Between 1986 and 1996 researchers at the AT&T Bell Laboratories Adaptive Systems Research Department curated thousands of images of handwritten digits assembled by the United States Postal Service to train and evaluate artificial neural networks. In academic papers and conference literature, and in conversations with the press, Bell Labs researchers, executives and company spokespeople deployed the language of neurophysiology to position the systems as capable of codifying and reproducing feats of perception. Interpretations such as these were pivotal to the formation of brain–computer imaginaries that surrounded the development of the systems, which obscured the institutional infrastructures, clerical and cognitive labour, and the manipulation and maintenance of data on which feats of ‘recognition’ depended. Central to building the group's networks was the development of data sets constructed in consort with the US Postal Service, which arbitrated between the practicality of conducting research and the representation of an extensive catalogue of possible forms and permutations of handwritten digits. These imaginaries, which stressed a likeness with the human brain, were compounded by the promotion of ‘successful applications’ that took place under the AT&T corporate umbrella and with the essential support of US Postal Service workers to correct system errors.

During the Cold War, UNESCO played a major role in promoting science education across the world. UNESCO's Programme in Integrated Science Teaching, launched in 1969, placed science education at the heart of socio-economic development in all nations. The programme planners emphasized the role of science education in the development of human resources necessary to build a modern nation state, seeking to build a scientific and engineering mindset in children. UNESCO's interest in science education drew inspiration from early Cold War curriculum reforms in the United States, where scientists, psychologists and teachers promoted science education as a way to enhance the scientific and technical workforce and to counteract irrational tendencies. While US curriculum reformers were concerned about the quantity and quality of science teaching in secondary school, UNESCO wanted to introduce science as a topic in primary, secondary and vocational schools, promoting integrated science teaching as the best way to do this. From the outset, the term ‘integrated’ meant different things to different people. It not only entailed less focus on scientific disciplines and scientific method strictly defined, but also more on teaching children how to adopt a curious, experimental and engineering approach in life. By the end of the Cold War, UNESCO abandoned the idea of integrated science teaching, but it has a lasting legacy in terms of placing ways of teaching science to children at the heart of modern society.

Drawing from recent theoretical insights on the circulation of knowledge, this article, grounded in real-world examples, illustrates the importance of ‘the site’ as an analytical heuristic for revealing processes, movements and connections illegible within either nation-centred histories or comparative national studies. By investigating place instead of project, the study reframes the birth of modern rocket developments in both China and India as fundamentally intertwined within common global networks of science. I investigate four seemingly disconnected sites in the US, India, China and Ukraine, each separated by politics but connected and embedded in conduits that enabled the flow of expertise during (and in some cases despite) the Cold War. By doing so, it is possible to reconstruct an exemplar of a kind of global history of science, some of which takes place in China, some in India, and some elsewhere, but all of it connected. There are no discrete beginnings or endings here, merely points of intervention to take stock of processes in action. Each site produces objects and knowledge that contribute to our understanding of the other sites, furthering the overall narrative on Chinese and Indian efforts to formalize a ‘national’ space programme. The focus on these four sites reveals a global network of science in motion, linked by the common goals of building powerful rockets to explore space.

This article explores an early computer algebra system called MACSYMA – a repository of automated non-numeric mathematical operations developed at the Massachusetts Institute of Technology from the 1960s to the 1980s, to help mathematicians, physicists, engineers and other mathematical scientists solve problems and prove theorems. I examine the extensive paper-based training materials that were produced alongside the system to create its users. Would-be users were told that the system would free them from the drudgery of much mathematical labour. However, this ‘freedom’ could only be won by adapting to a highly disciplined mode of problem solving with a relatively inflexible automated assistant. In creating an automated repository of mathematical knowledge, MACSYMA developers sought to erase its social context. However, looking at the training literature, we see that the social operates everywhere, only recoded. This article uses the paper-based training materials to uncover the codes of conduct – both social and technical – that coordinated between the users, developers and machines that constituted the system.

The analysis of a painting attributed to the rangaku scholar Shiba Kōkan is the occasion to recover the genesis of a stereotypical narrative of Japanese scientific modernization and to survey the role of books in the intellectual life of Tokugawa naturalists. For the practitioners of materia medica (honzōgaku), the knowledge of nature began and ended with, and in between continuously referred to, books – Chinese, Japanese and later ‘Dutch’. Canonical texts gave scholars terminology, taxonomy, philosophical justification and legitimation, but not all books had equal value in affecting scholars’ practices. A precise hierarchy, in fact, organized texts, from canonical encyclopedias to private fieldnotes, into ‘textual institutions’ that encouraged further research at the same time that they regulated and framed scholars’ cognitive claims.

Little historical work examines the problems, practices and values of ‘machine learning’ as it was understood and justified within pattern recognition research communities prior to the 1980s. This omission has led to a failure to appreciate how the efficacy of machine learning was often justified by its perceived capacity for ‘originality’ rooted in machine (and human) subjectivity. This paper examines why and how early 1950s pattern recognition researchers came to see ‘machine learning’ as a technical and epistemological set of nominalist strategies for performing induction and abduction given incomplete, complex or contradictory information that might also spur ‘creative’ insight in such diverse activities as political judgement and scientific inquiry. I document local research problems, epistemological commitments, institutional contexts and the circulation of ‘machine-learning’ practices and values through three cases of early-career researchers imagining, building and programming digital computers to ‘learn’ from 1950 to 1953. This machine learning implemented in learning programs came to be seen by some researchers as more efficacious descriptions of the natural and social world because these descriptions were perspective-dependent, profoundly contingent and contextually non-exhaustive. Often indexed as colloquial appeals to greater ‘generality’, conceptions of machine learning's efficacy as the capacity to make meaning from contradictory information continues to inform contemporary debates regarding artificial intelligence, society and possibility.