This remarkable book is the fruit of Sara Chiarini’s [C.] long-term involvement in the creation of the TheDeMa electronic database (Thesaurus defixionum Magdeburgensis), which was conceived by Martin Dreher as part of a larger project on cursing in the pursuit of the principal’s perceived rights, and funded by the Deutsche Forschungsgemeinschaft (2015-2018). The aim was to provide a digital resource that combined reliable versions of both Greek and Latin texts with a variety of search functions that would make it possible not just to capture individual words and word-strings but also potentially, by affording a neutral overview of all such items relevant to a specific point, to correct arguments based on lack of knowledge, selective citation or mere impressions1. The book forms the lightly-revised text of C.’s Habilitationschrift at Magdeburg, submitted in October 2019, and represents, in effect, a return to Audollent’s model of a substantive and enduring Mediterranean curse-tradition transcending differences of language (but of course including local styles and variants), after the recent series of monolingual studies and catalogues accompanied by texts printed in extenso, which have mainly focused on Latin/Italic texts2.

C.’s primary aim is to show that the traditional categorisation of curse-tablets by supposed genre —again a tradition deriving from Audollent— such as judicial, competitive etc., should be side-lined, though perhaps not entirely given up, in favour of a differentiated taxonomy based on three structural aspects of these written curse-texts, the object (i.e. the target) of the curse, the action envisaged, and the agent who is required to carry out the action (i.e. the explicit or implicit addressee of the text). After all, in almost half of all cases, especially where the text consists simply of names, the genre cannot be established, yet the strait-jacket of generic classification often impels scholars to attempt to assign one. C. also rightly warns repeatedly against assuming that the presenting curse itself represents the true grounds of dispute or enmity, which may have been quite different; but the curse-moment, say a court-case, represents a point at which the target is especially vulnerable to divine attack.

This major aim is flanked by four other positions sustained throughout the book. First, that the common assumption that curse-tablets are classifiable as ‘magic’, deriving from a late nineteenth-century crypto-Christian commonplace, is based on a fundamental misunderstanding of the subjective intention of the principals involved, which was to obtain justice for themselves by appealing to appropriate denizens of the (locally-conceived) Other World. Such a subjective view of justice did not mean re-establishing a supposedly level playing-field but rather the reversal, whether temporary or long-term, of an existing asymmetrical social situation to the benefit of the principal. The contested term magic thus hardly occurs in the book. Secondly, that, as appeals to gods aimed at helping the principal in an intractable situation, written curses are to be classed as (religious) prayers ( Fluchgebete ). A major implication of this is that the class of ‘prayers for justice’ can in fact be extended to cover all curse-tablets and not just the group identified as such by Henk Versnel 3 . Furthermore, that we can dismiss the idea that ‘binding curses’ as a rule constituted aggressive magic in the eyes of those who wrote them. This also reminds us to be critical of another widespread crypto-Christian assumption, that prayer ought not to benefit the speaker if it entails harming others. Thirdly, in consideration of the sacral or religious context, that we should adopt the term devotio , with its implicit link to the vow and the votive, in place of the traditional nominal term defixio , which has no connection with the notion of vows or prayers, but only with the ritual action of pinning down. At any rate, the entries in TLL suggest that devotio was a word in common use in the imperial period for curses in general 4 . Fourthly, in agreement with Olivier Dufault, that very few curses were written by ritual specialists rather than by the principal 5 . The sheer variety of styles, formats and hands would anyway support such an inference. Moreover, there is little evidence for the use of handbooks or manuals (p. 121). Granted that the divine addressees are assumed to be thoroughly familiar with ritual language, and indeed to expect it, familiarity with (aspects of) ritual language was very widespread in antiquity, so that the occurrence of such language in curse-texts by no means supports the assumption that the authors of such texts must have been ritual specialists (p. 44). 6

At first sight, the strategy chosen by C. to exemplify these claims, namely the detailed study of that well-worn topic, the formula, might well seem odd. It has, after all, been associated since Eugen Kagarow precisely with attempts to classify curse-tablets as magical texts —indeed, Amina Kropp actually tried to establish a type of speech-act she observed in her Latin texts, which she termed ‘transformatives’ as specifically ‘magical’ 7 . C. however evidently came to the conclusion that, in order to demonstrate that Greek and Latin curse-tablets were individual strategies for confronting otherwise intractable situations, they needed to be analysed in detail; and, given the sheer number of texts involved, the only way to do that was to simplify the problem by modelling the repeated rhetorical patterns they present, i.e. their so-called formulae. As it happens, in re-addressing Kagorow’s effort, Amina Kropp had recently (2008) provided a suitably sophisticated model for the Latin texts, using historical pragmalinguistics based on Austin-Searle speech-act theory. Kropp’s basic distinction was between formulae that incorporate addresses to divinities, who are in effect the agents, and those that do not, namely ‘manipulative’ formulae defined as a special class of ‘transformative’ directives. In addition, she recognised, but hardly discussed, a range of subsidiary formulae that function in different ways as illocutionary reinforcers 8 . C.’s take on formulae is somewhat different, since she views the principals’ ability to deploy such elements as an index of their competence in communicating the message to the addressee, explicit or implicit: she does not assume any substantial difference between the two modes.

In outlining her model of formulaicity (§ 2.1-3, pp. 40-47), C. makes use of the work of the applied linguist Alison Wray, who views the communicative function of formulaic sequences under three heads: the manipulation of the addressee(s) by means of commands, requests, politeness markers, bargains etc.; asserting one’s separate identity by means of turn-holders and personal turns of phrase; and establishing membership of an implicit group by using institutionalised forms of words and ritual phrases, while yet asserting a privileged place in that group, say by using threats, quotations, and special forms of address 9 . According to Wray, ‘formulae’ in this linguistic sense are word-strings of different types that seem to be stored and retrieved whole in the memory; in order to function as a competent, fluent, adult one needs to accumulate a wide range of such sequences – up to 70% of daily communication is indeed to some degree formulaic; addressees are more likely to respond to a message if they have heard the form or pattern before; and there can be no clear boundary between ‘holistic’ or ‘prefabricated’ communication of this kind and ‘creative’, i.e. analytically segmented and recombined, language, since competent speakers constantly adapt synthetic templates or patterns to new uses.

Use of this looser, more inclusive, model of formulaicity allows C. to escape from the rigidity of earlier approaches and explore both ‘prefabricated’ and ‘creative’ aspects of her material, with special emphasis upon the Greek evidence 10 . In quantitative terms, two-thirds of the book (§ 2.4, pp. 47-186) are devoted to the first, the detailed analysis of formulaicity in the texts, followed by a short section on the central communicative issue of prayer, viewed as a challenge to gain the ear of one or more deities (§ 2.5, p. 186-205). Creative, or rather subjective aspects of curse-texts, which add to, or go beyond formulaic elements, are discussed in Part II (§ 3 . 1-4, 205-86), under the rubric of individualisation and the Lived Ancient Religion approach associated with Jörg Rüpke in Erfurt 11 . The brief final section (§ 4, pp. 287-99) picks up the theme of a subjective sense of justice along the lines established by Martin Dreher and Werner Riess 12 .

The analysis of formulae is divided into three main sections of rather different lengths: those relating directly to the object (in her terms patiens , or target) of the curse (§ 2.4.1, pp. 48-95) ; the action (actio ) envisaged against the object (§ 2.4.2, pp.  95-173); and the agent ( agens ) who is to carry out the action (§ 2.3, pp.  173-186). In principle, she proceeds from the lowest to the highest degree of subjectivity in the language used, though this is not always practicable. Thus under patiens we pass from discussion of the social identity – naming practices, additional specifications such as domicile, status and profession – to the corporeality of the target, as represented by the different kinds and intensities of lists of body-parts (pp. 61-85). Here C. distinguishes between ‘holistic’ and ‘specific’ or ‘targeted’ lists, though in my view we should rather speak of local or regional styles and of periodisation here 13 . The section on actio (2.4.2, pp. 95-173) lays out the expressions detailing types of harm under three heads according to the grammatical subject chosen: the harm the target should suffer; that a deity should inflict; that ego (the principal) wishes. Here the characteristic rhetorical features are cumulation and listing, with small ‘curse-units’ added one to another, and different formulae with the same function deployed, following the recognition that repetition is the basic technique of insistence. The sub-section on the target’s sufferings (§2.4.2.1) is largely devoted to an expansive analysis of the sources of images for ‘persuasive similes’ (pp. 101-34). The different types of ego’s commission to the deity (transfer, threats, reward), the function of shifting temporalities, the mental process of projecting the target onto the material base of the text, and centrality of the individual’s own act of inscription are the topics of the third (§2.4.2.3, pp. 150-86). The much shorter section on the agents ( agens ) is mainly devoted to the issue of whether divine epithets are chosen for their relevance to the task, with a glance at the few recorded historiolae (§ 2.4.3, pp. 173-186).

The coda to Part 1, chapter 2, though short, forms a crucial step in C.’s overall argument. Here she attempts to shift curses out of the moral dustbin where they are usually placed into the much more positive category of prayer. In doing so, she relies on D. Aubriot-Sévin’s argument that εὔχεσθαι denotes an action for the benefit of the claimant, who implicitly feels justified in making the claim 14 . It is the modalities by which the divine powers are approached that provides evidence of the supplicant’s attitude; and the modalities of curse-texts do not differ significantly from those of the wider class of prayers (p. 189-91). The supposed morality of specific prayers cannot be the issue here, since prayer is not inherently about ethics but about timely communication with the divine in an existential situation. The argument is supported by an extended defence (pp. 193-99) of the position that curses were indeed sometimes considered by the principal as equivalent to vows, such that success merited a reward to the divinity. The very word devovere in Latin implies such a connection with the votive.

As I have mentioned, Part II, devoted to individual variation of curse-language, is a great deal shorter than Part I. Moreover it is not presented as a model but as a rather discrete group of individual tactics, and thus in the nature of things difficult to organise into a coherent argument. Rejecting speculation about the emotional functioning of curse-tablets, such as ‘catharsis’ or psychological relief, C. treats her chosen material in terms of the evidence it provides for individualised religious action. Again there are three sub-sections: the ‘situative level’; evidence of emotional states; and self-presentation as a juristic person with rights. The first is mainly devoted to the rarity of provision of justifications or reasons for action (half of all cases here are thief-texts, for which C. usually employs Dreher’s defixiones criminales , also Verbrechensflüche ), and the wide range of existential situations in which people thought that an injury done to them, or threatened, justified resort to writing a curse (§3.2, pp. 210-237). C. notes here (p. 216) the frequent imbalance (to our eyes) between the harm experienced or threatened and the punishment requested. After discussion of insults and accusations, the sub-section on emotions (§3.3, pp. 238-263) returns to this issue, emphasising that the suffering envisaged is not simply gratuitous, not just due to Gewaltlust , but understood as a means of furthering ego’s interests in a critical situation and thus in a sense classified as ‘collateral damage’ (p. 260). For obvious reasons, the juridical level is heavily formulaic, but we do nevertheless come across a number of original demands for punishment (§3-4, pp. 263-286). In offering to explain such diversity, C. suggests, perhaps rather lamely, that in some cases, the principal was simply ignorant of the appropriate formulaic repertoire; whereas in others, the variation was deliberate (p. 278).

The final section (§4), critical of earlier views of the rationale of cursing, rehearses the theme of subjective justification in an existentially threatening situation. Here the choice of a sacral-religious mode of action was rendered plausible by two main factors: the institutionalisation of the written curse as an optional resort (she does not discuss a possible relation to dolls or poppets), and the limited ability of local juridical procedures to afford help to plaintiffs in particular cases, especially where concrete proof might well be lacking. The institution of the written curse thus amounts to a subjective or individualised sub-sub-system of justice ( Rechtssystem ) in Mediterranean societies in which social power was very unevenly distributed; more precisely: a weak form of leverage, given the familiar vagaries of divine response to appeals. It was above all the principals’ sense of suffering, or being threatened with, injustice that in their eyes justified the punishment requested. The variety of different situations evidenced by the surviving material suggests that individuals’ motives were unique to the particular situation in which they found themselves, so there is no point in generalising about their precise aims and state of mind on the basis of mere surmises —and, as Martin Dreyer pointed out, not only are there different senses of ‘justice’, but the principals themselves were often ambivalent in their appeals to the divinity, their language oscillating between righteous imperatives and humble entreaties 15 .

In many ways, the book represents a tour de force of intelligent problematising of a notoriously slippery set of materials, which will —or should— have an enduring influence on future work. Among its achievements one can list: the critique of the traditional listing of ‘the’ categories of curse-tablets, all too often naturalised as the unquestionably primary means of approaching their analysis; the direct attack, following the lead of Martin Dreher, on singling out the category ‘prayers for justice’ from other curse-practices, and so validating what is at best an ideal type, with all the distortions and simplifications that are entailed in such constructions; the ingenious choice of the category prayer to side-line the contentious issue of the common classification of curse-texts as magic, without heed to the trace-consequences of the history of this term in western scholarship; the heroic labour of C.’s analyses at the micro-level – there are here acute, searching, often original, readings of dozens of texts, mostly quoted in extenso and with her own translation; the convincing interleaving of theoretical positions and arguments with such micro-analyses; the deployment of individualisation-theory to ground the insistence on viewing these texts as personal tactics (rather than strategies) in concrete social situations of subjectively-perceived crisis; C.’s ability to integrate her clear self-positioning behind a politico-social approach with what is essentially a model adapted from applied linguistics.

There is, however, one serious drawback relating to the reception the book ought rightly to have: it is in German —not, admittedly, very difficult German: it is not peppered with neologisms drawn from theory— but it is rather long and the print is fairly small. For that reason I have thought it worthwhile to give a rather extended account of its aims and achievement for the benefit of those whose German is not up to reading the entire book. Otherwise it will probably mainly be consulted for its detailed analyses of individual texts, of which I provide a select list at the end of this review (see the Appendix). Another potential problem is that C., having been employed to work on the creation of the TheDeMa electronic database, of course uses it as her primary point of reference: although she regularly gives the prior or conventional print-references at the first mention of a text she discusses, the sole index of sources includes the TheDeMa numbers only, and there is no general table of correspondences. 16 The problem is compounded by the demise of TheDeMa and the serious technical difficulties encountered in setting up TheDefix at Hamburg. C. is currrently employed on another project there, and has had little time to devote to these problems. As I write, TheDefix is actually off-line, and only time will tell whether one will in future be able to access the texts directly —previously it seemed possible only to access them via individual words.

That said, much as I admire the book’s achievements, I do have some criticisms. First, the key term ‘formulae’ itself. Despite C.’s use of Wray’s applied linguistics, it does often seem that she uses ‘formula’ in a much narrower sense than Wray, so that hers can at most constitute a specialised subset of all ‘prefabricated language’ in Greek and Latin. If the key concept is prayer, we could have expected some attempt to relate curse-language to the wider forms of prayer-language in the two languages. And in developing the theme of prayer, she might have made more use of the stimulating book she has found by William Fitzgerald, on the rhetoric of prayer, which not only styles prayer as a response to crisis but also —even from a Christian perspective— firmly admits the curse among his categories 17 . All in all, as I have argued elsewhere, it might have been better, once one had established the religious character of ancient curses (and indeed, their link to curses deployed by cities and public institutions, self-curses in oaths and so on), to proceed on the basis of personal religious competence in the formulation of private curses. Whereas an approach via formulae can only cope with one, admittedly large and interesting, class of curse-texts (C. virtually ignores bare lists of names), one based on religious and performative competence would produce a theoretical continuum from the null point of the uninscribed but rolled and deposited tablet all the way to the finest, most elaborate, examples of the class 18 . In other words, I think that C.’s notion of competence, though very important, is too narrowly committed to the notion of formulae rather than to other types of relevant social knowledge.

A related problem is that of the notion of a coherent ancient ‘tradition’ of written curses, suggested by the very notion of ‘formula’. There is, however, no discussion of how such a tradition might come into being, nor of how it might persist in the absence of written models, given that, once formulated and written down, such texts were not circulated but deliberately removed from circulation, so that no individual effort could be fed back into the ‘tradition’, in the manner of a product of a recognised literary genre, that would be recited before an audience and then copied and multiplied. What then might an oral tradition of cursing have looked like? Who might have carried it? Given the virtual disappearance of vernacular curses in Greek during the late Hellenistic period, and the extremely obscure origins of the practice in the Roman world, the very idea of a coherent tradition must be subject to considerable doubt. It is precisely this sort of problem that the idea of ‘magicians’ or ritual specialists is deployed to answer, however unconvincing that solution may be in the case of ‘vernacular’ curses.

In this connection, too, I think it was mistaken not to make a clear, if indeed not a fundamental, distinction between what I used to call ‘indigenous’ and now call vernacular curses, and those known from the Graeco-Egyptian tradition 19 . There can be little question that composing the latter required knowledge and skills quite out of the reach of the principals who wrote the great majority of the texts discussed by C. Does this suggest a coherent tradition? Of course the self-same indifference has characterised the study of these texts from at least the time of Wünsch’s preface to his Appendix to IG III iii and Audollent’s collection, where both types of texts were presented indiscriminately as though they somehow all belonged to the same tradition, simply because they were all ‘curses’ 20 . Failure to work with such a distinction leads C. to introduce ‘formulae’ such as the διαβολή (pp.  219-20) , or tactics such as the combination of νικητικόν and χαριστήριον on the gold-leaf from Bostra (IV p ), which is not even a curse (p. 261-62), when these resources, like many others she does not mention, belong exclusively to the Graeco-Egyptian tradition. Nevertheless, since she hardly makes any use of texts in that tradition in her analyses, and then only in §3, it is clear that in practice she does operate with the same distinction, or one like it.

Moreover, the dual-language perspective adopted by C. inevitably tends to blur the outlines of informal, if sketchy, local knowledge of ‘how to’, and even specific oral advice at particular sites, so well explored by Tomlin and Adams in the case of the tablets from the sacred pool at Aquae Sulis/Bath, and just recently on a larger scale by Papakonstantinou in relation to Classical Athenian lawsuits 21 . In an established tradition one would also expect to find implicit rules of proper use or propriety, degrees of restraint and so on, but C. offers no help on such issues —indeed, by emphasising individualisation so heavily, she tends to give the impression that a local tradition gave no such advice about appropriate frame-conditions of this kind. It is precisely this issue that the traditional classification-system implicitly addressed, even though it is never presented from that point of view.

I also wondered about the felicity of the new substitute for ‘defixio’ (which I too would like to be rid of). Given the existence of other terms in Latin, such as execratio, imprecatio, deprecatio, and the verbal phrase dira precari, we might have expected some discussion of the most suitable term on the basis of the relevant entries in TLL22. Moreover, given the book’s argument about a subjective Rechtssystem and the protection of ego’s social figure in a crisis situation, the choice of malefica seems positively strange, because it reminds one of Pliny’s rhetorical question, defigi quidem diris deprecationibus nemo non metuit? (HN 28, 19) rather than the idea of subjectively-defined self-defence. Besides, we might well doubt that there are any emic terms that fit the bill, so why choose Latin at all? And since the entire book is about written, not oral, curses, we might surely have expected some reference to the written as opposed to the spoken word here. More generally, since Wray includes a discussion in her second book, Pushing the Boundaries, of the relation between oral and written formulae23, C. might have devoted more space to the implications of such differences for her material; one of Wray’s points, for example, is that new situations — of which crisis is a prime example— can stimulate expressive creativity and autonomy.

Finally, it seems to me that a good number of cases discussed in §2.2-4 might easily have been dealt with in §3, on individual variations. This suggests that the broadening of the concept of formula, however welcome, will inevitably lead to a weakening of its heuristic value24.

It would, however, be wrong to end on a critical note. C.’s book deserves full praise for its big thinking and grand scope, its use of important ideas from outside the narrow field of curse-studies, and its high ambition finally to go beyond the scope of catalogues, essays and notes by adopting a model framework sufficiently capacious to allow her to explore dozens of individual texts in a way that no scholar in this field has ever yet managed to do. True, this model, like Kropp’s, cannot be used to classify individual texts in the way that the old categories have been used to do, but that is not the point: hers is not a classificatory scheme but a grand effort to locate curse-tablets in a convincing socio-moral context while at the same time keeping faith with the wide diversity of expressive choices made by the principals. In a word, by reinventing the notion of ‘formula’ in the light of current work in applied linguistics, C. has developed a brilliantly imaginative line of argument that manages, despite all the odds, to impose a subtle and flexible coherence on the unwieldy mass of material that we call the corpus of Greek and Latin curse-tablets.

Bibliography

Curbera, J.

• IG II/III(3).8.1. Defixiones Atticae, Berlin, 2024.

Dreher, M.

• „Gerichtsverfahren vor den Göttern? ‚Judicial prayers‘ und die Kategorisierung der defixionum tabellae“, in G. Thür (ed.), Symposion 2009: Akten der Gesellschaft für Griechische und Hellenistische Rechtsgeschichte 21, Vienna, 2010, 303-337.

Dreher, M.

• „Rechtliche Elemente in den antiken Fluchtafeln“, in G. Thür, U. Yiftak & R. Zelnick-Abramovitz (eds.), Symposion 2017: Vorträge zur griechischen und hellenistischen Rechtsgeschichte (Tel Aviv 20-23 August 2017), Vienna, 2018, pp. 289-312.

Kropp, A.

• Magische Sprachverwendung in vulgärlateinischen Fluchtafeln (defixiones), Tübingen, 2008.

Papakonstantinou, Z.

• Cursing for Justice: Magic, Disputes, and the Law-Courts in Classical Athens, Stuttgart, 2021.

Riess, W.

• Performing Interpersonal Violence. Court, Curse and Conflicts in Fourth-Century Athens, Berlin, 2012.

Sánchez Natalías, C.

• Sylloge of defixiones from the Roman West, Oxford, 2022.

Urbanová, D.,

• Latin Curse Tablets of the Roman Empire, Innsbruck, 2018.

Wray, A.

• Formulaic Language: Pushing the Boundaries, Oxford, 2008.

Appendix

Discussions of specific texts reprinted by C. in full or in part:

DTAud
25 = Mitford 1971 no. 30 = TheDeMa 142	C. pp. 92-95; 120, n. 292; 178, n. 435
43-44 = SEG 36:351-352 = TheDeMa 139 + 1202	C. pp. 118-119
40 = IG XII 7, p.1 = TheDeMa 215	C. pp.235-237
49 = TheDeMa 140	C. pp.163-165
51 = Ziebarth 1934 no. 24 = TheDeMa 653	C. pp. 65-68
79 = Ziebarth 1934 no. 25A-B = TheDeMa 562	C. p. 182, n. 449
85 = SEG 37:389 = TheDeMa 185	C. pp. 108-110
111-112 = ILASantons 104a-b = Sánchez 2022, p. 240, no. 160 = TheDeMa 190	C. pp. 116-118
134 = Sánchez 2022, p. 132, no. 50 = TheDeMa 51 2	C. pp. 84-85
135 = Sánchez 2022, p.133, no. 51 = TheDeMa 220	C. pp.63-65
137 = Sánchez 2022, p. 93, no. 3 = TheDeMa 515	C. pp.280-281
138 = ILLRP 1145 = Sánchez 2022, p. 91, no. 1 = TheDeMa 516	C. pp. 255-258
139 = ILLRP 1144 = Sánchez 2022, p. 92, no. 2 = TheDeMa 263	C. p. 120, n. 292
140 = CIL VI 33899 = TheDeMa 529	C. pp.221-222
183 = SEG 54:524 = TheDeMa 183	C. pp. 154-155
190 = CIL X 8249 = Sánchez 2022, p. 139, no. 56 = TheDeMa 510	C. pp. 82-83
198 = IG XIV 872 = AE 2003: 337 = TheDeMa 467	C. pp. 32-33
208 = IG XIV 859 = TheDeMa 173	C. p. 277
213 = KAI 89 = TheDeMa 130	C. pp. 111-113
DTAtt
96, 97 = TheDeMa 971 + 206	C. pp. 170-171
98 = TheDeMa 223	C. pp. 291-292
IG
XII,7 p.1 (unnumbered and not in PHI) (Arkesine, Amorgos) = TheDeMa 215	C. pp. 235-237
XIV 859 = DTAud 208 = TheDeMa 173	C. p. 277
XIV 872 = DTAud 198 = TheDeMa 129	C. pp. 132-133
SEG
4: 47 (Messina) = TheDeMa 290	C. pp. 240; 290
14: 615 (Rome) = TheDeMa 219	C. pp.145-147; 237
28: 1568 (Asia Minor) = TheDeMa 230	C. pp. 159-160
30: 326 (Agora, Athens) = TheDeMa 224	C. pp. 213-216
34: 952 (Lilybaeum) = TheDeMa 182	C. pp. 155-158
35: 220 (Agora, Athens) = TheDeMa 103	C. pp. 77-79
36: 351-352 = DTAud 43-44	C. pp. 118-119
37: 389 = DTAud 85	C. pp. 108-110
37: 673 = IGDOlbia 109 = TheDeMa 232	C. pp. 221-222
41: 1581 (Bostra) = Kotansky, GMA no.58 = IGLS 9474 = TheDeMa 1189	C. pp. 88-89; 261-263
43: 434 (Pella: γάμος Διονυσοφῶντος) = TheDeMa 236	C. pp. 158-159; 248-251; 252; 253; 281-283
48: 380 = 57: 313 (Aegina) = TheDeMa 423	C. pp. 121-123; 126
49: 320 (Attica) = TheDeMa 235	C. pp. 89-91
51: 979 (Olbia Pontica) = TheDeMa 1074	C. pp.168-170; 245-247
52: 1875 = TheDeMa 379	C. pp. 271-273
57: 332 (nr. Corinth) = TheDeMa 424	C. pp. 166-168; 171; 172
57: 665 (Euxine) = TheDeMa 830	C. p. 82
57: 905 (Gela) = TheDeMa 250	C. pp. 26-29
61: 1384 (Antioch) = TheDeMa 376	C. pp. 127-128
IGDS
I no.40 (Selinunte) = TheDeMa 246	C. p. 225
Curbera, 2000
52 (Messina) = TheDeMa 280	C. pp. 243-245
PGM
O 2 = TheDeMa 115	C. pp. 269-270
SupplMag
I no. 51 = TheDeMa 189	C. pp. 125-126
II nos. 59-60 (Sabinus) = TheDeMa 311-312	C. pp. 234-235
Stroud, Corinth
127 = TheDeMa 359	C. p. 273
ZPE
188 (2014) 231-236 (Jordan/Rocca/Threatte) = TheDeMa 945	C. pp. 128-131 (unconvincing)
CIL
VI 33899 = DTAud 140	C. pp.221-222
X 8249 = DTAud 190	C. pp. 82-83
XIII 111340vii = Sánchez 2022, p. 255, no. 181 = TheDeMa 721	C. p. 278 (unconvincing)
AE
1907: 99 (Ptuj) = Sánchez 2022, p. 450, no. 527 = TheDeMa 780	C. pp. 275-276
1941: 138 (?Rome) = Sánchez 2022, p.108, no. 16 = TheDeMa 536	C. pp. 119-121
1975: 449 (Cremona) = Sánchez 2022, p. 184, no. 105 = TheDeMa 552	C. pp. 229-230
1993: 1008 = 1995: 770 = 1999: 770 (Carmo) = Sánchez 2022, p. 209, no.129 = TheDeMa 571	C. pp. 158-159
2003: 337 = DTAud 198	C. pp. 32-33
2007: 260 (Rome) = Sánchez 2022, p.127, no. 48 = TheDeMa 517	C. pp. 123; 137-139; 141
2007: 1006a,b (Groß Gerau) = Sánchez 2022, p. 404, no. 482 = TheDeMa 260	C. pp. 231-234
2010: 108 (Alcácer do Sol) = Sánchez 2022, p. 202, no. 121 = TheDeMa 600	C. pp. 180-181
TabSulis
97 = TheDeMa 654	C. pp. 267-268
98 = TheDeMa 101	C. p. 88
DTMogunt
1 = TheDeMa 758	C. pp. 265-267
5 = TheDeMa 261	C. pp. 274-275

---

* Sara Chiarini, Devotio malefica. Die antiken Verfluchungen zwischen sprachübergreifender Tradition und individueller Prägung. Hamburger Studien zu Gesellschaften und Kulturen der Vormoderne, 15. Stuttgart: F. Steiner Verlag, 2021, 339 pp. [ISBN 978-3-515-12973-2].

1  With the retirement of Martin Dreher in 2018, the Institut für Alte Geschichte at Magdeburg was closed down, but a new home for the data-bank kindly provided by Werner Riess (Hamburg). Unfortunately, due to technical problems with the software, the new site, TheDefix (www.thedefix.uni-hamburg.de), with some 1700 texts, was only launched in August 2021, and offers a restricted set of search functions by comparison with the Magdeburg version. 

2  E.g. A. López Jimeno, Textos griegos de maleficio, Madrid, 2001 (in fact Va-IVa Attic; tr. only); E. Eidinow, Oracles, Curses and Risk among the Ancient Greeks, Oxford, 2007, pp. 352-454; A. Kropp, Defixiones. Ein aktuelles Corpus lateinischer Fluchtafeln, Speyer, 2008; eadem, 2008 (sylloge in CDRom); F. Murano, Le tabellae defixionum osche, Pisa-Rome, 2013; Urbanová, 2018, pp. 426-532; Sánchez Natalías, 2022, part 2.

3  E.g. H.S. Versnel, “En het grensgebied van magie en religie, het gebed om recht”, Lampas, 19 (1986) 68-96; “Les imprécations et le droit”, RIDA, 65 (1987) 5-22; “Beyond Cursing: The Appeal to Justice in Judicial Prayers”, in C.A. Faraone & D. Obbink (eds.), Magika Hiera: Ancient Greek Magic and Religion, New York, 1991, pp. 60-106; Fluch und Gebet. Magische Manipulation versus religiöses Flehen? Religionsgeschichtliche und hermeneutische Betrachtungen über antike Fluchtafeln, Berlin, 2009; “Prayers for Justice in East and West: Recent Finds and Publications”, in R.L. Gordon & F. Marco Simón (eds.), Magical Practice in the Latin West, Leiden, 2010, pp. 275-355.

4  E.g. Nep. , Alc. 4.5; 6.5; Suet., Calig. 31; Petr., Sat. 103.6; Tac., Ann. 12.65 etc. See also Kropp, 2008, pp. 38-41.

5  O. Dufault, ‘Who wrote Greek Curse-tablets?’, in R. Evans (ed.), Prophets and Profits. Ancient Divination and its Reception, London, 2017, pp. 31-49. For a different view in the case of Classical Athens: Papakonstantinou, 2021, p. 17 n. 9 (on p. 18); 38-41.

6  It should however be noted that this is a rigorously textual study, in which the materiality of the texts, the archaeological evidence for rituals associated with the deposition of curse-tablets, and the implications of the choice of site, play virtually no role. ‘Everyone to his own last.’

7 E.G. Kagarow, Griechische Fluchtafeln, Lvov/[Lviv] – Paris, 1929; Kropp, 2008, pp. 221-229; cf. 26-67. Urbanová, 2018, while consistently treating ‘curses’ separately from ‘prayers for justice’, seems not to emphasise a connection with magic.

8 Kropp , 2008, pp. 190-240.

9 A. Wray & M.R. Perkins, “The Functions of Formulaic Language: An Integrated Model”, Language and Communication, 20 (2000) 1-28 (esp. Table 2 on p. 14); Wray, 2008, pp. 259-84.

10 Pp. 31-39 are devoted to a critique of previous analyses based on formulae, essentially Kagarow, Kropp (2008) and Urbanová (2018). The main objections are 1) being essentially empirical approaches, they muddle different semantic and grammatical levels with one another; 2) in practice they treat any textual item as a formula, so that in the end it is quite unclear what a ‘formula’ might be.

11 E.g. J. Rüpke , “Lived Ancient Religion: Questioning ‘Cults’ and ‘Polis Religion’”, Mythos , n. s.5 (2011) 191-204; “Individualization and Individuation as Concepts for Historical Research”, in Id. (ed.), The Individual in the Religions of the Ancient Mediterranean , Oxford, 2013, pp. 3-38; J. Albrecht et al., “ Religion in the Making: The Lived Ancient Religion Approach ”, Religion , 48.4 (2018) 568-93: https://doi.org/1080/0048721X.2018.1450305.

12 Dreyer , 2018, pp. 297-301; Id . „ Magie als Mittel der Konfliktlösung “, in N. Grotlkamp & A. Seelentag (eds.), Konfliktlösung in der Antike , Berlin, 2021, pp. 169-78; Riess , 2012, pp. 169-76.

13 Here C., following Riess 2012, pp. 196-97, takes issue with Versnel’s distinction (“ ‘καὶ εἴ τι λ[οιπὸν] τῶν μερ[ῶ]ν [ἔσ]ται τοῦ σώματος ὅλ[ο]υ...’. An essay on anatomical curses”, in F. Graf (ed.) Ansichten griechischer Rituale, (Stuttgart-Leipzig, 1998, pp. 217-267) between ‘anatomical’ and ‘instrumental’ corporeal curses, and his inclination to downplay the emotional intensity of such lists in order to construct a specious difference from ‘prayers for justice’ (pp. 72-77).

14 D. Aubriot-Sévin, Prière et conceptions religieuses en Grèce ancienne jusqu’à la fin du Ve siècle av. J-C., Lyon, 1992. C. also rightly notes that H.S. Versnel, “Religious Mentality in Ancient Prayer”, in Id. (ed.), Faith, Hope and Worship, Leiden, 1981, pp. 1-64 (at 17-21) had made a similar point a decade earlier.

15 Dreher, 2010: 311; 325-326. Like Dreher there, C. criticises (p. 290 n. 684) Versnel’s over-extended ‘border-area’ category, designed to save rather a lot of phenomena.

16 Sánchez Natalías 2022, pp. 497-509 does provide an inclusive Table of Correspondences (which already appeared in an Appendix to R.L. Gordon, F. Marco Simón & M. Piranomonte (eds.), Choosing Magic: Contexts, Objects, Meanings. The Archaeology of Instrumental Religion in the Latin West, Rome, 2020, pp. 169-181) for the texts in Latin, but unfortunately it does not include TheDeMa. Urbanová 2018, however, does provide the TheDeMa number in her catalogues.

17 W. Fitzgerald, Spiritual Modalities: Prayer as Rhetoric and Performance, University Park PA, 2012.

18 On the lines of e.g. R.L. Gordon, “Imaginative Force and Verbal Energy in Latin Curse-tablets”, in C. Sánchez Natalías (ed.), Litterae Magicae: Studies in Honour of Roger S.O. Tomlin, Zaragoza, 2018, pp. 145-63 (at 148-150). However, we might well doubt the capacity of such an approach to cope with the mass of material that C.’s model can encompass.

19 E.g. R.L. Gordon (nominally with C.A. Faraone), “Introduction” to the issue on Curses in Context, 1: Curse-Tablets in Italy and the Western Roman Empire, Religion in the Roman Empire, 5.3 (2019 ) 319-339 (at 320-321); Id. “Curse-practices in the Late-Antique Roman Levant and North Africa”, Religion in the Roman Empire, 7.1 (2021) 3-18 (at 3-5); Id. “Introduction” to Id., F. Marco Simón & M. Piranomonte (eds.), Choosing Magic: Contexts, Objects, Meanings. The Archaeology of Instrumental Magic in the Latin West, Rome, 2021, pp. 15-22 (at 16-18). I owe the term vernacular to M. Bowman & U. Valk (eds.), Vernacular Religion in Everyday Life: Expressions of Belief, Abingdon, 2014.

20 R. Wünsch, Defixionum Tabellae = IG III iii, Appendix, Berlin, 1897; A. Audollent, Defixionum Tabellae ..., Paris, 1904 [repr. Frankfurt a.M., 1967]. We find the same practice casually employed in all subsequent general treatments of the practice, such as the generally fine book by M. Bailliot, Magie et sortilèges dans l’Antiquité romaine. Archéologie des rituels et des images, Paris, 2010.

21 Bath: R.S.O. Tomlin, “The curse-tablets”, in B. Cunliffe (ed.), The Temple of Sulis Minerva at Bath, 2: The Finds from the Sacred Spring, Oxford, 1988, pp. 59-277; J.N. Adams, “British Latin: The Text, Interpretation and Language of the Bath Curse Tablets”, Britannia, 23 (1992) 1-26; note also his excellent commentaries on some of the British texts in his Anthology of Informal Latin, 200 BC- AD 900, Cambridge, 2016; and M.C. Line, “Cursing in Roman Britain: Connectivity, Identity and Belief”, Ph.D. Durham UK (2019). Athens: Papakonstantinou 2021.

22 Word has perhaps now spread that the nominal defixio, spelled defictsio, has recently turned up in a poorly-written ‘thief-text’ of the second or third century from the Roman fort of Abusina (Raetia) near Eining (now part of Neustadt a.d. Donau, Kr. Kelheim) (DTAbusina 1): J. Blänsdorf, “Pathetic Lament on a defixion Tablet from Abusina (near Eining, Danube)”, in G. Rocca & G. Bevilacqua (eds.), Gift of a Book: Studi in memoria di David Jordan, Alessandria, 2020, pp. 97-102. The phonetic spelling implies that the principal had indeed heard the word used but never seen it written.  Hitherto, apart from an entry in a mediaeval Latin-Greek glossary, only the verbal form had been encountered in such texts.

23 Wray, 2008, pp. 37-60.

24 The book is well-produced, with virtually no misprints (but note ‘Hyppolitos’ [p. 215, x4], correct in the Index locorum; ‘rivarly’ [p. 259n. 619]). I would however mention that the word etisch is wrongly spelled ethisch at pp. 186, 187, 235 with n. 564; 251; 254 with n. 608; 289 n. 681, but spelled correctly on pp. 261; 291; 294; 295; 299. Since C. employs both terms in her argument, the confusion is no doubt due to a misunderstanding by the spelling-checker. It is also odd that C. still claims that the group from Amathous in Cyprus was found at Kourion (pp. 178nn. 433, 435, 437; 185n. 460; 231), no doubt because she used T.B. Mitford, The Inscriptions of Kourion, 1971 and missed the subsequent correction; nor does she cite the extremely critical review of Mitford by T. Drew-Bear, BASP 9 (1972) 85-107. The following works are missing from the bibliography: B. Brooten, Love between Women, 1996; W. Fitzgerald, Spiritual Modalities, 2012; M. Gluckman, Essays on the Ritual of Social Relations, 1967; Johnson 2004 [I cannot trace this]; C. Kotsifou, “Prayers and Petitions for Justice”, Tyche, 31 (2016) 167-99; C. Menke, Kritik der Rechte, Berlin, 2015; E. Rauch, Sprachrituale, 1992; G. Rocca, “Les defixiones siciliennes ...”, in C. Dupraz & W. Sowa (eds.), Genres épigraphiques et langues d’attestation ... , 2015, pp. 305-313. IGDS (Dubois) is missing from the list of acronyms. Naturally C. could not refer to Curbera 2024.