Ross Thompson gives us an insightful account of technological change and industrial development in the United States between roughly 1790 and 1865. He focuses on six key industries: textiles, clocks, printing, steam engines, firearms, and woodworking, but eight other industries receive attention. Thompson compares and contrasts the different industries (and their leading inventor-entrepreneurs) in order to tease out interesting similarities and differences in inventive paths, which include the direction and relative importance of cross-industry technological flows, different regional profiles, the differing educational backgrounds of leading actors, the “social basis of innovation,” and much more. Thompson's arguments are, for the most part, sensible, and his book is impressively researched and well written.

Thompson argues that, “Technological change was a process of institutionally structured learning in which innovators gained knowledge through established communication channels, and innovations spread knowledge in ways that fostered further innovation” (p. 4). Throughout he emphasizes the importance of “networks” and “technological centers” in spreading and deepening industrial development. In his view, the pace of innovation increased markedly after roughly 1835 in part because of the gradual emergence of more mature networks and the increase in “cross-industry spillovers” (p. 316). There was not a single outcome—rather, the nature of the technologies, industry peculiarities, location, and markets shaped different paths for different industries, with varying degrees of geographical concentration, spillovers into other industries, and the like.

I particularly commend Thompson's ability to step back from the detailed nuts and bolts of his story to reflect on the bigger comparative picture, including some strategic comparisons with changes in Western Europe. Many of his non-quantitative figures and tables, such as his family trees of textile, steam engine, and firearms inventors, add punch to his argument by detailing the enormous amount of inventive and business activity emanated from the ground-breaking enterprises of Samuel Slater, David Wilkinson, and a few other innovators. For example, a remarkably high percentage of important American textile machinists either worked for Slater or were trained by men who had worked for him.

The recurring theme on the role of “networks” deals with both public and private institutions. Government policy encouraged innovation in a number of ways. At one extreme was the firearms industry where the U.S. armories at Springfield and Harpers Ferry stimulated flow of technical and organizational information (in addition to contracts) to manufacturers. The armories promoted the technologies and methods that would eventually make interchangeable parts a reality. Manufactures toured the armories, exchanged ideas, and received access to important open-source technologies.

Thompson describes the contributions of a number of civic institutions that promoted the open exchange of ideas and technologies. Among the most significant was the Franklin Institute in Philadelphia (1824), which held seminars, classes, and fairs; assessed inventions; and published a journal. This institute was one of the many “networks” that Thompson calls to our attention.

Although my overall assessment of this book is extremely favorable, I do have some quibbles. The terms “networks” and “technological centers,” are a bit amorphous. These include the Patent Office, a host of other government agencies, patent agencies, loose ties among inventors, more formal ties created by civic organizations, occupational groups and even the occupations themselves, the educational establishments, newspapers and more specialized publications (Scientific American was perhaps the most prominent), nearby firms, and cities. Centers need not be physical places—”Machinists … had become a technological center” (p. 158). At times, Thompson is more restrained, but a little more precision in defining what he meant might have strengthened his argument.

Thompson too often equates innovation with patents and even with patents in his few select industries. Let me make it clear that he knows better; at points he notes that these identities do not hold, and elsewhere he explicitly confronts (although I think inadequately) Petra Moser's thesis that many useful industrial innovations were not patented.Footnote ¹ But he has many lapses. In addition to Moser's case, there was a beehive of unpatented transformative inventive activity in the agricultural sector.Footnote ² If one includes such activities, the South and West would not look nearly so barren relative to the North East and Middle Atlantic regions, and rural areas would not appear to be so relatively infertile compared to cities. This bears directly on Thompson's findings that place innovative centers (more narrowly defined) within cities with engineers, etc. The innovative process was far more diffuse and complicated than even Thompson makes out.

These caveats aside, this is an important book, on an interesting subject, by a serious scholar.