This is a quite remarkable book. Andrew Riggsby set out to examine what he terms Roman Information Technologies: this includes lists and tables and weights and measures and also depictions of spaces both real and imaginary. R.'s carefully structured argument proceeds through forensic analysis of key documents of all kinds, but at the heart of the investigation is the making and use of the objects in question. He has much to say of vanished documents too, of what Claude Nicolet termed La mémoire perdue. The result is a quite original account of how Romans of all classes measured up and understood their social world. Mosaics of Knowledge has implications for most fields of Roman cultural history, from art history to economics and government.
The chronological centre of the book is the early empire and its geographical focus is on the Latin-using societies of Italy and the western provinces. R. argues that some practices and conventions did differ between the Greek- and Latin-using parts of the empire. The documents he considers range from those that use a great deal of writing (fasti and alba for instance) to those are entirely figural (the sacro-idyllic landscapes). Many are graphic products that mix words and images, from monumental plans to inscribed weights and measures.
R. devotes a good deal of time to defining terms. Often, he argues, confusion has resulted from imposing anachronistic definitions on ancient evidence or simply from disagreements about what terms like ‘perspective’, ‘table’ and ‘map’ mean. Depending on how one interprets that last term, for example, the ancient world had no maps at all or else made wide use of them. R.'s solution, having discussed all the rival views, is that ‘A map describes some physical space by establishing parallels, piece by piece, between elements of the map and reality’ (180), but he allows that no one definition serves all purposes. A similar approach allows him to make good sense of ancient perspective, not as a system based on a single vanishing point, but as a set of practices that employed consistent and effective conventions to represent spatial relations between component parts. Even ‘tables’ turn out to be more complex than is often admitted: the columns need to be as meaningful as the rows, and a series of linear sentences in roughly the same form is not enough. R. finds few true tables or nested lists in the Roman world.
Several general themes emerge. Romans were rarely concerned with very precise measures, whether of time, quantity, or weight, and they invested little effort in establishing universal standards against which local ones could be controlled. What mattered in the marketplace was fairness, so use of the same local measures mattered more than adherence to universal standards. Approximate measures were often employed, especially in time-keeping. Ratios and proportions mattered more than exact quantities (hence the wide variation in archaeologically attested building ‘modules’ or grain measures). Exact dimensions of building materials mattered less than consistency within a given building. Similar principles extended to other economic activities, including tithing and sharecropping. More worryingly, quantities were often stylised, some numbers much preferred to others, numbers often rounded up or presented in excessive detail to suit rhetorical and ideological aims.
R. knows that technological innovation often depends on prior developments. One reason that Roman texts had so few illustrations, he suggests, was the difficulty in precise reproduction of the kind that was only easy after the development of wood blocks. Measuring very small quantities or very small intervals of time would have been difficult with the machinery available at the time. Historians of the Scientific Revolution often point to its dependence on prior advances in glass-blowing, metallurgy and metrology. But Mosaics of Knowledge does not present an argument for the primitivism of Roman knowledge, and certainly not for technological determinism. R. repeatedly emphasises the importance of cultural preferences. Ancient texts made little use of indices, of pagination and of numbered chapters, not because these paratexts could not be devised, but because they were not desired. R. is good at spotting exceptions to the rule. Ephemeral plans, diagrams, blueprints and maps were made and used by many kinds of specialists, but were rarely thought appropriate for inclusion in literary works. R. argues for what he terms ‘resistance to metonymy’, a preference for thinking in terms of real cups and paces than abstract quantities. He suggests connections were not easily made between measures of duration and time-keeping. Other connections that seem habitual to us were unfamiliar to them.
From our perspective, there is often a sense of lost opportunity. Romans were more interested in writing down than in looking up, they could make precision instruments but mostly used them as curios, they gathered a mass of data but did not look for patterns in it. Estate owners documented costs and yields but did not use past performance to inform future planning. When they needed to be precise—in land measurement, or in planning large buildings, managing aqueducts or supplying military bases—they had the tools they needed. But they rarely made more use of precision measurement than was needed locally for the task in hand.
