in 1800, largely as a consequence of in-migration. Its population of 6,000 in 1700 was roughly equivalent to that of Canterbury, Cambridge, Salisbury and Hull; standing at about 74,000 a century later, it had increased more than tenfold, vastly outstripping the growth of most other towns of its original size. Wrigley, ‘Urban growth’, p.686. Yet it is difficult to credit new technologies with a major role in this transformation from middle-ranking regional centre to great city. The economy of Birmingham had been dominated by metal work since the sixteenth century at least, owing in large part to its proximity to iron ore and coal in Staffordshire and Worcestershire. Eric Hopkins, The Rise of the Manufacturing town. Birmingham and the Industrial Revolution (London, 1989; repr. Stroud 1998), pp. 3-4. In the middle of the eighteenth century, three metal industries – gun-making, brass, and ‘toys’ (small articles of brass and iron, such as buttons, buckles, pins and trinkets) – stood out in importance. All three branches developed and expanded considerably in the century that followed, yet groundbreaking, macro-inventions are perhaps most conspicuous here by their absence. In the gun-making trade, there were only two technological developments of importance: the use of water- or steam-powered rollers to produce gun-barrels; and the invention of the percussion cap, which replaced the flint-lock for firing the gun. Most other elements of gun-making remained largely unchanged, and as the historian of Birmingham has observed, this period witnessed ‘no fundamental change in working techniques for the majority of gun-trade workers’. Ibid., p.42. Likewise, although there were a number of improvements to the making of buttons, there was no clear technological break with the past in any branch of the toy industry. Ibid., pp.48-51. Only the brass industry saw significant technical improvements: the crucible method for making brass and the production of seamless brass tubes, though these innovations came in the 1830s, rather late in the history of Birmingham’s industrial and population growth. Ibid., pp.46-8. None of this should be taken to suggest that innovation was unimportant to Birmingham’s metalworking trades. Cumulatively, myriad minor improvements helped to increase output, reduce costs, and raise profits. Nonetheless, Birmingham remained firmly in the world of skilled workmen operating their handheld tools in small workshops well into the nineteenth century. No matter how broadly we define the process of technical change, the overall impact of new technologies in Birmingham was relatively modest, and this town therefore offers us a path to industrialisation that did not involve major technical change. [Insert images 6.5. & 6.6 depicting gun-making in Birmingham -- 6.5. Welding the gun-barrels, Birmingham. Illustrated London News, 1851; 6.6. Grinding the gun-barrels, Birmingham. Illustrated London News, 1851 These two engravings illustrate the traditional nature of gun-manufacture, Birmingham’s foremost industry. Skilled welders heat the barrel in a furnace and use hammers to mould the heated iron into a perfect tube. The grinders’ task is facilitated by the use of a large, steam-powered grind-stone, yet even with the addition of steam-power, each man grinds just one barrel at a time and needs to exercise some skill to achieve a smoothly finished barrel.] Birmingham certainly stands out for the rapid growth it achieved on the basis of only minor (albeit numerous) innovations, but it was by no means unique in its marriage of industrial expansion and the continued use of older production techniques. The towns and industrial villages to the north and west of Birmingham that made up the Black Country were largely devoted to simpler forms of metalworking than were to be found in the regional capital, and they too developed largely on the basis of existing techniques. The Staffordshire pottery industry provides another well known example of an industry which expanded without the benefit of spectacular technological breakthroughs. Mokyr, ‘Technological change’, p.28. A similar pattern of growth occurred in the hosiery (knitting) industry based in Nottinghamshire, Leicestershire and Derbyshire. The ‘stocking frame’ that formed the mainstay of the industry had been invented in the late sixteenth century. It roughly imitated the work of hand knitting and was used for the production of stockings and other knitted goods, such as waistcoats and gloves. In what must now be a familiar pattern, the process of framework knitting was improved throughout the eighteenth and early nineteenth centuries by a series of inventions. Attachments that could be added to a stocking frame, such as those patented by Jedediah Strutt in 1758 and John Morris in 1763, enabled its operator to knit a more elaborate and attractive stitch. Furthermore, the knitting industry responded to changes in fashions for its goods by countless innovations in the range of goods produced, offering new types of fabrics and knitted garments – velvets and brocades for waistcoats, ladies’ silk mitts and woollen polka jackets, lambswool drawers, and zigzag patterned stockings, to name a few. Stanley Chapman, Hosiery and Knitwear: Four Centuries of Small-scale Industry in Britain, c.1589-2000 (Oxford, 2002), pp. 1-11, 65-71, 105-11. Despite these advances, however, down to 1850s, most of those employed by the knitting industry were working at frames that were fundamentally the same as those in operation two centuries earlier, and the industry’s historian has concluded that hosiery provides ‘an interesting case of an industry whose structure and organisation barely changed in the century 1750 to 1850’. Ibid., p.52. No matter how widely we interpret technological change, it seems we are left with something less than an industrial revolution. This is not to deny the extent of inventive activity. The century following 1750 was undoubtedly a period of great innovation, involving both a small number of groundbreaking inventions, as well as a far larger number of small improvements, adjustments, and modifications to existing techniques; furthermore, this inventive process revolutionised some aspects of British manufacturing. In the cotton industry, a series of radical inventions vastly speeded up production processes, leading to an increase in output and reduction in price on an unprecedented scale. In metalworking, inventions helped to create a new product – wrought iron – at an affordable price, which in turn facilitated the creation and growth of many related industries and spawned a radically new form of transport – the railways. Yet in other spheres of the economy, the overall contribution of technology is less evident. The steam engine, for example, was novel in conception, but it was not widely adopted until well into the nineteenth century, well after the date at which the industrial revolution was conventionally thought to have begun. And the countless micro-inventions of this period, though they reveal an inventive spirit at work at the heart of the British economy and certainly helped contribute to economic expansion, nevertheless remained small in scale and their significance should not be overstated. Finally, there are regions such as Birmingham and parts of Staffordshire, Nottinghamshire, Leicestershire and Derbyshire which underwent significant industrial growth yet without much in the way of a technical revolution at all. In fine, the emergence of new technology provides no more than a partial explanation for Britain’s industrial revolution. If we are to provide a full account of what caused this turning point in British history it will be necessary to look beyond the machines and technology that so vividly captured the imagination of many Victorian commentators. PAGE 128

6. ‘The Mechanical Age: technology, innovation and industrialisation. Were we required to characterise this age of ours by any single epithet, we should be tempted to call it, not an Heroical, Devotional, Philosophical, or Moral Age, but, above all others, the Mechanical Age. It is the Age of Machinery (Carlyle, ‘Signs of the Times’, 1829). Thomas Carlyle, ‘Signs of the Times’, Edinburgh Review, 49 (June 1829), 438-59. It is upon the excellency of machinery that the superiority of British manufactures chiefly depends. In other countries labour may be cheaper, and in some the raw material may be more easily obtained, but as yet no country can equal Great Britain in the speed and perfection of machinery (Lawson, Geography of the British Empire, 1861). William Lawson, Geography of the British Empire (London, 1861; repr. 1866, p.192. For many Victorians rapid advances in technology, in particular the use of machines to perform work that had previously been done by hand, was one of the most striking developments of the age. The mechanisation of the cotton industry, the invention of the steam engine and a myriad other ‘contrivances’ and innovations in many branches of industry were been taken as emblematic of nineteenth-century economic progress. This emphasis on technology and machines has also continued throughout much of the twentieth century. In the late 1940s, for example, the economic historian T. S. Ashton spoke of a cadre of ‘Inventors, contrivers, industrialists, and entrepreneurs … from every social class and from all parts of the country’ busy at work fashioning the inventions that were to drive the industrial revolution. He continued: ‘It was not only gadgets, however, but innovations of various kinds – in agriculture, transport, manufacture, trade, and finance – that surged up with a suddenness for which it is difficult to find a parallel at any other time or place. T. S. Ashton, The Industrial Revolution, 1760-1830 (Oxford, 1948; repr. 1996), p. 13, 48. Whilst not everybody shared Ashton’s generally rather rosy account of the industrial revolution, his emphasis on the transformative role of new technologies continued to resonate throughout the second half of the twentieth century. In the Cambridge Economic History of Europe, published in 1965, David Landes’ offered a broad definition of the industrial revolution based upon a number of technological advances, encompassing the substitution of human strength with machines; new sources of power (fossil fuels and the steam engine); and new materials (iron and minerals). David Landes, ‘Technological change and development in western Europe, 1750-1914’, in H.J. Habbakuk and M. Postan, eds., The Cambridge Economic History of Europe, vi. (Cambridge, 1965); Idem, The Unbound Prometheus. Technological Change and Development in Western Europe from 1750 to the Present (Cambridge, 1969), p.41. More recently still, Joel Mokyr has argued that if ‘European technology had stopped dead in its tracks – as Islam’s had by about 1200, China’s had by 1450, and Japan’s had by 1600’, then Europe would not have continued down its path of industrialisation in the two centuries following 1750. Britain’s industries, he adds, ‘displayed an unprecedented technological creativity that lay at the foundation of the British Industrial Revolution’. Joel Mokyr, The Lever of Riches. Technological Creativity and Economic Progress (Oxford, 1990), p.81; Idem, ‘Editor’s introduction: the New Economic History and the Industrial Revolution’, in idem. (ed.), The British Industrial Revolution: an Economic Perspective (Boulder, Colorado, 1993), p.18. For many, both contemporaries and historians, the rapid pace of technological change after 1750 is not simply a colourful historical curiosity; it is the key to understanding the world’s first industrial revolution. This amounts to a large claim for the significance of technology and in this chapter we shall consider whether inventions and technology do indeed deserve a place at the centre of our definitions of the industrial revolution. It should immediately be clear that this account of pervasive and transformative technological change has been seriously challenged by the recent, and hugely influential, macroeconomic analyses of British industrialisation. Crafts and Harley’s estimates of national economic growth suggested that productivity growth was heavily localised in two ‘modern’ industries – cotton, and, to a lesser extent, iron – with only meagre productivity gains elsewhere. See especially Harley, ‘Reassessing the industrial revolution: a macro view’, in Mokyr, ed., The British Industrial Revolution: an Economic Perspective (Boulder, Colorado, 1993), pp.171-226, pp.197-200; Crafts, British Economic Growth, pp. 84-6. The poor productivity gains in industries outside the two modern sectors, led them to infer that most ‘other industries remained largely unchanged’, and therefore to reject accounts stressing widespread technological change. Harley, ‘Reassessing the industrial revolution’, p. 200. We have already reviewed a number of criticisms of macroeconomic approaches to British industrialisation and cautioned that in the absence of reliable data to measure the various elements of economic growth the results must be viewed as subject to a sizeable margin of error. There are a number of further reasons why their description of a languishing manufacturing sector, if not incorrect, might be misleading. In the first instance, Crafts and Harley’s estimates for industrial productivity were never based upon a complete analysis of the manufacturing sector. Their indices of industrial production were initially based upon samples of around a dozen different industries, though the size of the sample was slightly extended in subsequent revisions. Harley, ‘British industrialisation before 1841: evidence of slower growth during the industrial revolution’, Journal of Economic History, 42/2 (1982), 267-89; Crafts, British Economic Growth, pp.17-25; N. F. R. Crafts & C. K. Harley, ‘Output Growth and the British Industrial Revolution: A Restatement of the Crafts-Harley View’, Economic History Review, 45/4 (1992), pp. 703-730. Nonetheless, a recent study of industrial output for twenty-six industries for the period from 1815-1850 acknowledges that these industries amount to only about 60 per cent of total industrial production – still leaving fully forty percent entirely out of the account. D. Greasley & L. Oxley, ‘British industrialization, 1815-1860: A disaggregate time-series perspective’, Explorations in Economic History, 37/1 (2000), 98-119, p.101. Most of those excluded were relatively small industries, and it has even been suggested that it was in precisely some of these small, dynamic industries that technological innovation was most pervasive – a claim that Crafts and Harley, unsurprisingly, dispute. See M Berg and P Hudson, ‘Rehabilitating the Industrial Revolution’, Economic History Review, 45/1 (1992), pp.24-50; Crafts & Harley, ‘Output Growth’, pp.711 At any rate, our current macroeconomic estimates do not measure the productivity of these smaller industries, and the absence of such measurement should force us to pause before deciding whether change in manufacturing was localised in the ‘modern’ industries or was in fact more widely spread throughout the sector. Peter Temin has turned to import and export data as an alternative way of gauging the extent and significance of technological change in some of the smaller industries not included in the Crafts-Harley analysis. Peter Temin, ‘Two views of the British industrial revolution’, Journal of Economic History, 57/1 (1997), 63-82. Between a quarter and a half of all manufacturing exports between the late eighteenth and mid nineteenth centuries were of goods other than textiles and iron: they included items such as cutlery, pottery, clothing, glassware, books, umbrellas, hats, and fishing tackles. If the manufacture of these goods was undergoing improvements in productivity, one should expect their export to expand: without these improvements, exports should stagnate or be replaced by imports. Temin’s analysis indicates that exports in such industries were keeping pace with exports in cotton, and he infers from this that technological progress must have occurred within them. This, it should be stressed, is not direct evidence for technological improvements within these smaller, older, industries, any more than Crafts or Harley ever provided direct evidence for its absence. Trade data reveals that Britain enjoyed a comparative advantage in these industries relative to her foreign neighbours, but does not indicate what underpinned this comparative advantage. Nonetheless, the existence of a buoyant export market of so many items outside the lead sectors of cotton and iron sits rather uneasily with the Crafts-Harley account of a stagnant traditional manufacturing sector. See, however, their response to Temin’s analysis. C. K. Harley & N. F. R. Crafts, ‘Simulating the two views of the British Industrial Revolution’, Journal of Economic History, 60/3 (2000), 819-841. The evidence from patenting also paints a rather different picture of manufacturing progress to that depicted by Crafts and Harley. A patent is a grant by the state of exclusive rights for the use of a new invention for a defined period of time (during this period it was set at fourteen years), and the granting of patents has been used by historians as a crude yardstick of inventive activity. The definitive history of patents remains Christine MacLeod, Inventing the Industrial Revolution: The English Patent System, 1660-1800 (Cambridge, 1988). A useful summary may also be found in Kristine Bruland, ‘Industrialisation and technological change‘, in Roderick Floud and Paul Johnson, eds., The Cambridge Economic History of Modern Britain: Industrialisation, 1700-1860, i. (Cambridge, 2004), pp.117-146. Care must be exercised in linking the patent series to technical and industrial change. A patent was expensive and difficult to obtain and some of the industrial revolution’s most significant inventions – James Hargreaves’ spinning jenny and Samuel Crompton’s mule, for example – were never successfully patented, though the omission of such major innovations from the patent series was in fact a rather unusual occurrence. Hargreaves sold a number of jennies before taking out a patent, and for this reason his claim was later rejected in the courts. Crompton was unable to patent his mule, probably because Arkwright’s patent on his ‘throstle’ or ‘frame’ barred the way. See Ashton, The Industrial Revolution, pp.58-60. At the same time as some major inventions slipped through the patenting system, not every invention that was patented signified a fundamental technological breakthrough: whilst a handful were obtained for a radical new invention, many others were obtained for relatively minor improvements to existing techniques, or even simply in an attempt to evade the restrictions of existing patents. Bruland, ‘Industrialisation’, pp.122-3. Furthermore, the mere existence of a patent does not provide evidence that the patented device was ever produced and marketed successfully. More than one patented idea has failed to make the transition from inventor’s workshop to commercially viable product. It is little wonder, therefore, that the historian of patents, Christine MacLeod, has cautioned that the patent series ‘related to technological change in an erratic and tangential manner’. MacLeod, Inventing the Industrial Revolution, p. 157. Despite these caveats, however, a number of clear and very interesting trends emerge from the patenting record. In the first instance, the scale and extent of patenting activity expanded considerably during this period, particularly in the years following 1750. The rate of change accelerated so sharply after 1762 that one historian has suggested that England “entered her ‘Age of Invention’” at this time, defined as a period of self-sustaining growth in technology. Richard J. Sullivan, "England's 'Age of Invention': The Acceleration of Patents and Patentable Invention during the Industrial Revolution," Explorations in Economic History, 26 (1989), pp. 424-52, p.447, p.445. Given the great variety of inventions that underlie the patent series, it is helpful to break it down further, and consider which sectors of the economy were patenting new ideas, and what kinds of inventions they were seeking to protect. An analysis of the series by Richard Sullivan indicates the particular importance of machinery and motive power inventions (steam engines and other devices for transmitting power). Between 1750 and 1850, about a third of all patents concerned machinery and machine parts, whilst motive power accounted for about 7 per cent in the first fifty years rising to 14 per cent in the fifty years thereafter. Sullivan, ‘England’s “Age of Invention”’, p.442. And whilst some of this machinery was for use in the textile industry, taken as a whole, the patent series is not dominated by the cotton industry in the way that productivity figures might suggest. Around fifteen per cent of patents for capital goods between 1750 and 1800 were for textile machines (106 patents in all); and textile machines made up no more than six per cent of all patents issued. MacLeod, Inventing the Industrial Revolution, p. 148. See also Trevor Griffiths, Philip A. Hunt, Patrick K. O'Brien, ‘Inventive Activity in the British Textile Industry, 1700-1800’, Journal of Economic History, 52/4 (1992), pp. 881-906. This left over one and a half thousand patents taken out on a very wide range of inventions and innovations spread across the economy – agriculture, shipbuilding, canals, chemical equipment to name a few. Once again, the evidence from patents sits rather uneasily with the claim that there was little technological change outside the cotton industry before about 1850. The error lies in assuming that measures of productivity are a good indicator of the industrial processes at work in the economy. New technology is expensive to purchase, often prone to failure, and requires new workers to be trained to its use. All these factors mean that manufacturers are often slow to purchase new equipment and likely to wait a considerable period before seeing much return on their outlay. To give one example, James Watt’s separate condenser undeniably improved the fuel efficiency of existing steam engines, yet many manufacturers preferred to continue with their fuel-hungry Newcomen engines, as the cost of purchasing a Watt engine and paying his annual premiums outweighed any fuel savings that could be made, at least before his patent expired in 1800. Michael W. Flinn, The History of the British Coal Industry: 1700-1830: the Industrial Revolution, ii. (Oxford, 1984), pp.124-6; Steven King and Geoffrey Timmins, Making Sense of the Industrial Revolution (Manchester, 2001), pp.84-5. Despite the fact that productivity gains were highly localised in two sectors of the manufacturing economy, the evidence from exports and patenting suggests that technological change was occurring on a much wider basis in the century following 1750. New inventions can be slow to diffuse and measurements of changes in the rate of productivity are unlikely to reflect the underlying changes in industrial techniques. It is one thing, however, to demonstrate that technological change was pervasive and quite another to evaluate its significance in powering the industrial revolution. As the history of earlier great civilisations demonstrates, it is possible to have extensive, and even revolutionary, technological change, without having an industrial revolution. In the case of Britain, industrialisation appeared to turn a switch, marking the end of a period of limited population growth and limited economic expansion and the beginning of an era in which both population and the economy appeared able to grow without limits. Our question concerns the role played by new technologies in turning that switch. Let us turn, then, from assessing the extent of inventive activity to evaluating its wider impact on the British economy. Ever since Toynbee’s popularisation of the term ‘industrial revolution’ in the late nineteenth century, the mechanisation of the cotton industry, and of cotton spinning in particular, has lain at the heart of historical accounts of British industrialisation. It is not difficult to understand why so much significance has been placed on this one industry. In the period 1772-74, England imported 4.2 million lb of raw cotton. By 1839-41, the annual average had risen by an astonishing one hundredfold to 452 million lb. Geoffrey Timmins, Made in Lancashire. A History of Regional Industrialisation (Manchester, 1998), p.85, table 6.1, p.159. Over roughly the same period, the price of cotton cloth dropped by 85 per cent. Mokyr, Lever of Riches, p. 111. See also M. J. Daunton, Progress and Poverty. An Economic and Social History of Britain (Oxford, 1995), pp.186-90 and ‘Statistical Appendix’, table 3.d.i.-ii, pp.586-7. In fact, as this period also witnessed the rapid growth of muslins and fine cotton cloths, requiring less raw cotton to produce, it is likely that cotton output actually increased yet more rapidly than the figures for raw cotton imports suggest. Certainly, by any measure, the manufacture of cotton textiles underwent an astonishing expansion in the century following 1750, experiencing an acceleration of growth that was unmatched in Britain’s earlier industrial history. It is clear that something exceptional was happening in the cotton industry in the late eighteenth and early nineteenth centuries, and there can be no better starting point for considering the importance of technological change to Britain’s industrial revolution. Whilst the myriad changes at work in the cotton industry escape simple classification, we must look to a series of major technological breakthroughs in the eighteenth century, causing the mechanisation of work that had previously been done by hand (and the subsequent movement of work out of the home into the factory), in order to understand the historical development of this one particular industry. Before cotton cloth can be woven, the yarn has to be spun into thread, and it was in this branch of the industry – the spinning industry – that some of the most significant advances were made. During most of the eighteenth century and earlier, cotton yarn had been spun by hand, between thumb and forefinger, at a small wheel turned by hand: it was usually women’s work, and it was generally performed at home. It was a labour intensive process, and so, despite the low wages generally paid to women, a relatively costly task. In the 1760s and 1770s, the process of spinning was revolutionised by a series of inventions: the spinning jenny, the water frame, and the mule, which together replaced the work performed by women’s hands with various mechanical devices. Good accounts of technological change in the spinning industry may be found in: Landes, Unbound Prometheus, pp. 82-86; Geoffrey Timmins, ‘Technological change’, in Mary B. Rose, ed., The Lancashire Cotton Industry. A History since 1700 (Preston, 1996), 29-62, p.44-5; Mokyr, Lever of Riches, pp.96-99; Bruland, ‘Industrialisation’, pp.135-6. James Hargreaves’ spinning jenny replaced the spinner’s one spindle with several (initially eight or sixteen), enabling the machine operator to spin the yarn onto several spindles at the one time, thereby considerably increasing the quantity of yarn that could be spun in a given period of time. Richard Arkwright’s ‘frame’ (or ‘throstle’) produced a stronger thread by using three sets of paired rollers to produce yarn and a set of spindles to twist the fibres together; it was too large to be operated by hand, and after some experimentation was powered by a water wheel instead, thereafter becoming known as the ‘water frame’. Samuel Crompton’s mule combined elements of both the jenny and the water frame to spin strong and good quality cotton thread, which in turn facilitated the weaving of fine cotton cloth on a large scale. The mule required a skilled operator, but Richard Roberts’ ‘self-acting’ mule, patented nearly fifty years later in 1825, made the operator unnecessary, and ushered in the first truly automatic machine. The mechanisation of elements of the process of cotton manufacture that had traditionally been performed by hand enabled industrialists to replace human skill and effort with machines and vastly increased the productivity of the industry within a matter of decades. Whereas a worker spinning cotton on a hand-operated wheel in the middle of the eighteenth century might take more than 50,000 hours to spin 100lb of cotton, by the 1790s the same quantity of cotton might be spun in just 300 hours by mule, and the self-acting mule reduced the figure to 135. Timmins, ‘Technological change’, pp.44-5; and Mokyr, Lever of Riches, p.99. Figure 6.1 provides an illustration of rising imports over the century 1750-1850, and clearly illustrates the great expansion of the industry that occurred during this period of rapid technological change. [insert Figure 6.1. Retained imports of raw cotton, 1750-1810, raw cotton consumption, 1810-1850. Source: Daunton, Progress and Poverty, ‘Statistical Appendix’, table 3.d.i.-ii, pp.586-7. ] [insert images 6.1, 6.2, & 6.3. 6.1 Woman spinning on the one thread wheel. Edward Baines, History of the Cotton Manufacture in Great Britain (London, 1835); 6.2. Arkwright, Hargreaves and Crompton’s spinning machines. John James, History of the Worsted Manufacture in England (London, 1857); 6.3. Mule spinning. Edward Baines, History of the Cotton Manufacture in Great Britain (London, 1835). Layout and text: the three images, 61.-6.3 to appear close together with the following text. 1. Woman spinning on the one thread wheel. A woman using a hand-operated spinning wheel could only spin one thread of yarn at a time. It was a labour intensive, and therefore relatively costly, procedure. 2. Arkwright, Hargreaves and Crompton’s spinning machines. These three inventions – Arkwright’s throstle or frame, Hargreaves’ spinning jenny and Crompton’s mule, which combined elements of the other two – enabled one operator to spin several threads at one time. 3. Mule spinning. This image shows the interior of a large, steam-powered spinning factory, using the self-acting mule, first patented by Richard Roberts in 1825. Here a handful of employers oversee the spinning of hundreds of threads at once. The advantages over the simple spinning wheel are clear.] Whilst the spinning industry witnessed the most rapid increases in productivity in the late eighteenth and early nineteenth centuries, a host of new innovations in the bleaching, dyeing and printing sections of the cotton industry helped to transform these sectors too. In the bleaching industry, for example, the traditional method of open-air bleaching using buttermilk took up to eight months to complete. In the eighteenth century, this method was superseded by new techniques developed in chemical industry, using at first dilute sulphuric acid, and then lime chloride to cut production times from months down to days. Timmins, Made in Lancashire, p.127. See, also, G. N. von Tunzelmann, ‘Time-saving technological change: the cotton industry in the English industrial revolution’, Explorations in Economic History, 32 (1995), 1-27, 11. New washing and drying machines introduced in the early nineteenth century speeded up the bleaching process yet further. In the printing branch, Joseph Bell’s mechanised copper rollers replaced the older method of block printing by hand. One contemporary estimate suggested that a cylinder printing machine operated by a man and boy could do the same work as a hundred block printers, each with a boy to assist. Timmins, Made in Lancashire, p.128. These innovations led to far-reaching changes in these industries, driving up both output and productivity, and changing the working patterns of the thousands of men, women and children employed in them. [Insert image 6.4 Calico printing. Cylinder printing using copper rollers. According to one contemporary: ‘The saving of labour … is immense: one of the cylinder printing machines, attended by a man and a boy, is actually capable of producing as much work as could be turned out by one hundred block printers and as many tear-boys! In consequence of the wonderful facility given to the operation, three-fourths of all the prints executed in this country are printed by the cylinder machine’ (Baines, History of Cotton Manufacture, p.266.) The cotton industry was in the van of technological progress, yet it is important to note even here the incomplete nature of technological change. Although productivity in the spinning and finishing branches of the cotton industry was quickly pushed up through the use of new powered machinery, change occurred much more slowly in the intermediate stage: weaving. With the advantages of powered machinery so evident in the spinning industry, cotton manufacturers held high hopes that machines might also replace human labour in the field of weaving and a series of power looms invented by Edmund Cartwright in the 1780s, William Horrocks (1803), and Sharp and Roberts (1822) appeared to herald the realisation of these hopes. In 1825, the Manchester Chamber of Commerce declared that the new power loom brought ‘the whole process of manufacture, from the raw material to the cloth, into one connected series of operations, by means of which, a cheaper, more uniform and better fabric has been produced’. Roger Lloyd-Jones and M. J. Lewis, British Industrial Capitalism since the Industrial Revolution (London, 1998), p.41 But the reality was rather different from this hopeful vision. Cartright’s and Horrocks’ power looms did not work properly, and although Sharp and Roberts power loom marked a noticeable improvement on its predecessors, even it could not weave fine or weak threads. Timmins, The Last Shift. The Decline of Handloom Weaving in Nineteenth-century Lancashire (Manchester, 1993), pp.157-9. Furthermore, early power looms required very close attention from the operator as they needed to be stopped as soon as a thread broke or the shuttle became empty. Most weavers could only operate one, or at best two, power looms at a time, so gains in productivity were less significant than their purchasers might have wished. Ibid., p.159 and Timmins, ‘Technological change’, pp. 45-7. Given the difficulties that manufacturers faced in developing an effective automated weaving machine, inventive activity continued to be focussed upon creating a more efficient hand-loom. William Radcliffe’s ‘dandy loom’, which enabled the woven cloth to be wound automatically onto a beam at the back of the loom marked a significant improvement on the existing handloom: raising the hand weaver’s productivity by as much as fifty per cent, it helped to sustain handloom weaving in the first half of the nineteenth century. Timmins, Made in Lancashire, p.130, p.170 In the mid 1830s, there were about twice as many weavers operating a variety of different handlooms as there were power-looms. Ibid., p.87 Determining the relative importance of the hand-powered branch of the industry on the one hand and of the water- or steam-powered branches on the other is difficult: the output of power looms was certainly greater than that of the handlooms, but the handloom weavers produced higher quality cloths with greater profit margins. Wherever the balance lies, it is certainly the case that the weaving trade was only partially mechanised before about 1830, and that even so late as 1850, the handloom weavers made up a sizeable minority of the total weaving workforce. Idem., The Last Shift, pp.108-118. Furthermore, despite the undoubted technological advances in some branches of the cotton industry, the endpoint of cloth manufacture – the turning of manufactured cloth into clothes, hats and accessories – was largely done by hand until the invention of the sewing machine in 1860. Mokyr, ‘Editor’s introduction’, p.12 Even in the cotton industry, then, where new technology undoubtedly ushered in some phenomenal advances, the technological revolution was not completed by the middle of the nineteenth century. Yet taken as a whole, the cotton industry, and the spinning industry in particular, demonstrate how powerful invention and innovation can be as a source of economic change and growth. Cotton manufacture industry combined a handful of path-breaking inventions with countless minor adjustments and modifications to existing processes to sharply cut the cost of producing cotton textiles. The mix of radically new inventions and small adaptations to existing machines underpinned explosive industrial growth of a kind that Britain had never seen before. Outside the cotton industry technical change proceeded far more slowly and with rather less spectacular results, but it is nonetheless possible to identify other industries that were revolutionised, at least to some degree, by the emergence of new technologies. The iron industry, for example, was the site of significant technological progress, and whilst this did not lead to growth rates as impressive as those of the cotton industry, it did help to revolutionise both this industry and other parts of the British economy. The production of wrought iron depends on two processes. First the raw iron ore is smelted in a blast furnace to produce pig iron. Owing to its high carbon content, pig iron is hard and brittle; it can be cast items such as into pots, ovens and cannon, but its uses are rather limited. The soft, malleable wrought or bar iron, which can be fashioned into nails, locks, tools, cutlery, horse shoes, machine parts, railway tracks, and countless other items has a far wider range of uses, so pig iron is therefore put through a second process, in order to refine it into wrought iron. Daunton, Progress and Poverty, pp.211-19; Mokyr, Lever of Riches, pp.93-5; J. R. Harris, The British Iron Industry, 1700-1850 (London, 1988), pp.30-40; Richard Hayman, Ironmaking: The History and Archaeology of the Iron Industry (Stroud, 2005), pp.34-63. New technologies significantly improved both of these elements of iron manufacture – smelting and refining – during the eighteenth century. Smelting was transformed by the replacement of charcoal with coke and by the development of the ‘hot blast’ furnace, which used the furnace’s own gases to heat the air inside. Refining was improved first by the Woods Brothers ‘potting and stamping’ process, and soon after by Henry Cort’s ‘puddling and rolling’ process, patented in 1783 and 1784. The work of refining pig iron had traditionally been performed by skilled ironworkers, who repeatedly heated and hammered the pig iron in small forges to beat out the impurities. Cort’s procedure used iron rods to stir and beat impurities out of the molten pig iron, and then passed what was left between iron rollers to press the final impurities away. The technique heralded the end of small forges and helped to produce a more uniform and cheaper product. These various improvements to iron manufacture helped to raise the output of wrought iron by an average of 4.5 per cent per year in the first half of the nineteenth century and led to steep reductions in its cost. Joel Mokyr, ‘Technological Change, 1700-1830’, in R Floud and D McCloskey, eds., The Economic History of Britain since 1700 (Cambridge, 2nd edn. 1994), pp.25-7; Roger Burt, ‘The extractive industries’ in Floud and Johnson, eds., Cambridge Economic History, p.448-9. Figure 6.2 provides an illustration of rising output over the century 1750-1850, and once again indicates that technological improvement during the period went hand in hand with a significant increase in output. [Insert figure 6.2 Pig-iron output, 1750-1850. Source: Daunton, Progress and Poverty, ‘Statistical Appendix’, table 3.c, p585.] In many respects, the changes that occurred in the iron industry were less impressive than those that occurred in cotton. In iron refining, for example, Cort’s puddling process was vital in increasing the production of wrought iron, but various rolling techniques had in fact existed for centuries, and as the foremost historian of technology, Joel Mokyr, has noted, ‘the conceptual novelty of the process was modest’. Mokyr, ‘Editor’s introduction’, p.22. Furthermore, improvements in smelting and refining iron ore, though they certainly led to a significant expansion of the industry, did not cause growth of the magnitude seen in the cotton industry. Whether measured in terms of its workforce, its output, or its growth rate, the achievements of the iron industry are overshadowed by those of cotton. Landes, Unbound Prometheus, p. 89; Nonetheless, the technical achievements of the iron industry are perceived by many as a cornerstone of the industrial revolution, as they made available a raw material – wrought iron – for which no substitute existed. Cotton textiles could be, and were, produced by hand. New spinning, weaving, bleaching and printing technologies all replicated, with greater or lesser success, processes that had previously been performed by hand: by improving production processes they radically lowered the cost of cotton cloth, but they changed the product in only minor ways. By contrast, there was no alternative to cheap, wrought iron, and as its price dropped it began gradually to replace the wood in bridges, ships, buildings, and machinery, and of course, it enabled the construction of new inventions, such as the steam engine and the railways. Indeed, so numerous were the uses of wrought iron that Cort’s puddling process has been singled out by one historian as ‘a crucial invention which made the industrial revolution possible’. P. Deane, The First Industrial Revolution (Cambridge, 1965), p. 130. In their different ways, the cotton and iron industries both lend support to the claim that technological change was the driving force behind the industrial revolution. Elsewhere, however, it is more difficult to demonstrate the primacy of new technologies in powering British industrialisation. The difficulty lies not in identifying inventive activity, which was present, to some degree, in almost every section of the manufacturing economy; but rather in evaluating the overall significance of this inventive activity. Nowhere are the difficulties of placing new technologies at the heart of the industrial revolution more apparent than with the example of the steam engine. Engines of various kinds had been available since the late seventeenth century, and their numbers had grown since the invention of Thomas Newcomen’s self-acting atmospheric engine in the early eighteenth century – the first engine properly to use steam to operate machinery. But Newcomen’s engine was large and noisy, operated in a jerky motion, and had a voracious appetite for fuel, all of which effectively limited its use to pumping water from mines, where the size and noise of the engine posed few difficulties and fuel was plentiful. As a result, most other industrial processes continued to be powered by other means: by wheels driven by water or by horses; or by small machines operated by hand or by foot. The value of steam power to manufacturing industry was greatly enhanced by James Watt’s realisation that the two phases of the engine’s cycle – the heating and cooling – could be separated. This enabled him to create an engine in the 1760s that was considerably more fuel efficient (four times so) and that ran with a smoother motion than its predecessors. Mokyr, Lever of Riches. Both factors contributed to its early adoption in the mining industry and helped to facilitate its spread to the textile industry in the early nineteenth century. Timmins, Made in Lancashire, p.131. By the 1830s steam had largely replaced the waterwheels that had powered the cotton industry through most of the eighteenth century; it was indeed an integral part of the technical revolution that occurred in that industry. Outside the mining and cotton industries, however, steam power penetrated far more slowly, and its contribution to the industrial revolution is far less clear. The mining and textile counties of Cornwall, Durham, Lancashire, Northumberland, Shropshire, Staffordshire and Yorkshire had between them well in excess of a 1,000 engines by the end of the eighteenth century, but several others – for example, Bedfordshire, Dorset, Hertfordshire, Suffolk, Sussex, and Wiltshire – had not a single one. See J. Kanefsky and J Robey, ‘Steam engines in eighteenth-century Britain: a quantitative assessment’, Technology and Culture, 21 (1980), pp.161-86, table 5, p.176. The historian of the west Midlands, where steam certainly did penetrate in the early nineteenth century, has nonetheless concluded that ‘many of the midland manufactures had no use for such a large measure of power which could not easily be turned on and off. For the majority of processes and in many works the traditional power of wind, water, man and animals continued to be used not only because they were cheaper but also because they were more appropriate and efficient in the particular context’. Marie Rowlands, The West Midlands from AD 1000 (Harlow, 1987), p.236. In the 1970s, von Tunzelmann set out to put some figures to the importance of steam engine by counterfactual analysis, or the ‘social savings method’, which seeks to measure the economic contribution of an innovation by calculating the saving in costs compared with the earlier alternative technology. His assessment of the number and use of steam engines based upon Watt’s ideas in the early nineteenth century reveals that had they never existed, national income around 1800 might have been lower by about 0.11%; without steam engines of any kind, it would have been 0.2 per cent lower – clearly very negligible quantities. G. N. Von Tunzelmann, Steam Power and British Industrialisation to 1860 (Oxford, 1978), p.286-7. These figures have received more recent confirmation from Nick Crafts. His calculations suggest that steam’s contribution to improvements in labour productivity was never more than 0.02 per cent per year prior to 1830, rising to 0.2 per cent per year over the next two decades; the contribution of steam power to labour productivity growth, he concludes, was trivial before 1830. Nicholas Crafts, ‘The first industrial revolution: Resolving the slow growth/rapid industrialization paradox’, Journal of the European Economic Association, vol. 3 no.2/3 (2005), pp.525–534 p. 528. Inevitably, measuring something so complex as the contribution of steam power to labour productivity in the early nineteenth century cannot be done with any great precision and the exact figures that von Tunzelmann and Crafts have provided should be taken with a pinch of salt. Nonetheless, the broad outlines of their estimates fit neatly with a wide range of qualitative studies, and we might readily concur with Patrick O’Brien that the ‘”age of steam” … remained imminent rather than dominant during the first stages of the industrial revolution’. P. K. O’Brien ‘The deconstruction of myths and reconstruction of metanarratives in global histories of material progress’, in Benedikt Stuchtey and Eckhardt Fuchs, Writing World History (Oxford, 2002). Yet the history of the steam engine also poses some difficulties for accounts that place technology at the base of the industrial revolution. The evidence concerning national economic growth rates and population growth and movement considered in the previous three chapters all points towards deep-seated change dating from as early as 1700; by 1800 Britain’s economy and population appear to be following a trajectory different to that of both the rest of Europe, and that of its own the past. Yet steam power had no real impact until at least the 1830s, and prior to then Britain’s economy was powered largely by its existing mix of waterwheels and hand power – indeed the period 1760-1830 has been labelled the ‘Age of Water Power’ by one historian. A. E. Musson, The Growth of British Industry (London, 1978), p.109. It is not simply, then, that the technology was slow to diffuse, as that so much was achieved without it. We will return to this problem in the next chapter where we consider coal – that vital fuel that made steam power possible. For the present, however, it must be emphasised, that the steam engine, one of the most significant technological inventions of the entire period, does not fit within the timeframe of our traditional narratives of industrialisation. Von Tunzelmann’s demonstration that Britain’s early industrialisation proceeded largely without the benefit of steam power has encouraged a generation of historians to turn attention away from the dramatic technological breakthrough and to emphasise in addition the significance of small scale invention and of successive improvements and modifications to existing techniques. These incremental innovations are sometimes termed ‘micro-inventions’ in contrast to ‘macro-inventions’ – the much rarer, groundbreaking inventions that open the possibility of performing tasks in an entirely new way. The distinction is helpfully described in Mokyr, ‘Editors introduction’, pp.17-24. By expanding our framework for technological advance, it is possible to develop a much broader view of technological progress during the industrial revolution. In this vein, hand power is not viewed simply as the poor cousin of the steam engine, but as a viable alternative equally capable of technological improvement. Early steam engines were not well adapted for many manufacturing processes and machines or tools powered by human muscles offered a vital degree of precision or dexterity with which steam powered machinery could not compete. We have already seen how this was the case with cotton weaving: the power looms could weave large quantities of coarse cloth but could not weave fine and fancy cloths as effectively as the handloom. Such examples can be multiplied endlessly. In ship-building, for example, timber needed to be sawed with an accuracy that machinery was as yet unable to impart, so hand-operated saws remained the mainstay of the industry. As one commentator explained: ‘it might at first thought be imagined that machine-worked saws would be used; but the curvatures and angles of the timber are so extremely varied, not only in different timbers, but also in different parts of the same timber, that the precision and regularity of machinery would here be thrown away, and indeed unavailable’. Quoted in King and Timmins, Making Sense, p. 75. And of course, hand tools and innovation are not exclusive. On the contrary, hand tools could be, and were, improved by incremental technological progress, in ways that were more or less significant. Raphael Samuel, ‘Workshop of the world: steam power and hand technology in mid Victorian Britain’, History Workshop Journal, 3 (1977), pp. 6-72. Once again, we have already described one such example from the cotton industry – the dandy loom – but countless other examples abound. Glassmaking, for example, was improved by the cylinder method adopted by Chance Brothers at their Birmingham works in the 1830s. The cylinder method enabled the production of larger panes of glass than existing methods, yet it still remained a hand technique performed by a skilled workman rather than a machine. King and Timmins, Making Sense, pp.71-2. Likewise, this more expansive approach to the role of technology has encouraged historians to look beyond machines and industrial processes and to focus attention on the finished product instead. This reminds us that part of the inventive process lay in providing attractive and desirable goods to consumers: new fabrics, coloured and patterned cloths, shiny buttons, cheap buckles, affordable tableware – the list of new products designed to appeal to the changing tastes of consumers is endless. One study of the cotton industry has suggested that inventive energies were equally divided between devices aimed at reducing the costs of labour and raw materials on the one hand and product innovation – improvements to the nature and appearance of the finished cloth – on the other. Trevor Griffiths, Philip A. Hunt, Patrick K. O'Brien, ‘Inventive Activity in the British Textile Industry, 1700-1800’, Journal of Economic History, 52/ 4 (1992), pp. 881-906, p.892. Similarly, Maxine Berg’s study of patents in the metal-wares, glass, ceramics, furniture, clocks and watches demonstrates that one quarter of the patents specified new products, improvements in ornamenting and finishing, or imitations of existing goods. M. Berg, ‘From imitation to invention: creating commodities in eighteenth-century Britain’, Economic History Review, 55/1 (2002), pp. 1-30. These improvements involved inventions such as Keen and Schmidt’s method of binding gold and silver to woollen cloth, John Peele’s method of printing images onto linen handkerchiefs, and Henry Clay’s manufacture of black, lacquered buttons. Griffiths, Hunt and O’Brien, ‘Inventive activity’, p.895; Berg, ‘Imitation to invention’, p.23. See also, more generally, Maxine Berg, Luxury and Pleasure in eighteenth-century Britain (Oxford, 2005). It is doubtless important to emphasise that hand powered technology was more than a simple precursor to steam and that it was often the site of significant improvement and innovation. It is also helpful to shift attention away from machines altogether and consider the range and extent of improvements to goods and products as well. Yet placing hand technology and product improvement at the heart of the industrial revolution would arguably be placing more weight on these small scale improvements than they can really bear. No matter how much hand technology was refined and improved, it still remained hand technology, an essentially older way of doing things. And whatever inventiveness was displayed in designing more attractive consumer products, it is also possible to remain unconvinced about the wider significance of new techniques to decorate cloth with gold, print images on handkerchiefs, and manufacture lacquered buttons. Enterprising producers had always sought to improve their existing methods of production and searched for new ways to make their goods more attractive and it is likely that during the eighteenth century they became more inventive in their quest to do so. For our purposes, however, the critical question remains what role this played in promoting British industrialisation. By any measure, the industrial revolution marked a great discontinuity with the past, and however numerous and pervasive micro-inventions and product improvements may have been, it is difficult to see how they could have played more than a small role in forcing this momentous transition. If extensive, yet ultimately small-scale, invention seems insufficient to account for the industrial revolution, it is also interesting to consider the example of towns and regions that managed to industrialise without a corresponding leap in inventive activity. Whilst the fit between technology and industrialisation is convincing in the case of Lancashire, the industrial history of many other towns and regions demonstrates the complex relationship that existed between technical innovations on the one hand and the process of industrialisation on the other. Consider, for example, the history of Birmingham. As we saw in the previous chapter, Birmingham had grown to become England’s fourth city in 1800, largely as a consequence of in-migration. Its population of 6,000 in 1700 was roughly equivalent to that of Canterbury, Cambridge, Salisbury and Hull; standing at about 74,000 a century later, it had increased more than tenfold, vastly outstripping the growth of most other towns of its original size. Wrigley, ‘Urban growth’, p.686. Yet it is difficult to credit new technologies with a major role in this transformation from middle-ranking regional centre to great city. The economy of Birmingham had been dominated by metal work since the sixteenth century at least, owing in large part to its proximity to iron ore and coal in Staffordshire and Worcestershire. Eric Hopkins, The Rise of the Manufacturing town. Birmingham and the Industrial Revolution (London, 1989; repr. Stroud 1998), pp. 3-4. In the middle of the eighteenth century, three metal industries – gun-making, brass, and ‘toys’ (small articles of brass and iron, such as buttons, buckles, pins and trinkets) – stood out in importance. All three branches developed and expanded considerably in the century that followed, yet groundbreaking, macro-inventions are perhaps most conspicuous here by their absence. In the gun-making trade, there were only two technological developments of importance: the use of water- or steam-powered rollers to produce gun-barrels; and the invention of the percussion cap, which replaced the flint-lock for firing the gun. Most other elements of gun-making remained largely unchanged, and as the historian of Birmingham has observed, this period witnessed ‘no fundamental change in working techniques for the majority of gun-trade workers’. Ibid., p.42. Likewise, although there were a number of improvements to the making of buttons, there was no clear technological break with the past in any branch of the toy industry. Ibid., pp.48-51. Only the brass industry saw significant technical improvements: the crucible method for making brass and the production of seamless brass tubes, though these innovations came in the 1830s, rather late in the history of Birmingham’s industrial and population growth. Ibid., pp.46-8. None of this should be taken to suggest that innovation was unimportant to Birmingham’s metalworking trades. Cumulatively, myriad minor improvements helped to increase output, reduce costs, and raise profits. Nonetheless, Birmingham remained firmly in the world of skilled workmen operating their handheld tools in small workshops well into the nineteenth century. No matter how broadly we define the process of technical change, the overall impact of new technologies in Birmingham was relatively modest, and this town therefore offers us a path to industrialisation that did not involve major technical change. [Insert images 6.5. & 6.6 depicting gun-making in Birmingham -- 6.5. Welding the gun-barrels, Birmingham. Illustrated London News, 1851; 6.6. Grinding the gun-barrels, Birmingham. Illustrated London News, 1851 These two engravings illustrate the traditional nature of gun-manufacture, Birmingham’s foremost industry. Skilled welders heat the barrel in a furnace and use hammers to mould the heated iron into a perfect tube. The grinders’ task is facilitated by the use of a large, steam-powered grind-stone, yet even with the addition of steam-power, each man grinds just one barrel at a time and needs to exercise some skill to achieve a smoothly finished barrel.] Birmingham certainly stands out for the rapid growth it achieved on the basis of only minor (albeit numerous) innovations, but it was by no means unique in its marriage of industrial expansion and the continued use of older production techniques. The towns and industrial villages to the north and west of Birmingham that made up the Black Country were largely devoted to simpler forms of metalworking than were to be found in the regional capital, and they too developed largely on the basis of existing techniques. The Staffordshire pottery industry provides another well known example of an industry which expanded without the benefit of spectacular technological breakthroughs. Mokyr, ‘Technological change’, p.28. A similar pattern of growth occurred in the hosiery (knitting) industry based in Nottinghamshire, Leicestershire and Derbyshire. The ‘stocking frame’ that formed the mainstay of the industry had been invented in the late sixteenth century. It roughly imitated the work of hand knitting and was used for the production of stockings and other knitted goods, such as waistcoats and gloves. In what must now be a familiar pattern, the process of framework knitting was improved throughout the eighteenth and early nineteenth centuries by a series of inventions. Attachments that could be added to a stocking frame, such as those patented by Jedediah Strutt in 1758 and John Morris in 1763, enabled its operator to knit a more elaborate and attractive stitch. Furthermore, the knitting industry responded to changes in fashions for its goods by countless innovations in the range of goods produced, offering new types of fabrics and knitted garments – velvets and brocades for waistcoats, ladies’ silk mitts and woollen polka jackets, lambswool drawers, and zigzag patterned stockings, to name a few. Stanley Chapman, Hosiery and Knitwear: Four Centuries of Small-scale Industry in Britain, c.1589-2000 (Oxford, 2002), pp. 1-11, 65-71, 105-11. Despite these advances, however, down to 1850s, most of those employed by the knitting industry were working at frames that were fundamentally the same as those in operation two centuries earlier, and the industry’s historian has concluded that hosiery provides ‘an interesting case of an industry whose structure and organisation barely changed in the century 1750 to 1850’. Ibid., p.52. No matter how widely we interpret technological change, it seems we are left with something less than an industrial revolution. This is not to deny the extent of inventive activity. The century following 1750 was undoubtedly a period of great innovation, involving both a small number of groundbreaking inventions, as well as a far larger number of small improvements, adjustments, and modifications to existing techniques; furthermore, this inventive process revolutionised some aspects of British manufacturing. In the cotton industry, a series of radical inventions vastly speeded up production processes, leading to an increase in output and reduction in price on an unprecedented scale. In metalworking, inventions helped to create a new product – wrought iron – at an affordable price, which in turn facilitated the creation and growth of many related industries and spawned a radically new form of transport – the railways. Yet in other spheres of the economy, the overall contribution of technology is less evident. The steam engine, for example, was novel in conception, but it was not widely adopted until well into the nineteenth century, well after the date at which the industrial revolution was conventionally thought to have begun. And the countless micro-inventions of this period, though they reveal an inventive spirit at work at the heart of the British economy and certainly helped contribute to economic expansion, nevertheless remained small in scale and their significance should not be overstated. Finally, there are regions such as Birmingham and parts of Staffordshire, Nottinghamshire, Leicestershire and Derbyshire which underwent significant industrial growth yet without much in the way of a technical revolution at all. In fine, the emergence of new technology provides no more than a partial explanation for Britain’s industrial revolution. If we are to provide a full account of what caused this turning point in British history it will be necessary to look beyond the machines and technology that so vividly captured the imagination of many Victorian commentators. PAGE 128