In a book stemming from her 2014 doctoral dissertation, Patt takes on the so-called Formulae imperiales, a group of models, mostly for imperial diplomas, that stem from the reign of the Carolingian emperor Louis the Pious (r. 814–40). The formulas, fifty-five in number, were copied in Tours sometime around 830, entirely in Tironian notes, into a rather disorganised manuscript that also includes theological texts and capitularies (BNF, Paris, ms lat. 2718). They get their name from their most recent editor, Karl Zeumer, who in 1886 published them in his edition of the early medieval formula collections for the Monumenta Germania Historica (MGH LL 5). There is no question that they are closely connected to Louis's chancery; they were only inconsistently anonymised, enough to tell us that most were taken from real, actually issued diplomas. Some of these diplomas still exist, but some have been lost, so that the Formulae imperiales both significantly expand the range of Louis's surviving diplomas and tell us more about the legal matters with which Louis's chancery concerned itself. Moreover, in contrast to the surviving diplomas, a striking number of the Formulae imperiales deal only with laypeople. They represent, therefore, a very important resource for studying everything from imperial politics and the practices of the imperial chancery to Carolingian society at large.

A central problem with using the Formulae imperiales as a source, however, is grasping their relationship to actual chancery practice. Their name indicates how they have traditionally been understood: as an official collection of models issued by the imperial chancery that chancery scribes were supposed to rely on. Patt tests this assumption by comparing the formulas comprehensively with Louis's over 400 surviving diplomas, an effort made possible by her participation in the project headed by Theo Kölzer at the University of Bonn to complete the long-awaited new edition of Louis's diplomas for the MGH (published in 2016).

Patt first sets the Formulae imperiales against the backdrop of the other early medieval formula collections. She engages thoroughly with Alice Rio's Legal practice and the written word in the early Middle Ages: Frankish formulae, c. 500–1000 (Cambridge 2009), which stands as the most comprehensive study of the early medieval formulas to date. Patt agrees with Rio in seeing the formula collections as a Carolingian phenomenon that was oriented towards actual legal practice. She also agrees with Rio that many of the collections that Zeumer identified and edited for the MGH are editorial fictions that must be used with extreme caution. Nevertheless, Patt is reluctant to press Rio's scepticism too far; some of what Zeumer produced, she argues, is perfectly sound. This includes his texts of the Formulae imperiales, which – given that they survive in only one manuscript – do not pose the same editorial issues as other formula collections.

Then we are introduced in depth to Louis's diplomas, to the editing and publishing history of the Formulae imperiales, and to the history of their interpretation. Along the way, Patt repeatedly points out where past scholars have run ahead of their evidence, for example by too firmly identifying the author of the Formulae imperiales as the chancery director and abbot of Tours, Fridugis (d. 834). She rejects the proposition that Paris, ms lat. 2718 – a heterogenous and disorganised collection of texts copied into a codex made up of irregular parchment scraps – was a product of a so-called leges scriptorium at Tours dedicated to turning out legal handbooks in the 820s and 830s. Indeed, she is sceptical that such a project existed at all (pp. 123–9).

The meat of the book lies in chapter v, where Patt compares the Formulae imperiales with the diplomas. She anchors her discussion with an extensive appendix – just over one-third of the book's total length – that provides her textual evidence. She finds some complete correspondences between formulas and diplomas (Vollentsprechungen), but notes that they are relatively rare, as are partial correspondences (Teilentsprechungen). She identifies most of the diplomas as Mischformularen, i.e. texts that mix and blur material corresponding to several of the Formulae imperiales, to the point that it is often impossible to link particular language to a particular formula. In these cases it appears that chancery scribes reached for a large body of formulaic material that they knew extremely well, and put individual pieces together variably according to their specific needs, without necessarily having to resort to fixed models. This picture undermines, she argues, the picture of the Formulae imperiales as an official collection that was supposed to serve chancery scribes as a binding model (pp. 171–2). Instead, they were probably copied gradually into a personal reference book by someone, probably at Tours, who had worked in the chancery and who wanted models or examples for his own use.

This is not a book for a wide readership. It is a sober, careful and thoroughly documented research tool for people with an interest not only in the specifics of Patt's subject, but also in the history of documentary culture in the Carolingian period. Nevertheless, I enjoyed reading it; the arguments are strong and well-supported, and Patt's prose is clear and engaging. My critiques mostly concern details about the early medieval formula collections in general. I would disagree with Patt, for example, when she distinguishes between formula collections that show a heavy late Roman influence and ‘true early medieval formula collections’ (p. 24); late Roman legal language remained part of the process of drafting authoritative documents deep into the early Middle Ages. As for her main project, while Patt discusses the other contents of Paris, ms lat. 2718 briefly, I would have liked to see her do so in more depth, to place the Formulae imperiales more firmly against the backdrop of the other things in which the manuscript's compiler was interested.

These, however, are very minor complaints. This is a model study showing how a Carolingian-era collection of documentary formulas was used. It echoes, reinforces and carries forward what has emerged from recent research on other Carolingian formula collections and illuminates the practices of Louis's imperial chancery. It must rank for the foreseeable future as the definitive study of the Formulae imperiales; it is also an excellent resource for the state of the art on early medieval formulas and Carolingian documentary practices in general.