University of Michigan Press (2021) 360pp £85.95 (ISBN: 978-0472132409)

R. is Associate Professor Emeritus in the Department of Historical Studies at the University of Toronto and holds classics-based degrees from Toronto, Oxford, and Harvard Universites. Her output has mostly involved comparative approaches to the use of numbers and quantified information in classical texts.  Quantifying Mentalities is the end-product of a 25-year research project, which developed a relatively simple coding system that ostensibly provides a numeric profile for any author or work.  It was pleasing to read that the many research assistants involved in this project across the years were acknowledged by name.  They handsomely deserve the recognition.

The book begins by describing the methodology of the project and explaining that it is essentially philological and stylometric.  It goes on to set out the results of the application of the method to six of the earliest, largely surviving works of Ancient Greek historiography: Herodotus, Thucydides, Xenophon (Anabasis and Hellenica), Polybius, and Diodorus Siculus.   Chapter 2 deals with 'Time' numbers and subsequent chapters explore "Distance and Size', 'Military', 'Population' and 'Money'.  Further chapters (7-9) then develop R.'s thoughts on the qualification of numbers, the problems associated with the translation of numbers and the numeric profiles of the writers whose work(s) she has analysed.  Each chapter dealing with a particular type of number has a succinct conclusion which distills the overall findings.  There is also an extremely readable, short 'Conclusion' chapter which signposts the various ways that the data from this project could be used further.  In a sense, this book is setting out a challenge for historians in the future.  Whether the challenge is accepted remains to be seen.

R. writes in an engaging first-person style that is easy to follow and helps to keep the general thrust of the narrative in focus.  At the start of most of the chapters she uses some very good, modern examples to illustrate and contextualise what she and her team were attempting to do with the ancient texts.  Indeed, Appendix L dissects a press report about the Obama/Clinton US presidential campaign from 2008 using the same methods.

Virtually every page has several footnotes which greatly augment the conversation occurring in the text proper.  These footnotes function very well; they tend to de-clutter the text which, given that it contains many numbers, might have veered towards unreadable without them.  The bibliography is extensive and extremely well laid out.   The appendices (of which there are 12) are highly detailed and provide a great deal of potential source material for future historians.  Approximately one-third of this book is devoted to these appendices.  The basic data tables are available for consultation on R.'s dataverse site: https://dataverse.scholarsportal.info/dataverse/rubincam where there is a folder containing over 50 spreadsheets (Excel format) to download.  Although the link above is active for the time being, the Scholars Portal Dataverse is currently changing to Borealis.  The same data is available at: https://borealisdata.ca/dataset.xhtml?persistentId=doi:10.5683/SP2/Y4JDLP

There are many interesting findings from R.'s studies.  For example, in the chapter dealing with "Time Numbers" she points out that "The tendency for numbers to concentrate at the lower end of the range is immediately obvious.  On average, numbers ≥20 make up only 33%, and numbers ≥100 only 8% of the total." and she goes on to explain that this pattern is (more or less) replicated across the five historians.  R. also relates (in "Distance Numbers") an amusing tale from the 1960's of Prof. Winter (University of Toronto) demonstrating by actual experiment that his feet were a few centimetres longer than those of Heracles.  Elsewhere, R. points out that the use of "Money numbers" in the ancient texts is remarkable for the scarcity of the evidence.  The average across all the works studied is only 4.7% of the totals.  It is also the category with the least variation which suggests that the quantification of money was less important for the Greek historians than the quantification of troop numbers, casualties, time and distance.

The book is lacking any diagrams and charts which, in some instances, might have been a better way to convey the underlying messages.  It needs to be recognised that this book contains a great deal of data, but it is a different matter to transform this into information - R. is gently challenging others to do this.  For example (Table K.1) , the numeric profiles assigned to the five ancient authors (there are two entries for Xenophon) are presented as percentages.  It is easily within the ability set of statisticians to discuss the significance of these numbers.  The fact that 24% of the numbers in Herodotus are of 'Time' type, but only 16% of the same are present in Thucydides is data.  What it actually means and whether or not the difference is significant is information.  Perhaps R.'s follow-up works will deal with this; perhaps others will take up the baton.  Indeed, she hopes that "...it should have opened the eyes of every reader to some important aspects of ancient historiographic texts that have not hitherto attracted systematic scholarly attention."

At nearly £86, this is not a book for the casual reader but it certainly has a place in university libraries.

John Timney