Both volumes under review (hereafter referred to as GEMF) originate from the research project ‘The Transmission of Magical Knowledge’, led by the editors at the University of Chicago. Their main aim is to produce – in collaboration with other scholars – authoritative editions and translations of Greek-Egyptian magical texts preserved (mainly, but not exclusively) on papyrus. Throughout GEMF, the philological attention given to the texts is constantly paired with their treatment as material objects, with their format, layout and paratextual qualities playing an essential role. This is for the greater part an innovative and desirable approach, complementing the study of magical texts through analysis of their textual content.

In contrast to previous collections that in one aspect or another paved the way for GEMF, K. Preisendanz's editions in Papyri Graecae Magicae (1928–1931), H.D. Betz's Greek Magical Papyri in Translation (1986), and R.W. Daniel and F. Maltomini's Supplementum Magicum (1990–1992), these volumes focus solely on the formularies – i.e. the handbooks that provide instructions to perform magical-religious rites (similar, e.g., to collections of medical treatments or cooking recipes) with the so-called activated texts (the curses, amulets or rites specifically naming the individual to attack, protect or heal) excluded. Most of these formularies are (parts of) rolls, codices, single sheets of papyrus or (less commonly) parchment or paper found in Egypt, but other writing media are included: GEMF 27, for example, is a spherical gemstone from the Black Sea region dated to the second/third century ce. The specific focus of GEMF on the formularies is aimed at highlighting the evolution of magical-religious knowledge over time in a longue durée process. This end is served by two key editorial choices: the chronological ordering of the texts and the inclusion, together with the Greek texts, of Egyptian and bilingual (Egyptian-Greek) formularies. The chronological disposition – which in Volume 1 spans the second century bce to the fourth century ce – allows readers to follow the development of the magical formularies in a diachronic perspective, while the inclusion of bilingual and Egyptian texts has many advantages over the usual compartmentalised treatment of individual texts in a single language and provides a broader overview of the topic (some Demotic formularies that will be studied by J.F. Quack have not been included to avoid unnecessary overlap).

A thread that unites all the contributions in GEMF is the attention paid to the material aspects of the magical formularies. This ‘material approach’ not only acknowledges that books in antiquity are also a witness to the people who ordered, produced and used them – and as such are not only conveyors of texts, but are also artefacts informing us of the cultural environment and of the local and historical circumstances that led to their production, a process of historical significance in itself – but has also allowed for a series of immediate improvements. For example, it has been recognised that PGM VI and PGM II form a single magical document (now edited as GEMF 30), just as do PGM LVII and PGM LXXII (= GEMF 8). Additionally, the proposed dating of a few formularies could be rectified: for example, PGM I (= GEMF 31) has been convincingly assigned to the second half of the third century ce, about a century earlier than the date given in PGM.

GEMF Volume 1, Text and Translation, provides 54 re-editions of previously published magical formularies. Each text receives a GEMF serial number, but references to the numeration of previous editions (e.g. in PGM or Supplementum Magicum) are provided. It is worth noting that some texts are referred to as PGM LXXXII and higher (e.g. GEMF 13 is equated with PGM CX), although numerations higher than LXXXI do not exist in PGM proper. These PGM LXXXII-and-higher numbers were used in Betz's GMPT. But since this work does not contain critical (re-)editions of the texts, the proper citation of these papyri should be their latest critical edition. For the sake of clarity, it might be a good idea to refer to GMPT numbers as such. The text just mentioned, then, would be indicated as GEMF 13 (= P.Wash.Univ. II 73, GMPT CX).

The editions begin with an introduction providing information about the formulary's dating, provenance and repository, previous editions, translations, and discussions of the text in question. This information is thorough and complete. One misses, however, information about the existence of printed and online photographical documentation of the texts – although the preface (p. xxiii) signals the intent to provide a collection of online images. Following the introduction, the text is given, provided with an extensive and thorough apparatus criticus (more below) and a translation. Particularly welcome is the aim to translate the termini technici as such; a recurring example is the word πράξιϲ, translated as ‘procedure’ rather than ‘spell’ (LSJ s.v., II 4), cf. furthermore the preface at p. xx. In the style of GMPT, footnotes in the translation give the contributors an opportunity for commentary. These are rich in additional information and bibliographical references, although it may be inconvenient for those interested in the original text that remarks on grammar or textual difficulties are found only in these footnotes.

As a collection of new critical editions of magic formularies, the quality and thoroughness of the editions of GEMF must be highlighted. Cases in which the text can be improved are rare and mostly stem from typographical errors: GEMF 15.169, a passage in a rite to obtain a divinatory dream, should run χρη(μάτιϲον) τῷ δ(ε)ῖ(νι) (as in PGM, and confirmed by the picture in R. Daniel, Photographic Edition [1991], p. 9), whereas the new edition mistakenly places the monogram χρη(μάτιϲον) between article and substantive, although the translation correctly has: ‘prophesy to him, NN’. Rare are also the cases where something goes unnoticed in the apparatus and in the translation: GEMF 15.200–1, the end of a divinatory ritual, is printed: χρ(ημάτιϲον) εἰ περῖ τοῦδε, περ[ὶ π]άν|των πυνθάνω. The apparatus records the reading in PGM, περ[ὶ π]άν|των, ‹ὧν› πυνθάνω, but does not inform us that PGM also has χρ(ημάτ)ει(ϲον), corrected into χρ(ημάτ)ι(ϲον), comparing PGM XII 115 and 120. εἰ was printed as a conjunction in Leemans's editio princeps (p. 21), followed by Dieterich (Jahrb. klass. Phil. Suppl. 16 [1888], 802). The translation, if the edition does favour εἰ as a standalone element, should be modified accordingly.

As pointed out, the materiality of these texts is given high priority in GEMF. This also means that the editors have focused on reproducing the layout of the originals in the edition, displaying abbreviations, marginalia, diacritical signs, drawings and symbols in the text. Although this approach is a welcome choice, the adherence to this principle is in some cases ‘reader unfriendly’: non-grammatical forms are printed in the text and corrected in the apparatus, but in GEMF diaereses as well as symbols and abbreviations stylised in their graphic form are inserted in the text, with their resolutions placed in the apparatus, which creates a hybrid between a diplomatic transcription and a running text edition. This applies to the structure of the apparatus itself. The apparatus is thorough in recording readings and emendations. These are combined with grammatical corrections and resolutions of symbols or abbreviations: with such a large amount of material printed in a small font size and with narrow interlinear space, it can be at times difficult to appreciate fully the great amount of work that the team members invested in the apparatus. A more visually readable apparatus would have been welcome. Consideration might perhaps have been given to two separate apparatus, one with grammatical corrections and graphic replications of symbols and abbreviations (as in editions of documentary papyri), another containing readings proposed by different scholars, emendations and similar matters.

In a small number of instances, the apparatus could have profited from revision. A case in point is GEMF 31.5 (p. 384): there the imperative ἀποθωϲον is a misspelling of ἀποθέωϲον (‘deify’, i.e. ritually kill an animal by drowning). The apparatus runs: ‘l. ἀποθ[έ]ωϲον Pr’. This is probably an amalgamation of ‘l. ἀποθέωϲον: ἀποθ[έ]ωϲον Pr’. In ll. 6–7 an undyed strip of cloth to be used in the ritual is mentioned in the dative, ῥάκει ἀχρωτίϲτῳ. The adjective ‘undyed’ (ἀχρώτιϲτοϲ, a syncopated form of ἀχρωμάτιϲτοϲ) was written on the papyrus as ακρωτιϲτωϲ. A χ was then added above the κ, the final ϲ is superfluous (a mistake that originated at some point as a result of misunderstanding an ι originally adscriptum?). GEMF, too, prints a χ above the κ, to represent the papyrus’ appearance, but also enclosures the κ in double square brackets, as well as noting in the apparatus: ‘χ corr. ex κ’. The adjective being a dative, an ι subscriptum is added in the text, but the superfluous ϲ is left unaltered. To realise that the editors correctly regard the final ϲ as a blunder, readers need to look into the apparatus, which, as just mentioned, reduplicates the information given in the text by the brackets: ‘ἀχρωτίϲτῳ χ corr. ex κ’. The apparatus criticus sometimes gives the reading of the text first and the rejected readings afterwards, sometime vice versa, cf. for example the two magical words between line-ending and line-beginning at GEMF 15.205–6. But only in rare cases are the remarks in the apparatus concerning readings and emendations imprecise: the apparatus to GEMF 15.195 (p. 88) ἐπίλεγε τὸν [λό]γο(ν) provides an example, incorrectly ascribing the unrestored reading [. .]γō to R. Daniel, Photographic Edition (1991), which gives [λό]γ̣ō (cf. p. 10). Especially since the editors took the trouble to write the apparatus in Latin, there are cases where the diction could have been refined; for example, at p. 322 to GEMF 21.35 the wording εἶ (2x) might better have been given as εἶ bis, and in other texts ter instead of (3x) and quater instead of (4x). At p. 384 in a note on GEMF 31.12 with a vowel sequence of four upsilons the apparatus notes: ‘υυυυ: alterum υ corr. ex α’. Here, secundum would have been clearer. Or does this notation refer to l. 11, where indeed only two upsilons appear? A confusion also seems to have occurred at p. 422 in the apparatus to GEMF 34.46: the line number has been given incorrectly, an abbreviation is not resolved, and a typo in what follows makes it unintelligible.

It must be stated that these are clearly minor details that do not diminish the overall quality of the new editions. They are nonetheless worth noting since the production of authoritative critical editions with an ambitious and thorough apparatus criticus involves such small details as well.

The second volume under review, Libraries, Books, and Individual Recipes, is a collection of eleven essays, most of which derive from the work conducted on the formularies re-edited for GEMF. Here, too, the contributors display how fruitful it can be to study the magical formularies from Graeco-Roman Egypt as material objects rather than concentrating only on their language and the theological, ritual and technical topics emerging from their texts. This is, to a great extent, a novel approach that complements the traditional historical-philological treatments of the material.

Chapter 1, ‘Anatomy of the Magical Archive’ by K. Dosoo and Torallas Tovar, defines the concept of ‘magical archive’ and analyses the collection, redaction and transmission of magical handbooks from three different archives (the so-called Theban Magical Library, the Hermonthis Magical Archive and the Kellis House 3 Archive), showing how the study of the social context in which these texts were produced improves their general understanding.

Chapter 2, ‘Roll vs. Codex: The Format of the Magical Handbook’ by Dosoo and Torallas Tovar, is a survey dealing with the format evolution of magical formularies seen as part of general scriptorial trends. The data presented will prove to be of fundamental importance, among others, for scholars interested in the evolution of book-practices.

Chapter 3, ‘The Paleography and Dating of the Magical Formularies from Roman Egypt’ by A. Nodar, discusses the hurdles of dating paraliterary texts such as the magical formularies on palaeographical grounds, nonetheless providing useful guidelines for such an endeavour.

Chapter 4, ‘Compositional Patterns in the Paris Magical Codex (GEMF 57 = PGM IV)’ is an abridged version of the groundbreaking contribution by L.R. LiDonnici on the compositional patterns observed in the Paris Magical Codex PGM IV. The study established the view that its 3,274 lines were assembled directly or indirectly from five (or six) previous collections which served as different Vorlagen.

Chapter 5, ‘The Composition of the Demotic Magical Papyrus of London and Leiden (GEMF 16 = PDM/PGM XIV)’ by Dosoo, and Chapter 6, ‘GEMF 60 (= PGM XIII): A Study of Material, Scribal, and Compositional Issues’ by R. Gordon and R. Yuen-Collingridge, apply the methodology developed by LiDonnici to two other magical handbooks. The first deals with the largest surviving magical roll from Roman Egypt, which possibly derives from an original core expanded according to subject and the practical interests of the practitioner(s). The structural analysis of this text suggests that its redactor actively collected recipes and may have profited from collaboration with experts within a larger community of practitioners. The second study focuses on a codex that presents itself as a coherent and authoritative book (‘The Eighth Book of Moses on the Holy Name’) containing two versions of an extremely complex Graeco-Egyptian formulary, possibly produced by combining different Vorlagen of related but at times divergent rituals. The numerous and complex paratextual features in this text are accurately described, showing how a broader and structure-based approach can be fruitful for understanding history and meaning of this text.

Chapter 7, ‘GEMF 74 (= PGM VII): Reconstructing the Textual Tradition’ by Gordon and R. Martín Hernández, deals with a long magical roll, apparently a compendium that combined different Vorlagen at different stages. By analysing the four thematic blocks recognisable in the roll, as well as their content and structure, the authors convincingly show how GEMF 74 places itself at an intermediate point between the more thorough and ‘ambitious’ formularies and the shorter, clearly client-oriented collections of recipes (or single-sheet recipes).

Chapter 8, ‘GEMF 15 (= PDM/PGM XII): Production and Use of a Bilingual Magical Formulary’ by P. Sarischouli, focuses on a second-century roll containing 28 procedures in both Greek and Demotic. The structure of this roll is examined, showing that, whereas some material must have been incorporated from loose sheets, other sections are linked in thematic groups. The compiler of this text seems to have been someone versed in both languages and in both religious and mythical traditions. GEMF 15 is thus characterised as a ‘one-volume library’ produced by someone associated with the milieu of the Egyptian temples for a patron in the Graeco-Roman upper class.

Chapter 9, ‘The Composite Recipes in GEMF 57 (= PGM IV) and How They Grew: From Practical Instructions to Literary Narratives’ by Faraone, deals with the sections of PGM IV where hexametric and iambic hymns appear and different ritual performances are combined (by expanding an existing formulary with additions at its end or by insertions). In a few cases, such composite rituals are introduced by fictional rubrics naming the auctoritates who invented or collected them. Faraone argues that some of these longer recipes, rather than having a ritual purpose, can be regarded as narrative stagings that mix Greek myths with an Egyptian hieratic setting, thus aiming to captivate readers.

Chapter 10, ‘The Rationale of Multi-Purpose Praxeis in the Formulary Tradition’ by Gordon, deals with 19 special procedures (out of the 443 listed in Betz's ‘Table of Spells’), in which a ritual practice is not specifically aimed at a particular scope, but rather is presented as efficient for multiple necessities. Gordon identifies in the ‘multi-purpose procedures’ an opportunity for the practitioners to rearrange elements of the magical-religious Egyptian tradition in an original way, updating them in Greek. The main goal, rather than producing employable formularies, might have been to preserve in written form a tradition that was perceived as threatened by its appreciators.

Chapter 11, ‘The Traffic in Magical Recipes: Single-Sheet Formularies as Prompts for Oral Performance’ by Faraone, concentrates on rituals that circulated on single papyrus sheets (or on other materials). A comparison is given with Galen's description in his Περὶ ἀλυπίαϲ of how he had collected (and distributed) recipes for a treatise of his on drugs. Single-sheet formularies might have been extracted from handbooks according to the specific needs of a client, but authoritative recipes that circulated as single texts may also have been added to larger collections. By considering a series of single-sheet formularies, among which are three lamellae found outside of Egypt, Faraone highlights the preponderance in such texts of formulas to be ritually uttered. As such, they might have served as prompts for oral rituals.

In conclusion, these first two volumes of GEMF, despite the few weaknesses noted, will be appreciated for updating a corpus that has become, in several ways, obsolete. GEMF offers authoritative editions of magical formularies, combining their philological treatment with painstaking attention paid to the material aspects of the objects that carried the texts. How fruitful the results of this approach can be, is evidenced by the second volume, whose contributions are likely to spark new debate. Thus, the series will be positively received both by specialists in the field, whether their main interest is more philological or more historical, and by scholars from neighbouring disciplines who wish to expand their research in so-called magical texts. To mention a single such example: the attention paid to the material and paratextual features in these artefacts will prove to be a treasure-chest for linguists interested in multimodal aspects of ancient texts and their semiotic features. The inclusion and consideration of texts written in Greek as well as Egyptian and the diachronic treatment of these texts covering an extensive time frame complete the picture of GEMF's innovative traits, which will become a reference point for anyone dealing with Greek-Egyptian magical texts, whatever their scholarly approach to these texts may be.