Business History Review 87 (Spring 2013): 95–120. doi:10.1017/S0007680513000159 © 2013 The President and Fellows of Harvard College. ISSN 0007-6805; 2044-768X (Web). I Alessandro Nuvolari and James Sumner Inventors, Patents, and Inventive Activities in the English Brewing Industry, 1634–1850 This article examines the relationship between patents, appro- priability strategies, and market for technology in the English brewing industry before 1850. Previous research has pointed to the apparent paradox that large-scale brewing in this pe- riod showed both a self-aware culture of rapid technological innovation and a remarkably low propensity to patent. Our study records how brewery innovators pursued a wide variety of highly distinct appropriability strategies, including secrecy, selective revealing, open innovation and knowledge-sharing for reputational reasons, and patenting. All these strategies could co-exist, although some brewery insiders maintained a suspicion of the promoters of patent technologies, which faded only in the nineteenth century. Furthermore, we fi nd evidence that sophisticated strategies of selective revealing could support trade in inventions even without the use of the patent system. n traditional narratives of the Industrial Revolution, the brewing in- dustry occupies a marginal role; yet large-scale commercial brewing in England underwent rapid technical change across the eighteenth and into the early nineteenth century. To some extent, this marginalization is a simple consequence of the industry’s expansion taking place at a lei- surely pace: from 1700 to 1830, the average growth of real output was about 0.5 percent per year, far lower than in the most celebrated “lead- ing sectors” such as cotton or iron.1 These fi gures, however, conceal a remarkable concentration of production that Peter Mathias, in his monumental 1959 study of English brewing in this period, highlighted as revolutionary.2 The technical element of this revolution, Mathias wrote, consisted of two interlinked trends: the adoption of mechanized 1 Peter Mathias, The Brewing Industry in England, 1700–1830 (Cambridge, UK, 1959), 542–43. 2 Ibid., 12–21. Alessandro Nuvolari and James Sumner / 96 equipment for large-scale production and process monitoring using quantitative instruments such as the thermometer and saccharometer. This vigorous inventive culture, however, did not lead to growth in the use of patents to protect inventions. Indeed, Christine MacLeod, in her authoritative 1988 study of the English patent system during the eighteenth century, considers brewing as the archetype of a sector in which inventive activities went on outside the patent system. Certainly, some brewing-related patents were fi led, yet they were remarkably in- frequent given the concentration of the market and the technical nature of the process.3 Before the institution of the Patent Offi ce in 1852, se- curing a patent involved time-consuming bureaucratic procedures and large fees (approximately £100 for England and Wales), and the out- comes of litigation were notoriously unpredictable. Therefore, some economic historians have suggested, patenting activity was skewed to- wards capital-intensive inventions resulting in large pieces of equip- ment, for which royalties were easier to collect, cases of infringement were more likely to be detected, and evidence could more easily be pre- sented in court.4 Since the largest and most highly mechanized brewer- ies displayed not only conscious technological prowess, but also a very high degree of capital intensity—being, in fact, among the largest man- ufacturing concerns of any description—the unpopularity of patents may appear altogether perplexing. Part of our aim in this article is to scrutinize this apparent anomaly in the light of recent broader interpretations of the functioning of pat- ent systems. Some scholars have noted that the establishment of rela- tively secure, formalized property rights on innovations, such as those provided by a patent regime, represents a foundational prerequisite for the rise and consolidation of “markets for technologies.”5 The advan- tage of a reliably functioning market for technology, the argument runs, is that it enhances the effi cient use of technological knowledge through- out the economy. In particular, it encourages the emergence of a fruit- ful division of innovative labor in the market between, on the one hand, fi rms and individual inventors specialized in inventive activities, and, on the other, fi rms and entrepreneurs with advanced capabilities in the development and commercialization of innovations. 3 Christine MacLeod, Inventing the Industrial Revolution: The English Patent System, 1660–1800 (Cambridge, UK, 1988), 108–9. 4 Ibid, 147–49; B. Zorina Khan and Kenneth Sokoloff, “Patent Institutions, Industrial Or- ganization, and Early Technological Change: Britain and the United States, 1790–1850,” in Technological Revolutions in Europe: Historical Perspectives, ed. Maxine Berg and Kristine Bruland (Cheltenham, 1998), 292–313. 5 Ashish Arora, Andrea Fosfuri, and Alfonso Gambardella, Markets for Technology: The Economics of Innovation and Corporate Strategy (Cambridge, MA, 2001). Patents in the English Brewing Industry, 1634–1850 / 97 From this perspective, then, the advantage of the patent system lies not only in the direct incentive to individual inventors to invest in in- ventive activities, but in the emergence of a general commercial and social context that makes invention and innovation more convenient. The fi ndings of Sokoloff and Lamoreaux (1999) on the culture of late nineteenth-century US inventors have stimulated these considerations. They suggest that the American patent system, with its comparatively low fees, limited bureaucracy, and effi cient administration through the courts, allowed “creative but ambitious inventors to focus on what they did best,” leading to “the most technologically fertile period in Ameri- can history.”6 Comparative work by Khan and Sokoloff (1998) pre- sents the English system as much less effective in this regard.7 The ear- lier investigations of Harold Dutton (1984), however, show that even the pre-1852 regime in England hosted an identifi able class of “quasi- professional inventors”: individuals who specialized in inventive activi- ties broadly considered, taking multiple patents relating to different in- dustries.8 Dutton’s survey also documents a signifi cant number of cases in which patent rights were traded by means of licenses or assign- ments.9 It appears, then, that even the unreformed English system, frus- trating as it often was to inventors and investors, functioned well enough to sustain emerging markets in invention. More recent contributions in the economics of innovation literature have pointed out that inventors may be able to exploit their inventions commercially in situations of very weak patents or even in the absence of patent rights.10 6 Naomi Lamoreaux and Kenneth Sokoloff, “Inventors, Firms, and the Market for Tech- nology in the Late Nineteenth and Early Twentieth Centuries,” in Learning by Doing in Mar- kets, Firms and Countries, ed. Naomi Lamoreaux, Daniel M. G. Raff, and Peter Temin (Chi- cago, 1999); Naomi Lamoreaux and Kenneth Sokoloff, “Market Trade in Patents and the Rise of a Class of Specialized Inventors in the Nineteenth-Century United States,” American Eco- nomic Review 91 (2001): 39–44; Naomi Lamoreaux and Kenneth Sokoloff, “Intermediaries in the US Market for Technology, 1870–1920,” NBER Working Paper no. 9017 (2002), quo- tation on 5; and see also B. Zorina Khan, The Democratization of Invention: Patents and Copyrights in American Economic Development, 1790–1920 (Cambridge, UK, 2005); Arora, Fosfuri and Gambardella, Markets for Technology, discuss the contributions of Lamoreaux and Sokoloff on pp. 23–27. 7 Khan and Sokoloff, “Patent Institutions,” 309–13; Khan, Democratization, 37–39. 8 Harold Dutton, The Patent System and Inventive Activity during the Industrial Revo- lution, 1750–1852 (Manchester, 1984), 112–17. 9 Dutton, Patent System, 122–42. See also Sean Bottomley, “The British Patent System during the Industrial Revolution, 1700–1852,” PhD diss., University of Cambridge, 2012, 101–54. 10 For a general discussion of “appropriability strategies” for innovations in different busi- ness contexts, including weak intellectual property rights, see David Teece, “Profi ting from Technological Innovation: Implications for Integration, Collaboration, Licensing, and Public Policy,” Research Policy 15 (1986): 285–305; James Anton and Dennis Yao, “Expropriation and Inventions: Appropriable Rents in the Absence of Property Rights,” American Economic Review 84 (1994): 190–209, provides a theoretical discussion of effective appropriability strat- egies in the absence of intellectual property rights. The historical relevance of “appropriability Alessandro Nuvolari and James Sumner / 98 In what follows, we present a fresh look at the connection between patenting and inventive activities in the English brewing industry, fo- cusing on the extent to which appropriability strategies not based on patents could be effective both for protecting innovations and for trad- ing them in “markets for technologies.” We begin with a quantitative survey of brewery-related patents in the period in question. This serves to confi rm the low incidence that prior authors have noted, but it also highlights the complexities of assigning particular inventions to partic- ular industries. Next, we survey the various alternative appropriability strategies used in brewing, before discussing how early patents were promoted, chiefl y by outsiders; and how, ultimately, such patents began to be accepted in the trade. Patents in the Brewing Industry, 1634–1850 Two principal sources, if carefully interpreted, offer a convenient means to identify patents relevant to the brewing industry. The fi rst derives from the Patent Offi ce itself. In 1854, the Superintendent of Specifi cations, Bennet Woodcroft, oversaw the publication of a Subject- Matter Index that attempted to categorize all identifi able patents ac- cording to the trades or industries to which they referred. The category for “Brewing, distilling, rectifying and preparatory processes” is further divided into subcategories including malt preparation, mashing and wort management, clarifi cation of drink, and the more wide-ranging “Brew- ing and fermenting: malt liquors.”11 Woodcroft’s index gives only the ti- tles of patents, but in the 1860s the Patent Offi ce began a long-running project to publish abridgements of specifi cations, again classifi ed by in- dustry.12 Class 99 in this series, covering “brewing, wine-making, and distilling alcoholic liquids” in the period 1634 to 1866, appeared in 1881 under the supervision of Woodcroft’s successor, Henry Reader Lack.13 The second major source is the practitioners’ literature on brewing, strategies” alternative to patents during the British Industrial Revolution has also been dis- cussed by David Landes, “What Do Bosses Really Do?” Journal of Economic History 46 (1986): 585–623. In particular, Landes points to “fi rst mover advantage” as a particularly ef- fective strategy in that historical context. For a discussion of the “fi rst mover advantage” lit- erature, see Fernando Suarez and Gianvito Lanzolla, “The Role of Environmental Dynamics in Building a First Mover Advantage Theory,” Academy of Management Review 32 (2007): 377–92. 11 Bennet Woodcroft, Subject-Matter Index of Patents of Invention, vol. 1 (London, 1854), 85–97. 12 John Hewish, Rooms Near Chancery Lane: The Patent Offi ce under the Commission- ers, 1852–1883 (London, 2000), 41–44. 13 H. Reader Lack, Patents of Inventions: Abridgments of Specifi cations Related to Brew- ing, Wine-making, and Distilling Alcoholic Liquids (London, 1881). Patents in the English Brewing Industry, 1634–1850 / 99 which by the 1860s was becoming increasingly systematic and technically oriented. The monthly Brewers’ Journal, launched in 1865, featured a regular “Brewer’s Engineer” column with descriptions and engravings of recent inventions, patented and otherwise; in 1869, the Journal seri- alized a listing of “every patent from 1635 to 1868” under classifi ed head- ings, a project that was updated regularly in later years.14 The Journal and its rival the Brewers’ Guardian were aimed at, and largely written by, self-consciously “progressive” and “scientifi c” brewers, to whom it was important to affi rm a technical history for the trade. This agenda culminated in the work of Henry Stopes (1852–1902), who in the 1880s established himself both as a specialist brewers’ engineer and as an an- alyst of the malting and brewing trades, contributing historical surveys to the Journal and other periodicals.15 In 1885, alongside an epic bibli- ography of malting and brewing literature, Stopes published a consoli- dated “List of Patents” since 1634 that he deemed relevant to the indus- try.16 Of the 176 patents listed by Stopes in the period up to 1850, 174 also appear in the 1881 Patent Offi ce volume.17 Careful analysis of the Woodcroft, Lack, and Stopes lists, however, shows that the defi nition of a “brewing patent” is complicated on sev- eral levels. To begin with, a clear distinction between brewing and dis- tilling inventions is frustrated by the fact that several operations are common to both. These include the heating of water in bulk, the crush- ing or grinding of malt, and the “mashing” (infusion in hot water, usu- ally with mechanized stirring) that the malt undergoes to produce fer- mentable wort. In some cases, such as the mashing machine patented in 1793 by Edward Biley (himself a brewer), external evidence shows that the patentee was purely or primarily concerned with the beer- brewing market.18 In other patents, such as one that James Tate, a Lon- don ironmonger, took out in 1794 for an “invention of applying fi re to the coppers [boilers] of brewers and distillers,” the reference to brewing may be entirely speculative. A further problem arises because there was no consensus on whether inventions in malting should be considered a subset of those in brewing. Though some brewers, at all levels of production, had always made their own malt, the two processes were customarily distinguished both com- mercially and technically. Most malt was prepared by non-brewing malt- sters who were either pure specialists or involved in various agricultural 14 Brewers’ Journal 5 (1869): 33–34, and subsequent issues. 15 Lynn Pearson, British Breweries: An Architectural History (London, 1999), 192–94, and see for instance Brewers’ Journal 16 (1880): 375–76. 16 Henry Stopes, Malt and Malting (London, 1885), 568–607. 17 In the list constructed by Lack, patent 10973 is mistakenly reported as patent 10963. 18 Mathias, Brewing Industry, 94–95. Alessandro Nuvolari and James Sumner / 100 enterprises and who might also supply distillers.19 It was quite possible, however, for a technological change in malting to have its chief practi- cal effect in the brewery. The process for making highly roasted black malt, which the chemist Daniel Wheeler tried (ultimately unsuccess- fully) to protect by patent in 1817, eventually became a highly restricted niche activity with no relevance to most maltsters’ practice. To the major porter-brewers who adopted it, however, it was central to a drive for cheaper materials and greater consistency, which fundamentally al- tered the character of the drink.20 Moreover, to the self-consciously “sci- entifi c” brewers of the nineteenth century, the single most important goal was a full understanding of the nature of malt and its possible sub- stitution. They therefore championed greater vertical integration or at least the close involvement of the brewer in technical decisions. Corre- spondingly, the trade press declared its intended readership to include not only the brewers but also the “allied trades,” the Journal adopting the full title of The Brewers’ Journal, and Hop and Malt Trades Re- view in 1867. Accordingly, when Henry Stopes published his aforemen- tioned patent list in a volume entitled Malt and Malting, devoted mainly to the construction and management of malthouses, he assumed that many of his readers would be brewers. Stopes made no attempt to dis- entangle the two trades in his bibliography, and his patent list, unlike those of the Patent Offi ce, is uncategorized. From his perspective, all the listed patents evidenced the growth of a common technical culture.21 It is also important to understand that in assigning patents to in- dustries, the Patent Offi ce clerks worked without deep knowledge of the industries concerned and erred persistently on the side of inclusion, en- tering individual patents under as many categories as possible. The 1854 Subject-Matter Index was prepared “from Titles only” (i.e., without ex- amination of the full specifi cations); any mention of malt, brewing, or distillation was suffi cient to ensure inclusion. The brief descriptions, furthermore, were worded so as to favor relevance to the category, so that the patentees’ intentions often appear more specialized than was re- ally the case. Andrew Meikle and Robert Mackell’s 1768 patent, for in- stance, is noted under “Brewing” as describing a “Machine for dressing 19 Ibid., 466–70. An early case of vertical integration at the highest level of production is that of the Calvert family, for whom see James Sumner, “Status, Scale, and Secret Ingredi- ents: The Retrospective Invention of London Porter,” History and Technology 24 (2008): 289–306, on 291. 20 William Ford, An Historical Account of the Malt Trade and Laws (London, 1849), 68– 70; Mathias, Brewing Industry, 420–23. For Wheeler’s diffi culties see the report of Wheeler v. Malins, Court of Chancery, 1818, reported in the Times, 19 Aug. 1818; and Rex v. Wheeler, Court of King’s Bench, 1819, reported in William Carpmael, Law Reports of Patent Cases, vol. 1 (London, 1843), 394–400. 21 Stopes, Malt and Malting, 409. Patents in the English Brewing Industry, 1634–1850 / 101 and cleansing malt,” but also addressed much more generally under “Agriculture” as a “Machine for dressing wheat, malt and other grain.” Examination of the patentees’ backgrounds reveals them as millwrights courting a primarily agricultural audience.22 Milling and steam-boiler technologies, too, were incorporated on the basis of a passing mention of possible brewery or distillery application. This tendency extends to the abridgement volumes, where the most relevant sections of the spec- ifi cations were abstracted for the industries in view. Even Stopes, whose professional context made him much more selective about the patents of his own time, had to rely on outline descriptions for pre-1800 pat- ents and made some questionable inclusions as a result.23 It would not be useful, therefore, to draw conclusions from the list- ings of Lack and Stopes unmediated. In recording faithfully all patents with any suggestion of a malting or brewing context, however, the two sources provide a useful dataset from which to construct a more careful categorization. Our method has been to appraise each patent based on the full specifi cation, where this is readily available, or, failing this, the Patent Offi ce abridgements or summaries found in periodicals such as the Repertory of Arts and Manufactures and Newton’s London Jour- nal of Arts and Sciences (both of which reproduced many specifi cations in full), the Retrospect of Philosophical, Mechanical, Chemical, and Agricultural Discoveries, and the Chemist, or in published accounts of litigation. In a number of (mostly pre-1800) cases, specifi cations were not enrolled or cannot be traced; for these, we have performed as full as possible a cross-comparison of any descriptions available. We have also surveyed the principal secondary sources on British brewing and distill- ing in the period and have supplemented these with primary research undertaken by one of the authors in preparing a monograph-length study on the development of scientifi c concepts in the brewery.24 In this way, we have been able to assign most of the patents listed in the Stopes and Lack selections to the following categories, chosen to minimize ambiguity. 22 Woodcroft, Subject-Matter Index, vol. 1, 31, 85; John P. Shaw, “Meikle, Andrew (1719– 1811)” Oxford Dictionary of National Biography (Oxford University Press, 2004), http:// www.oxforddnb.com/view/article/18516, accessed 3 Feb. 2013. 23 For instance, Stopes includes Walter Taylor’s 1786 patent as describing a “Malt-mill.” Woodcroft excludes the same patent from the brewing category, elsewhere deeming it to re- fer to “Machines for grinding grain” or “grinding starch for hair-powder.” 24 Mathias, Brewing Industry; Pearson, British Breweries; Terry Gourvish and Richard Wilson, The British Brewing Industry, 1830–1980 (Cambridge, UK, 1994); H. S. Corran, A History of Brewing (Newton Abbott, 1975); Ian Donnachie, A History of the Brewing Indus- try in Scotland (Edinburgh, 1979); Christine Clark, The British Malting Industry since 1830 (London, 1998); James Sumner, Brewing Science, Technology, and Print, 1700–1880 (Lon- don, forthcoming 2013). Alessandro Nuvolari and James Sumner / 102 1. Malting: innovations in the malt-making process or the con- struction of malthouses (not including innovations in the crush- ing or grinding of malt, which was always carried out by the brewer or distiller). 2. Brewing: limited to those innovations that are by nature ap- plicable only in beer-brewing (such as hop extraction or beer cleansing), those specifi ed as intended particularly for beer- brewing, and those known to have been promoted only to beer-brewers. 3. Distilling: limited to those innovations applicable only in distill- ing (such as still-head constructions, condensers, and rectifi cation techniques), those specifi ed as intended particularly for distill- ing, and those known to have been promoted only to distillers. 4. Brewing and distilling: innovations applicable in both indus- tries (such as mashers, boilers, and wort refrigerators) and ex- plicitly specifi ed or promoted as relevant to both. 5. Milling, etc.: innovations for the dressing, crushing, or grinding of grain, noted as applicable to malt (and thus potentially useful to brewers) but also to products used in other industries. 6. Boilers, etc.: boilers (including steam-boilers), furnaces, kilns, and heat transmission technologies, specifi ed as relevant to brew- ing and/or distilling, but also to other industrial processes. This categorization excludes a number of patents noted by Stopes or Lack that relate to wine- or cider-making, or that we found to describe general engineering or measurement technologies with no specifi c drinks-production context. Since our main concern is with industry attitudes to the decision to patent, our selection criteria privilege the intentions of patentees, where these are known, and exclude the unintended consequences of patent- ing. We assign the 1780 patent covering the pattern of the Dicas hy- drometer, for instance, to “distilling,” despite evidence that the pattern was later used in the brewery, because the chief aim of the patent was to secure a legislative monopoly that only affected the distillery.25 Con- versely, we assign Matthew Wood’s 1802 patent for a malt-based color- ing agent to “brewing,” despite its reference to “spirits, wines, and other liquors,” since the patent was contrived specifi cally with an eye to the brewery market.26 On this basis, we have constructed two tables. Table 1 reports our assessment of the patents contained in Lack’s and Stope’s lists based on 25 Mathias, Brewing Industry, 68–69; William Ashworth, Customs and Excise (Oxford, 2003), 275. 26 Mathias, Brewing Industry, 420. Patents in the English Brewing Industry, 1634–1850 / 103 the specifi cations and the descriptions contained in the contemporary literature. Table 2 reports the incidence of relevant patents in different periods, both in absolute numbers and as a proportion of total patent- ing. To the best of our knowledge, ours is the fi rst attempt to reconstruct in detail the volume of brewing-related patents in this historical phase. Overall, our results confi rm MacLeod’s fi nding that the industry Table 1 Patents by Type of Invention in the Lists Constructed by Lack and Stopes (1634–1850) Years 1634–1700 1701–1750 1751–1800 1801–1850 Total Brewing 3 2 15 39 59 Brewing/distilling 1 0 6 23 30 Malting 7 7 5 20 39 Boilers, etc. 0 1 8 13 22 Milling, etc. 0 0 4 7 11 Distilling 3 6 9 88 106 Sources: H. Reader Lack, Patents of Inventions: Abridgments of Specifi cations Related to Brewing, Wine-making, and Distilling Alcoholic Liquids (London, 1881); Henry Stopes, Malt and Malting: An Historical and Scientifi c and Practical Treatise, Showing as Clearly as Ex- isting Knowledge Permits What Malt Is and How to Make It (London, 1885). Table 2 Patents of Closest Identifi ed Relevance to the Brewing Industry, by Share of Total (1634–1850) Years 1634–1700 1701–1750 1751–1800 1801–1850 Total Total patents 366 292 1,804 10,972 13,434 Combined “brewing” and “brewing/distilling” patents 4 2 21 62 89 Share of total (%) 1.1 0.7 1.2 0.6 0.7 Combined “malting,” “brewing,” and “brewing/distilling” patents 11 9 26 82 128 Share of total (%) 3.0 3.1 1.4 0.7 1 Sources: Henry Stopes, Malt and Malting: An Historical and Scientifi c and Practical Trea- tise, Showing as Clearly as Existing Knowledge Permits What Malt Is and How to Make It (London, 1885); H. Reader Lack, Patents of Inventions: Abridgments of Specifi cations Re- lated to Brewing, Wine-making, and Distilling Alcoholic Liquids (London, 1881). Total pat- ents from Bennet Woodcroft, Titles of Patents of Invention Chronologically Arranged, 1617– 1852 (London, 1854). Alessandro Nuvolari and James Sumner / 104 displayed an extremely low propensity to patent throughout the period 1634 to 1850. If we consider as genuine “brewing” patents the patents assigned to the fi rst two rows of Table 1 (i.e., the categories “brewing” and “brewing/distilling”), we fi nd that the share of brewing patents in total patenting is less than 1 percent (see also Table 2). We also attempt to extend the analysis to consider the occupations of “brewing” and “brewing/distilling” patentees, drawing this informa- tion from the chronological index of English patents constructed by Woodcroft.27 Table 3 reports the percentage shares of different occupa- tions. As noted by MacLeod, the titles “gentlemen” and “esquire” may be regarded as an accurate description of social status only for the sev- enteenth and early eighteenth centuries: in later years, they were widely adopted as courtesy titles and may conceal specialist industrial back- grounds.28 The main point of interest emerging from the table, how- ever, is the signifi cant share of patentees who were identifi ably “outsid- ers” to the brewing trade. Even after excluding self-described chemists, engineers, millwrights, and merchants—groups that might have had some direct overlap with the brewery where their inventive activities were concerned—the remaining outsider occupations still account for almost 25 percent of all patents. This fi nding leads us to two conjec- tures: fi rstly, that “insider” invention developed principally in a man- ner invisible to the patent system; and secondly, that the brewing sec- tor held a particular appeal for inventors with backgrounds in other Table 3 Occupations of Brewing Patentees (1634–1850) Number % “Gentleman” 12 10.71 “Esquire” 4 3.57 Brewer 24 21.43 Chemist 8 7.14 Engineer/millwright 21 18.75 Merchant 9 8.04 Not specifi ed 6 5.36 Other “outsiders” 28 25.00 Total 112 100 Source: Occupations have been retrieved from Bennet Woodcroft, Titles of Patents of Inven- tion Chronologically Arranged, 1617–1852 (London, 1854). 27 Bennet Woodcroft, Titles of Patents of Invention Chronologically Arranged, 1617–1852 (London, 1854). The 89 brewing patents we have identifi ed in Tables 1 and 2 were taken by 112 patentees. 28 MacLeod, Inventing, 117. Patents in the English Brewing Industry, 1634–1850 / 105 fi elds who hoped, by patenting, to benefi t from some form of “trade in invention.” With these conjectures in mind, in the following sections we pro- pose a preliminary taxonomy of the appropriability strategies, patent- based and otherwise, adopted by the inventors who targeted the indus- try. Our survey is based on the secondary sources mentioned above (note 24) and on primary investigations of published literature, brew- ing records, and correspondence held in archives. We distinguish fi ve principal approaches. First, of course, there was the traditional recourse to secrecy. Second, there was also a traditional rhetoric of openness, in- voked by those who professed to have no commercial interest in their achievements and by others who claimed that it was impossible, for various technical or economic reasons, to appropriate their work even given the fullest possible disclosure. Some inventors combined ele- ments of these two approaches in a third, the use of selective revelation as a promotional tool. Fourth, there was patenting as applied by outsid- ers, who in practice often struggled to make headway against non- p atenting competitors from inside the trade. Finally, however, and only towards the end of the period under discussion, there emerged a cate- gory of insiders who saw benefi ts in patenting themselves. The follow- ing sections consider each of these approaches in turn. Secrecy The most obvious alternative to patenting was silence. Tradition- ally, the principal techniques of beer-brewing had been considered uni- versal common knowledge, as for baking or other domestic arts, but commentators from around 1700 began to refer to differences in the methods and products of the large well-capitalized brewhouses that served the concentrated markets of London and other major towns.29 Technical knowledge of processes was typically protected by oral com- munication, direct demonstration, and guild affi liation. Some London apprenticeships around 1740 carried premiums of several hundred pounds: the Brewers’ Company did not control the right to brew, but the fee bought training and a set of connections that were diffi cult to gain by any other means.30 The most notable technical change to emerge within this culture was the rise of the distinctive, well-hopped, and dark style of beer known as 29 [Jeffrey Boys], Directions for Brewing Malt Liquors (London, 1700); [William Ellis], The London and Country Brewer, 1st ed. (London, [1735?]), 36–49. 30 Richard Unger, Beer in the Middle Ages and the Renaissance (Philadelphia, 2004), 211; [Ellis], London and Country Brewer, 35; A General Description of all Trades, Digested in Alphabetical Order (London, 1747), 34–36. Alessandro Nuvolari and James Sumner / 106 London porter. Porter production combined effi cient use of materials with greatly improved stability in storage, making it uniquely suited to large-scale production; the main points of the innovation were changes to the malt and hop profi le and a greatly increased storage time, often by the use of newly developed storage vessels.31 As the style spread from London to the urban centers of Britain and Ireland across the eigh- teenth century, porter-brewing gained a tantalizing reputation as a prof- itable “secret” known to few producers.32 A popular story emerged that equated the “invention of porter” with a single discrete innovation, sup- posedly made by an unsung brewer of genius in 1722. This account in fact telescoped and projected backwards a century of gradual change in a cul- ture that was not necessarily much concerned with the identifi cation or reward of individual inventors.33 Typical of the transmission mecha- nism is the case of Nathaniel Chivers, “bred to the porter brewing in Lon- don,” who was paid a hundred guineas to bring his knowledge to Glasgow in 1775 under conditions (later breached) of local exclusivity.34 Indeed, there are several accounts by both insiders and outsiders to brewing that present secrecy as a defi ning element of its culture. These usually focus on the extreme lengths to which brewers would go to pro- tect production details such as quantities of materials, times, and tem- peratures. Instrument-makers of the nineteenth century routinely sup- plied brewers with the so-called “blind thermometer,” featuring a traveling index marker and detachable scale. With the scale in place, the brewer could set the marker to a particular value, such as the desired mashing temperature; he would then remove the scale before handing the thermometer to his servant who would then set the temperature without being able to communicate it.35 Robert Brakspear, who kept a middle-sized brewery at Henley, Oxfordshire, from 1779, made notes using an esoteric system of coded symbols: a square stood for “hours,” a theta for “mild beer,” a diamond for “isinglass,” and so on.36 At the major 31 The standard account is Mathias, Brewing Industry, 12–16, 53–62; see also the modifi - cation proposed in James Sumner, “Status.” Not surveyed in this article is an important pri- mary source on storage vessels: Victualling Commissioners to Philip Stephens, 24 Jan. 1774, in Navy Board Victualling Offi ce out-letters, ADM 110/26 folios 101–15, the National Ar- chives, Kew, Richmond, UK. 32 For instance, George Watkins, The Compleat English Brewer (London, 1767), 122; Gentleman’s Magazine 60 (1790): 801. 33 Sumner, “Status,” 297–99; Martyn Cornell, Beer: The Story of the Pint (London, 2003), 92–95. 34 [Robert Reid], Glasgow: Past and Present, vol. 3 (Glasgow, 1856), 415–19. 35 Alexander Morrice, A Treatise on Brewing, 1st ed. (London, 1802), appendix; John Tuck, The Private Brewer’s Guide, 2d ed. (London, 1822), 32, 60; William Loftus, The Brewer: A Familiar Treatise (London, 1863), 143. 36 Brakspear papers, II/i/1, passim, and II/i/3, Oxfordshire History Centre, Oxford, UK. See also Francis Sheppard, Brakspear’s Brewery, Henley on Thames, 1779–1979 (Henley on Thames, 1979). Patents in the English Brewing Industry, 1634–1850 / 107 London fi rm of Reid’s, written records were “partly faked,” presumably according to an agreed upon and orally communicated system.37 Such colorful strategies are not necessarily representative of the wider culture. The surviving evidence of brewery production logs at all scales of operation suggests that most were not intentionally cryptic. All employ some degree of esoteric shorthand, but only so far as is normal in any specialized activity. It is also easy to make too much of the exotic codes and shibboleths of guild culture, which were of declining rele- vance. Writing in 1760, one long-serving member of the London trade claimed that the traditional function of the Brewers’ Company as a reg- ulatory body had broken down entirely around 1730, giving way to an “anarchy” of new entrants looking to get rich quickly.38 Protecting inno- vative advantage typically relied on a more mundane strategy of keep- ing sensitive records and equipment behind closed doors, paying high salaries to skilled staff, and otherwise trusting in the barrier formed by the high degree of tacit knowledge that was needed to turn a profi t from most new equipment or methods. If brewers could be secretive, they were often skeptical of “secrets,” in the sense of prescriptions that paper communication revealed. What Nathaniel Chivers’s Glasgow employers contracted to buy from him, for instance, was not a written specifi cation of the London method, as much as on-site practical demonstration. A market for secret specifi cations did, however, operate to some ex- tent. In 1762, Humphrey Jackson (1717–1801), an apothecary and man- ufacturing chemist, proposed a course of lectures in which he would ex- plain methods for malting, the analysis of malt, hops, and yeast, “the real Causes of Cloudyness in Malt Liquors,” “an easy practicable Method of Brewing Porter in Summer,” and “a most useful new improved Ther- mometer adapted to the Brewery.” Jackson was probably inspired by his father-in-law, Benjamin Martin, who delivered public lectures to promote sales of the scientifi c instruments he retailed. Jackson’s course, however, was emphatically not public. “As soon as an adequate Sum is subscribed,” he advertised, The Proposer will execute his Proposals, by way of Lecture, sup- ported with Experiments, at any Time and Place agreed on; after which he will print them at large; and by leaving certain blank Spaces in the Book, to be fi lled up afterwards in Writing, or by Ref- erences printed separately, they will be rendered useful to the Sub- scribers only.39 37 Walter Pearce Serocold, The Story of Watney’s (London, 1949), 19. 38 London Chronicle, 22 Dec. 1760, 603. 39 Humphrey Jackson, “Proposals for Communicating and Explaining the Real Causes of Cloudyness in Malt Liquors,” printed circular, June 1762, papers of James Best of Chatham, U480/B874, Medway Archives, Strood, Kent, UK (hereafter, Best papers). Alessandro Nuvolari and James Sumner / 108 The book with blanks was a variation on an old guild device. In the seventeenth century, the Company of Distillers in London had circu- lated printed lists of process directions to its freemen with crucial infor- mation disguised by symbols that became intelligible once handwritten keys were added. The system had an obvious drawback: the text was later pirated and republished with the keys included (“for the Publike good”).40 Jackson, however, was looking to increase the pace of techni- cal change in the brewhouse and presumably reckoned on being able to keep ahead of any printed disclosures. In the event, it seems that the proposed printing did not take place. Jackson became advisor to the major London porter brewer, Henry Thrale, and the lecture scheme was abandoned. Around 1770, however, Thrale sanctioned a partial revival of the lecture scheme, provided Jackson advised only “Country Brew- ers” far from Thrale’s territory.41 The brewers who were instructed under such terms could generally be trusted to keep confi dentiality agreements, as any breach would potentially aid their own rivals. Almost all attempts to gain a premium from secrecy relied on some form of direct personal interaction between inventor and client, in the lecture-room or brewhouse, but there were also attempts to promote privileged knowledge—or at least the promise of it—through correspon- dence, directions copied in manuscript, or even print. The brewer George Lloyd Worthington, in 1812, sold a book of guidance privately at fi ve guineas (£5.5.0), ten times the price of a more compendious con- temporary manual available to the public.42 The intention, of course, was that the extent of the buyer’s investment would itself secure his support in protecting confi dentiality. George Adolphus Wigney, brewer and inventor of an unpatented refrigeration system, went much further in 1835, offering a comprehensive treatise privately to major brewers at £100. Supported by testimonials from satisfi ed customers and employ- ers, he was able to meet his subscription target.43 Openness Some innovators, by contrast, professed an attitude of complete openness regarding the content of their innovations, for a variety of 40 The Distiller of London (London, 1639); The Distiller of London, with the Clavis to Un- lock the Deepest Secrets of that Mysterious Art (London, 1652). 41 Humphrey Jackson to James Best, 21 July 1770, Best papers. 42 George Lloyd Worthington, The Brewer’s Guide: A New Work (London, 1812); cf. Alex- ander Morrice, A Treatise on Brewing, 2d ed. (London, 1802). 43 George Adolphus Wigney, A Practical and Theoretical Treatise on Malting and Brew- ing (Brighton, 1835); circular announcements, EK-U1453/B2/40/588, at Kent History and Library Centre, Maidstone, Kent, UK. The rise of cheap print was no barrier to the recurrence of this strategy: see George Stewart Amsinck, Practical Brewings: A Series of Fifty Brewings (London, 1868), priced twelve guineas. Patents in the English Brewing Industry, 1634–1850 / 109 reasons.44 A few claimed—perhaps sincerely—to seek no reward from their discoveries beyond the glory of achievement. A key example is Michael Combrune (d. 1773), the fi rst commercial brewer to publish in detail on trade practices. Combrune’s most crucial achievement was in popularizing the use of thermometers in malting and brewing. Though he had no proprietary interest in any particular device, his system of management itself, based on tabulated relationships between malt col- ors, mashing heats, and the desired keeping qualities of the beer, could have been made the subject of a patent application or promoted pri- vately as a secret. Instead, Combrune fi rst offered his system to the London Company of Brewers, to which he belonged, to be held in com- mon for the eyes of Company freemen only. To his surprise, Company offi cials (perhaps not sharing his opinion of its signifi cance) pronounced him “at full liberty to dispence of it to the public.”45 This he did in two volumes of 1758 and 1762, which were widely excerpted in the review- ing press.46 Combrune occasionally acted as a private consultant to other brewers, dispensing information about practical thermometry not found in his books, but made it clear that he did not see this as a princi- pal source of income.47 His chief aim, in publishing, was to gain accep- tance as a chemical theorist. To others, building a reputation within the trade might have more commercial signifi cance. James Baverstock, a brewer on a middling scale at Alton in Hampshire, discovered (according to his own version of events) the use of the fl otation hydrometer to measure wort strength in the 1760s. He initially approached several leading brewers privately with the knowledge. Like Humphrey Jackson, he secured the attention of Henry Thrale, who later presented Baverstock with a silver hydrome- ter in recognition of his achievements. Most brewers, then, were not ri- vals. Only those competing over the same geographical and product markets were in a position to damage each other’s livelihoods, and the greatest brewers could safely exercise a paternalistic attitude towards the rest of the trade. A brewer in Baverstock’s position could, by shar- ing his innovations, place himself effectively under the patronage of a 44 For a discussion of openness and knowledge sharing among inventors in different historical contexts, see James Bessen and Alessandro Nuvolari, “Knowledge Sharing among Inventors: Some Historical Perspectives” in Revolutionizing Innovation: Users, Communi- ties, and Open Innovation, ed. Dietmar Harhoff and Karim Lakhani (Cambridge, MA, forthcoming). 45 Michael Combrune to James Best, 17 Aug. 1762, Best papers; Michael Combrune, Es- say on Brewing (London, 1758), [v]–[vi]. 46 Combrune, Essay; Michael Combrune, The Theory and Practice of Brewing (London, 1762). 47 Combrune to Best, 17 Aug. 1762, Best papers; Council minute book entries for 25 June and 16 July 1772, Royal Dublin Society Archives, Dublin, Ireland. Alessandro Nuvolari and James Sumner / 110 relative giant such as Thrale. The reward might be entry into the Lon- don market for his product (under terms agreeable to his patron) or the likelihood of employment in the event of the failure of his own busi- ness. Like Combrune, Baverstock claimed to be acting in the general interest of the trade. Yet unlike Combrune, he did not publish until John Richardson, the promoter of a proprietary hydrometer, claimed the novelty of the discovery for himself in 1784, stinging Baverstock into asserting his priority. Baverstock also responded by engaging an instrument-maker to develop a hydrometer to his own specifi cation, but from 1786 he was managing partner in an increasingly profi table brewhouse at Windsor and apparently had neither time nor need to consider marketing innovations as a source of income.48 The dominant porter brewers of London were particularly inclined to make a strategic show of openness given that, by around 1800, they were heavily insulated from rising competition by the immense quan- tity of capital needed to break into the market at their level of produc- tion. They would often fl aunt their technical innovations as statements of forward-thinking economy and reliability, aimed not only at drinkers but also at each other. In 1784, Henry Goodwyn, who ranked as Lon- don’s seventh brewer in terms of output, became the fi rst to commis- sion a Boulton and Watt engine; an order from Whitbread, the leading brewer, followed a month later. Boulton and Watt’s design itself was protected, of course, but its installation, requiring various adaptations in the brewhouse, must have generated valuable experience. We might expect Goodwyn to have sought patents or simply to have kept the in- formation to himself. Instead, he went out of his way to offer himself as an advisor to others planning to introduce steam, including his brewery rivals.49 The point was to confi rm Goodwyn’s status—notwithstanding that his output was only half of Whitbread’s—within the charmed circle of large-scale brewers. Goodwyn’s magnanimity was intrinsically re- stricted to that circle. For all their one-upmanship, the greatest brewers made collective agreements on issues such as retail pricing and were in many ways almost a cartel, safe in the knowledge that no competitor could raise the capital to oppose them.50 Unrestricted disclosure, then, was harmless so long as no outsider was in a position to use the infor- mation disclosed. 48 James Baverstock, Hydrometrical Observations and Experiments in the Brewery (London, 1785); James Baverstock, Treatises on Brewing (London, 1824), xvi–xix, 299– 306. 49 Mathias, Brewing Industry, 89–90. 50 For the only notably successful—and yet short-lived—project to breach this capital bar- rier, see Mathias, Brewing Industry, 243–51. Patents in the English Brewing Industry, 1634–1850 / 111 Partial Publication without Patenting Some innovators sought to combine the advantages of secrecy and openness, publishing accounts of discoveries that were suffi cient to demonstrate their expertise, but that could not be applied in practice without private consultancy. The most infl uential exponent of this ap- proach was John Richardson (1743–1815), mentioned above as the rival of James Baverstock. Richardson had been advertising privately as a consultant for some time when in 1777 he published a short volume, modeled loosely on contemporary chemical treatises, offering “Hints” towards principles that, he said, would direct a more effi cient prac- tice.51 In 1784, now established as a successful brewery proprietor in Hull, Richardson engaged a prominent London instrument-maker, John Troughton, to produce a custom-designed hydrometer, which he pro- moted through a second, more detailed treatise.52 Richardson dubbed the new device a “saccharometer,” emphasizing its particular applica- tion to the dissolved sugars in brewers’ wort, and claimed to be the pio- neer of wort hydrometry. In fact, some brewers had known of this ap- proach since the 1740s, and Richardson’s priority claim triggered fi erce public objections from Baverstock.53 Nonetheless, by carefully present- ing himself as a champion of brewers’ interests, Richardson established the saccharometer as an industry standard. To affi rm that it was not merely an adapted distillery device, he published a detailed history of the instrument’s calibration, as performed in his own brewhouse using barrels of brewers’ wort.54 There is no evidence that Richardson attempted to patent the sac- charometer. From a technical standpoint, he could probably have done so. The design contained novel features such as a sliding “regulator,” designed to correct mechanically for differences in local water, and was at least as distinctive as a hydrometer patented by John Dicas in 1780.55 Yet Richardson instead chose to prioritize building up his reputation 51 John Richardson, Theoretic Hints on an Improved Practice of Brewing Malt-Liquors (London, 1777). 52 John Richardson, Statical Estimates of the Materials for Brewing, or a Treatise on the Application and Use of the Saccharometer (London, 1784). 53 Richard Clarke, Notice Is Hereby Given, to all Dealers in Brandy, Rum, Malt, or Melasses-spirits, Arrack, &c. that the Hydrometer, or Brandy-Prover, Being the Production of many Years Study and Experiments, is now Brought to its Utmost Perfection (London, [1746]); James Baverstock, Hydrometrical Observations and Experiments in the Brewery (London, 1785). 54 James Sumner, “John Richardson, Saccharometry and the Pounds-per-Barrel Extract: The Construction of a Quantity,” British Journal for the History of Science 34 (2001): 255–73. 55 Richardson, Statical Estimates, 3, 30; and cf. Mathias, Brewing Industry, 69, no. 2; Dicas, John. Patent 1259, 27 June 1780. The date given for all patents cited is the date in which the patent was granted (“sealed”). Alessandro Nuvolari and James Sumner / 112 among the brewers. His publications cast him as a “scientifi c” (a word that appeared in his titles from 1788) authority, while stressing that he remained a commercial brewer and would advise primarily on a com- mercial basis.56 The physical technology of Richardson’s saccharometer was only one element of an interlinked system, which also included the published treatises, direct instruction by personal attendance, and se- cret manuscripts. The text of one such manuscript found its way into print some years after Richardson’s death. It does not describe any novel apparatus, but consists largely of operational directions to be car- ried out using conventional equipment, guided by Richardson’s sac- charometer and a thermometer of standard Fahrenheit pattern, along with directions for record-keeping. Richardson’s 1784 treatise retailed at fi ve shillings, and the saccharometer itself at three guineas (£3.3.0); yet the manuscript commanded 150 guineas (£157.10.0) plus a guaran- tee of twenty years’ secrecy.57 It is likely, therefore, that Richardson considered he had little to fear from the emulation of his technology by others, provided his authority remained intact. The name “saccharome- ter” and Richardson’s brewery-specifi c scaling were, indeed, appropri- ated by more established hydrometer-makers in his lifetime.58 None- theless, Richardson’s reputation as a key brewing theorist grew and outlived him, as his consultancy, and his own brewery, prospered.59 Richardson’s strategy was refi ned by later consultant brewers such as William Black, who published an extensive treatise in 1835. Black emphasized the importance of chemical theory to effective brewing, paying particular attention to contemporary research suggesting that electrical action could infl uence the progress of fermentation. Con- vinced that the unpredictable souring still widely experienced in the most up-to-date breweries was an effect of atmospheric electricity, Black began to promote the comprehensive removal of iron vessels and fi ttings and their replacement with electrically insulating materials.60 The same interpretation was later to form the basis of two patents, in- cluding one by Andrew Crosse, an amateur whose electrical research was famous in his day.61 Yet Black, despite setting up several breweries with non-metallic equipment, never patented anything himself. 56 Richardson, Theoretic Hints, 1–2. 57 Richardson, Statical Estimates, 241; [David Booth], The Art of Brewing (London, 1829), [33]. 58 George Skene Keith, “Observations on the Papers Presented to the House of Com- mons,” Farmer’s Magazine (1807), 476–500, on 487–88; Friedrich Accum, A Treatise on the Art of Brewing (London, 1820), 104. 59 [Booth], Art of Brewing, 11–15; William Black, A Practical Treatise on Brewing, 1st ed. (London, 1835), 37; Pat Aldabella and Robert Barnard, Hull and East Riding Breweries (Beverley, 1997), 80–81. 60 Black, Practical Treatise, 11–14, 29–31, 94–97. 61 Crosse, Andrew. Patent 11604. 2 Mar. 1847; Van Kempen, Peter, communicating an in- vention of Gerrit Abraham Cramer. Patent 14015. 8 Mar. 1852. Patents in the English Brewing Industry, 1634–1850 / 113 For a variety of reasons, then, the promoters of brewery innova- tions—in particular, those who were themselves brewers—often focused on strategies that had no use for the kind of protection afforded by the patent system. Indeed, a consultant such as Black, who expected to live by tutoring and consultancy fees rather than royalties, might positively welcome the early appropriation of his innovations by others. Patenting as Used by “Outsiders” By the late eighteenth century, large-scale brewers increasingly found themselves dealing with the installation and management of patent technologies aimed at a broad range of industries, such as steam engines and milling machines. There were also a few patents by non- brewers that, though of broad potential scope, were in practice of pri- marily brewery signifi cance, such as Matthew Wood’s 1802 process for making artifi cial coloring.62 Humphrey Jackson, discussed above as a promoter of brewing systems under conditions of secrecy, provides an earlier example. He fi rst came to the attention of the brewing commu- nity when he obtained, in 1760, a patent on the production of isin- glass.63 This material, prepared from sturgeon and other fi sh, was used in a variety of industries, but its chief use was in fi ning (removing cloud- iness from) London porter. By one report, leading brewers combined to gift Jackson “a considerable sum of money” towards the perfection of his process.64 Jackson was later to invoke his brewery connections in a further scheme involving a chemical process for hardening timber, pat- ented in 1769. Though his chief intended market was always shipbuild- ing, his circulars to brewing clients pointed out the importance of casks and wooden storage vats to their operations, and it was Jackson’s chief brewery patron, Henry Thrale, who largely funded the project.65 John Long (d. 1807), a few years later, made patenting more cen- tral to his approach. Long’s background is obscure: he is described as a “Merchant” of Longville, County Dublin, and was apparently a protégé of John Beresford, First Commissioner of the Revenue in Ireland. He 62 Wood, Matthew. Patent 2625. 31 May 1802. 63 Jackson, Humphrey. Patent 749. 26 Mar. 1760. 64 Robert Dossie, Memoirs of Agriculture, vol. 1 (London, 1768), 275–77. See also Math- ias, Brewing Industry, 51–53. 65 Jackson, Humphrey. Patent 910. 9 Dec. 1768; Humphrey Jackson, New Art of Harden- ing and Preserving Wood (London, 1770); John Appleby, “Humphrey Jackson, F.R.S., 1717– 1801: A Pioneering Chemist,” Notes and Records of the Royal Society 40 (1986): 147–68, on 152–56; Katharine Balderston, ed., Thraliana: The Diary of Mrs. Hester Lynch Thrale, 1st ed., vol. 1 (Oxford, 1951), 53 and 312; Abraham Hayward, ed., Autobiography, Letters and Literary Remains of Mrs. Piozzi, 1st ed., vol. 1 (London, 1861), 257–58; James Clifford, ed., Dr. Campbell’s Diary of a Visit to England in 1775 (Cambridge, UK, 1947), 51. Alessandro Nuvolari and James Sumner / 114 was thus able, on traveling to England in 1789, to tour the major brew- eries and distilleries, making investigations into their operations that inspired his two patents of the following year.66 Long’s specifi cations are notable as the fi rst to describe an entire system of brewing, includ- ing innovations in large parts of the process, as opposed to a method or apparatus for an individual task. Most of Long’s proposals involved the use of submerged pipe coils, similar to distillers’ worms, through which hot or cold water would circulate to achieve temperature control in var- ious brewery vessels. In Long’s case, again, we see that patenting was not always viewed as the optimum strategy. Returning to Dublin in 1792, Long petitioned the Irish Parliament, stating that he was in pos- session of a full systematic knowledge of English production, and he was willing to negotiate disclosure based either on the granting of an Irish patent to himself or on remuneration in exchange for a general public disclosure to the Irish brewers. Parliament favored the latter course, which probably explains Long’s appointment as “General Ex- aminer and Inspector of Breweries” by 1800.67 Another patentee of a system of brewing was Richard Shannon, who styled himself “M.D.” but was apparently active as a chemist, dis- tiller, and metalware dealer. Shannon’s system of 1798, like Long’s, fo- cused on temperature control, but relied on steam-jackets and similar enclosures to convey hot water, cold water, or air.68 In 1805, he issued a treatise on brewing, distilling, and vinegar-making that was, for its time, the most extensive and expensive work on the subject, running to over one thousand pages of lavishly produced quarto and sold by sub- scription at £2.12.6. The apparently comprehensive work was, in fact, simply a promotional device on a grand scale. The text was built around Shannon’s patent specifi cations and other descriptions of his equip- ment, bulked out with lengthy quotations from Richardson, Baverstock, and other authors.69 The activities of such outsiders inevitably attracted the suspicions of some within the trade, to whom a reliance on patenting might sug- gest impracticality or imposture. Humphrey Jackson’s involvement with Henry Thrale ended in disaster. Thrale over-committed himself 66 Long, John. Patent 1754. 4 June 1790; Long, John, and Thomas Harris. Patent 1769. 28 July 1790. The second patent, registered jointly with a distiller, consolidates the specifi cation of wort-cooling apparatus described in general terms in the fi rst. For biographical details see Monthly Magazine 24 (1807): 413. 67 Votes of the [Irish] House of Commons (Dublin, [1792]), 365–66, 446–47; Parliamen- tary Register, or, History of the Proceedings and Debates of the House of Commons of Ire- land 12 (1793): 259–60, 299; John Long, A Treatise on Malting (Dublin, 1800), title page. 68 Shannon, Richard. Patent 2212. 1 Feb. 1798. 69 Richard Shannon, A Practical Treatise on Brewing, Distilling, and Rectifi cation (Lon- don, 1805). Patents in the English Brewing Industry, 1634–1850 / 115 fi nancially to the timber-hardening scheme, and the brewhouse staff blamed Jackson when, in 1772, one of the wooden storage vessels burst apart and the remainder of the year’s production was found to be sour, leaving the fi rm on the brink of collapse.70 Instructors and innovators with a brewery background had an obvious interest in excluding the outsiders. Jackson was also attacked by John Richardson who produced an anonymous pamphlet assailing his “secret system” as dangerous quackery.71 John Long, in similar fashion, ran afoul of George Blake, a trained ale-brewer who operated as a traveling consultant, initially in private. Long had approached Blake as a potential partner, but Blake derided Long’s patent system as absurdly impractical and was moved to publish for the fi rst time in order to dissociate himself from it.72 (The experience apparently converted Blake to the value of print. In later years he published a volume describing the innovations on which he advised privately, again without patenting.)73 Shannon, it appears, was more convincing: the seven hundred subscribers to his treatise included most of the trade’s conspicuous leaders in London and provincial cen- ters such as Burton upon Trent .74 Patenting as Used by “Insiders” No patentee of an evidently brewing-related invention gave his oc- cupation as “brewer” until 1784 when Sutton Thomas Wood of Oxford took out a patent on an approach that could be applied to wort-boiling and to comparable boiling stages in other manufactures. The boiler was to be covered and the steam conveyed by pipes to perform various use- ful tasks, such as working a steam engine or providing heat elsewhere in the process.75 The specifi cation built on a slightly earlier patent in which Wood described modifi cations to steam-engine design, and it was followed by another extending the approach to various other man- ufactures. A few years later Wood obtained a fourth and fi nal patent on a steam-driven weaving engine.76 Wood’s innovations did not, there- fore, focus notably on brewing. He might rather be classed with the 70 Balderston, Thraliana, 1: 53, 312; Hayward, Autobiography, 1: 257–58; Hester Piozzi, ed., Letters to and from the Late Samuel Johnson, vol. 1 (London, 1788), 78–79. 71 [John Richardson], Observations on the Art of Brewing Malt Liquors, in a Series of Strictures on a Secret System (London, 1775). 72 George Blake, Strictures on a New Mode of Brewing (London, 1791). 73 George Blake, Theoretical and Practical Remarks on G. Blake’s System of Malting and Brewing (London, 1817). 74 Shannon, Practical Treatise, [xxvii]–xxxii. 75 Wood, Sutton Thomas. Patent 1455. 17 Nov. 1784. 76 Wood, Sutton Thomas. Patent 1447. 20 Aug. 1784; Wood, Sutton Thomas. Patent 1492. 27 July 1785; Wood, Sutton Thomas. Patent 1860. 15 Mar. 1792. Alessandro Nuvolari and James Sumner / 116 numerous contemporary engineers concerned with improving the effi - ciency of steam-heating.77 A self-described “brewer,” in this period, could be anything from an expert in brewhouse technology to a sleeping part- ner whose interests lay elsewhere, and without fi rm biographical evi- dence it is diffi cult to assess how far Wood could appeal to brewers as a knowledgeable insider. Nonetheless, when the largest breweries began to install steam engines in the 1780s, Wood was in contention as a po- tential rival manufacturer to Boulton and Watt.78 From Wood’s time, as suggested by Table 3, a signifi cant minority of brewery-relevant patents came from identifi able brewers. In 1787, James Walker of Dover took out the fi rst patent covering a mechanized mashing-rake. William Ker (d. 1807), said to have erected “one of the completest breweries and distilleries” in Peebles, patented in 1788 a scheme to capture and condense the steam from hop-boiling in order to recover the essential oil of hops that would otherwise be lost to the at- mosphere.79 Walker’s patent specifi ed particular technical contrivances for carrying out known tasks; Ker’s patent outlined the general prin- ciple of a proposed reform that could be carried out by various means. Both innovations inspired a number of later patentees, but it is impos- sible to gauge to what extent the patents were actually worked. The same is true of the patent obtained fi fty years later by Peter Walker, a Liverpool brewer. Walker’s fermentation cleansing apparatus has since been claimed as the origin of the Burton Union system, a distinctive mode of arrangement and management of fermentation that stands as an icon of the nineteenth-century wave of brewery industrialization centered on Burton and the West Midlands. Paul Bayley’s review of the literature, however, suggests that Walker’s prominence may simply be an artifact of the relatively high survival and traceability of patent spec- ifi cations. The development of the Burton Union system, like that of London porter, evidently involved the gradual assimilation of numer- ous innovations. One of these, developed by the London brewer R. W. Dickinson around 1823, received not a patent but a Society of Arts medal for the disclosure.80 The fi rst London brewer to patent was Richard Hare, whose porter brewery at Limehouse had been in his family since at least the 1730s. In 77 MacLeod, Inventing, 178. 78 Richard Hare to Boulton and Watt, 11 Jan. 1785. In “Industrial Revolution: A Docu- mentary History” (microfi lm series. Marlborough: Adam Matthew Publications, 1993 on- wards), series 1, part 13, reel 243, 3/392, item 16. 79 Ker, William. Patent 1641. 4 Mar. 1788; John Sinclair, ed., The Statistical Account of Scotland, vol. 12 (Edinburgh, 1794), 8. 80 Paul Bayley, “An Evaluation of the Number and Distribution of Burton Unions,” Brew- ery History 129 (2008): 39–72; Walker, Peter. Patent 7658. 31 May 1838; “Apparatus for Clearing Beer,” Transactions of the Society, Instituted at London, for the Encouragement of Arts, Manufactures, and Commerce 42 (1823): 23–25. Patents in the English Brewing Industry, 1634–1850 / 117 1790 it was ranked the fi fteenth-greatest in London on an annual out- put of 23,000 barrels, putting it on a par with the largest provincial breweries.81 Hare’s patent of 1791 describes a closed boiler whose stated purpose, as in Ker’s patent, is to minimize the loss of volatile hop oils, although the design was probably more inspired by Wood’s engine: Hare had consulted Wood in 1758. Hare installed a version of his appa- ratus at his own brewery, and when he sold the business in 1792, he li- censed the invention to his successors for an annuity of £100 for the re- maining term of the patent. On learning of Wood’s prior activities, however, the successors suspected that Hare’s patent was not secure, and by ceasing to pay the annuity, they triggered an action from Hare in 1803. With Wood testifying on their behalf, Hare’s patent was duly voided. While the specifi ed purpose of Hare’s apparatus was unques- tionably distinct from Wood’s, its operation was held to fall entirely within the detailed conceptual scheme Wood had submitted.82 Later surveys noted that “Hare’s apparatus is now universally employed in all extensive breweries: the construction is more complete than Wood’s.”83 The art of patenting, like that of brewing, contained many traps for the inexperienced and the poorly connected. The two brewers with the highest profi le to hold patents were Ed- ward Biley and Henry Goodwyn, Junior, who both enrolled mashing machines in, respectively, 1793 and 1797. Members of the coterie of major brewers were evidently somewhat amenable to paying each other royalties under the patent system. Whitbread introduced a Biley masher after 1796.84 Goodwyn, son of the above-mentioned steam-engine pio- neer, was the sixth- or seventh-greatest producer around the time of his patent. Yet the very greatest and most highly capitalized brewers such as Whitbread, Barclay Perkins, and Meux Reid held no patents despite the fact that their frenetic growth led to numerous innovations. Patent- ing was one possible revenue stream among many and an uncertain one at that. The London giants’ attentions were far more focused on build- ing larger and more reliable markets by “tying” public houses to their product through leaseholding or loans to publicans.85 Nonetheless, the major brewers’ interest in steam enmeshed them indirectly in some of the highest-profi le debates on the nature and scope of patent protec- tion. While most of its competitors dealt with Boulton and Watt, Meux 81 Thomas Pennant, Of London (London, 1790), 279. 82 Repertory of Arts, 2d ser. 3 (1803): 232–35. A subsequent action confi rmed Hare’s right to the money already paid out: New Reports of Cases Argued and Determined in the Court of Common Pleas [ from 1804 to 1807] vol. 1 (1826), 260–63. 83 Report from the Select Committee on the Law Relative to Patents, 12 June 1829, 195. 84 Mathias, Brewing Industry, 95. 85 For the origins of the tied trade in London see Mathias, Brewing Industry, 117–38; and for its subsequent growth, Gourvish and Wilson, British Brewing Industry, 128–46. Alessandro Nuvolari and James Sumner / 118 Reid commissioned one of Hornblower and Maberley’s engines; this was the principal case cited when Boulton and Watt successfully sued their rivals for infringement.86 This case is also signifi cant for the light it sheds on the beginnings of what became the specialist class of “brewer’s engineer.” The engineer appointed to the project, Arthur Woolf, found the Meux Reid site par- ticularly productive for testing new refi nements. The brewers, for their part, found various uses for his talents, so that Woolf stayed on as a sal- aried resident consultant until 1806, cementing the relationship by marrying Mrs. Meux’s maid.87 Woolf’s projects at Meux’s included an apparatus for heating water by waste steam erected in 1800.88 His fi rst patent, for a tubular boiler in 1803, was based directly on his experi- ence of installing two boilers at Meux’s. These boilers were initially de- signed to replace open-fi ring of the brewing coppers, the invention being adapted later to supply steam engines.89 From the early nineteenth century, there began to emerge a num- ber of dedicated “brewers’ engineers” who typically had general me- chanical engineering training and might or might not work exclusively on brewery projects. Unlike Woolf, these practitioners were not re- tained by an individual fi rm, but served a wide variety of brewers as consultants. They were as likely as other mechanical engineers to take out patents, working them directly, with their consultancy clients pro- viding a captive market. One of the most active was William Tizard, who positioned himself in the 1840s as an experienced brewer who no longer had vested interests in any one brewery, prepared to consult for anyone who agreed to his terms. Tizard took it for granted that his in- ventions should be protected under patent. His fi rst specifi cation, in 1841, describes a system of innovations across the whole process from malt-crushing to fermentation, including the whimsically named “hys- tricon” (a rotary device to agitate the grain while sprinkling it with water), “Caloriphagon” (a lamp enclosed in a cooling water jacket to avoid heating the fermenting room), and “Pneumatic Life-protector” (a signal light to show excess build-up of carbonic acid gas in subterranean vaults).90 Tizard described all these innovations in detail in an extensive 86 Henry Dickinson and Rhys Jenkins, James Watt and the Steam Engine (Oxford, 1927), 69, 307. 87 Mathias, Brewing Industry, 95–96. 88 “Description of an Apparatus for Heating Water by Waste Steam, Invented by Mr. Arthur Woolf,” Journal of Natural Philosophy 1 (1802): 203–4, and see Thomas Harris, Arthur Woolf: The Cornish Engineer, 1766–1837 (Truro, 1966). 89 “On Mr. Arthur Woolf’s Improved Apparatus, Applicable to Steam Engines and Other Purposes of Art and Manufacture; Including a Description of Two Boilers Now Erecting at Messrs. Meux’s Brewery,” Philosophical Magazine 17 (1803): 40–47; Woolf, Arthur. Patent 2726. 29 July 1803. 90 Tizard, William. Patent 8921. 5 Apr. 1841. Patents in the English Brewing Industry, 1634–1850 / 119 treatise on brewing, fi rst published in 1843, in which he adapted the strategy adopted earlier by Richard Shannon. However, where Shan- non had merely bulked out his specifi cations with hack borrowing, Tiz- ard based his work on an exceptionally comprehensive survey of the lit- erature, carefully updating the text between editions to showcase his mastery of the latest knowledge—and to denigrate the approaches and apparatus offered by rivals.91 Tizard never licensed his patents, instead managing the installation of all apparatus privately on his own account, alongside his more general advisory consultancy. Conclusions Recent studies of patenting activities, both contemporary and his- torical in focus, have pointed to signifi cant intersectoral differences in the effectiveness of patents as a tool for protecting inventions.92 Our study shows that it is important to acknowledge differences in appro- priability strategies even within the same industry. The case of the En- glish brewing industry displays widespread and persistent behavioral heterogeneity. We might perhaps regard the industry as a complex eco- logical system in which inventors could successfully develop different strategies fi t to ensure their survival in specifi c niches. No one approach, therefore, was “typical” or “characteristic” of the industry at any given time.93 There is, however, evidence of a widespread insider presump- tion against patenting, which was often associated with unreliable out- siders, until around the turn of the nineteenth century. To account ade- quately for the later shift to embrace patent protection would require further study, although it seems reasonable to suppose that the increas- ing appeal to brewers of technologies developed in more patent-friendly settings—most obviously, the steam engine—played a major role. Perhaps our most important fi nding, as demonstrated by the case of John Richardson, is that careful selective revealing of information, and a sophisticated approach to the communication of inventions, could allow a “trade in inventions” to thrive even without the use of the patent 91 William Tizard, The Theory and Practice of Brewing Illustrated, 1st ed. (London, 1843). Further editions followed in 1846, 1850, and 1857. 92 Wesley Cohen, Richard Nelson, and John Walsh, “Protecting Their Intellectual Assets: Appropriability Conditions and Why US Manufacturing Firms Patent (or Not)” NBER Work- ing Paper no. 7552 (2000). For some nineteenth century evidence, see Petra Moser, “Innova- tion without Patents: Evidence from World’s Fairs,” Journal of Law and Economics 55 (2012): 43–74. 93 For an insightful discussion of the persistent heterogeneity of innovation strategies within industrial sectors, see Martin Shrolec and Bart Verspagen, “The Voyage of the Beagle into Innovation: Explorations on Heterogeneity, Selection, and Sectors,” Industrial and Cor- porate Change 21 (2012): 1221–53. Alessandro Nuvolari and James Sumner / 120 system. Furthermore, even when inventors opted for an appropriability strategy based on patent protection, patents were only one ingredient of the mix. Promotion through technical books and pamphlets, drawing on the past strategies of both patenting and non-patenting innovators, remained a key strategic factor. Given the growing interest in markets for technologies, it would be instructive to establish whether this pat- tern was peculiar to the English brewing industry or is representative of a more widespread phenomenon. . . . ALESSANDRO NUVOLARI is associate professor of economic and social history at Sant’Anna School of Advanced Studies, Pisa, Italy. His research in- terests are mostly focused on the development and diffusion of steam-power technology and on the relationship between patents and innovation during the Industrial Revolution. His recent publications include “Bennet Woodcroft and the Value of English Patents, 1617–1841,” Explorations in Economic His- tory (2011) (jointly with Valentina Tartari) and “Technical Choice, Innova- tion, and British Steam Engineering, 1800–1850,” Economic History Review (2009) (jointly with Bart Verspagen). JAMES SUMNER is lecturer in history of technology at the University of Manchester, United Kingdom. His current research addresses how special- ized knowledge based on scientifi c authority comes to be accepted in com- mercial production cultures, focusing on the relationship between chemistry and beer-brewing in Britain and Ireland. Pickering & Chatto will publish his monograph, Brewing Science, Technology, and Print, 1700–1880, in sum- mer 2013.