2. The scholarly context of the Archimedes Project

The Archimedes Project is the digital library component of a major research project of the Max Planck Institute for the History of Science dealing with mental models in the history of mechanics. In the context of this project, research is focused on questions concerning the long-term development of mechanics such as the following:

• How could atomism have been invented in Greek antiquity long before the first experiments revealed the real existence of atoms?
• Why did Aristotelian natural philosophy dominate scientific thinking for more than 2000 years in spite of obviously fallacious assumptions such as that of the speed of fall is proportional to the weight of the falling body?
• How was the complicated technology of architecture and machine construction possible without the scientific base that we associate today with such achievements?
• How was it possible to overcome the concepts of classical mechanics by a radically new theory of space and time as it was formulated by Einstein long before physical and astronomical evidence suggesting such a theory was available?

Such questions can hardly find satisfactory answers in the study of the universal structure of the science, as is customary in philosophy, nor in studies focused on specific disciplines during specific historical periods based on a small selection of sources as is customary in historiography.

The Max Planck Institute for the History of Science has therefore undertaken a study of the long-term development of mechanical knowledge which encompasses the vast historical period from the third millennium B.C. until the decline of classical mechanics and currently focuses on scientific as well as practical knowledge from contrasting cultures such as ancient Mesopotamia, classical Greece, ancient China, the world of the Arabic and Latin Middle Ages, and the early modern period of the European engineer scientists.

In this vast period mechanical knowledge has completely changed its character several times. Mechanics has its origins in everyday experience built up in roughly the same way by individuals under extremely different cultural settings such as shipbuilding in the South Pacific or the use of ballistic weapons ranging from blast pipe technologies to throwing boomerangs. For a long time even highly-developed technologies of this kind were not accompanied by explicitly documented mechanical knowledge of the kind familiar from later mechanical treatises.

When the first treatises on mechanics emerged, their practical relevance was negligible. Rather, they profited more from existing mechanical knowledge when formulating theoretical concepts with a potentially unlimited range of application than they contributed to the structuring and advancement of the practical knowledge embodied in professional activities. This relation underwent change only during the Renaissance, when challenging new tasks required that the entire realm of mechanical technologies be drawn on, which led to a realization of the implicit potential of the theoretical resources accumulated since antiquity. This realization was the beginning of an age in which the growing integration of theoretical and practical knowledge accompanied the advancement of technology from a marginal phenomenon to a central element of social productivity in the economy of modern capitalist society. The changing role of theoretical mechanical knowledge was foreshadowed in the realm of ideas where the so-called mechanistic world view emerged and then developed into an almost universal and intuitively self-evident doctrine. This mechanical world view was abandoned only when the rich results of science, to which it had contributed, could no longer be fitted into its framework and when twentieth-centry physics made its seemingly timeless conceptual foundations obsolete.

The study of such long-term processes of the development of scientific knowledge requires not only a general theoretical framework of a sort unlikely to emerge from specialized historiography or from rational speculative syntheses. Such a study obviously also hinges on the analysis of an enormous corpus of sources that serve as its empirical validation. These sources have to cover mechanical knowledge far beyond what is usually taken into account by a history of science that focuses on individual discoveries. What the project at the MPIWG is attempting to reconstruct for the different historical periods covered by the project are the following dimensions of knowledge:

• theoretical scientific knowledge,
• the practical knowledge implicit in the productive use of technology which is a common precondition of both science and technology, and
• the universal knowledge embodied in cultural practices similar in all human cultures.

This may explain the two main aspects of the research which forms the foundation of the Archimedes Project :

• the application of methods and techniques of cognitive science in order to capture the structures that organize the interaction of these three types of knowledge,
• the use of computer-assisted analysis that enables us to extract information relevant to the reconstruction of such structures on the basis of historical sources.