Arts of Memory, Ancient Manuscript Technologies, and the Aims of Theology

Round 1

Reviewer 1 Report

Review: “Arts of Memory, Ancient Manuscript Technologies, and the Aims of Theology”

This is a strong essay. The argument is lucid, significant, and (with a few minor caveats) persuasively presented.  Several revisions would strengthen the article and I recommend that these are taken into account before submission of the final manuscript.

First, the engagement with the key scholarship on textual organisation in the Roman Mediterranean, on early Christian paratexts, and on late ancient textuality should be enriched. I suspect that the author has attempted to streamline their engagement with secondary scholarship in the interests of brevity. Nonetheless, important recent contributions have been omitted. In particular, note the following:

• The author cites Duncan’s delightful 2021 book on p. 2. This is well and good, and Duncan’s book is on the whole excellent. But Duncan is not a particularly reliable guide in his discussion of pre-modern phenomena. Moreover, the author has omitted the most important recent discussion of pre-modern information management, the 2019 OUP monograph of Andrew Riggsby, Mosaics of Knowledge: Representing Information in the Roman World (see esp. chs. 1 and 2). Similarly missing here, at least from my perspective, is the foundational work of Blair, A. 2010. Too Much to Know: Managing Scholarly Information Before the Modern Age. New Haven: Yale University Press. Moreover, when it comes to the vertical file one should cite Robertson, C. 2021. The Filing Cabinet: A Vertical History. Minneapolis: University of Minnesota Press.
• On p. 2, the author would do well to cite Andrist’s crucial article on the definition of the paratext in manuscript cultures: Andrist, P. 2018. “Toward a Definition of Paratexts and Paratextuality: The Case of Ancient Greek Manuscripts.” In Bible as Notepad: Tracing Annotations and Annotation Practices in Late Antique and Medieval Biblical Manuscripts, edited by L. I. Lied, and M. Maniaci, 130–50. Berlin: de Gruyter. Moreover, on early Christian manuscripts and their paratextual features (with particular relevance to the notion of ‘nonlinear’ access), note Coogan, J. 2021. “Transforming Textuality: Porphyry, Eusebius, and Late Ancient Tables of Contents.” Studies in Late Antiquity 5 (1): 6–27.
• p. 2 n. 5 should include Coogan, J. 2017. “Mapping the Fourfold Gospel: Textual Geography in the Eusebian Apparatus.” Journal of Early Christian Studies 25 (3): 337–57.
• The discussion of Pliny on p. 3 should engage Riggsby (both his 2019 book and his earlier 2007 article in the Whitmarsh / König volume) and also the important monograph on Pliny and the ToC: Doody, A. 2010. Pliny’s Encyclopedia: The Reception of the Natural History. Cambridge: Cambridge University Press. (esp. ch. 3).
• p. 4 n. 8: On the ‘gains and losses’ of codex technology, there has been a substantial body of recent scholarship that has refined and at points critiqued the Cavallo cited here. Note especially:
• Harnett, B. 2017. “The Diffusion of the Codex.” Classical Antiquity 36 (2): 183–235.
• Larsen, M. D. C., and M. D. Letteney. 2019. “Christians and the Codex: Generic Materiality and Early Gospel Traditions.” Journal of Early Christian Studies 27 (3): 383–415.
• Coogan, J. 2021. “Gospel as Recipe Book: Nonlinear Reading and Practical Texts in Late Antiquity.” Early Christianity 12 (1): 40–60.
• p. 15: A minor bibliographic matter: Why is the edited König and Whitmarsh 2007 volume cited in the bibliography to the expense of the actual König and Whitmarsh essay from which the citations are derived?

Second, several questions arise about the relationship between the mnemotechnical model advanced here and the broader contexts of late ancient grammatical education and textual scholarship. This is the part of the essay that is least well developed and it warrants further attention before publication.

• p. 2: Do mind and written surfaces afford different modes of knowledge organisation because of differences between their media? This article proposes (rightly to my mind) the significance of mnemonic organisational strategies for the method adopted by Priscillian, but it does not explore the potential differences afforded by the written medium that Priscillian deploys. To transfer from mind to text is never automatic or one-to-one, but that fact does not emerge into visibility here.
• Nor does this article discuss Priscillian’s own intellectual formation, which matters in the present context: Education in the fourth century was not the same as education in the fourth century BCE (Aristotle) or the first century CE (Quintilian).
• p. 3 discusses ‘the larger intellectual implications’ of paratextual phenomena. Here and at other points in the article, the author’s approach reflects the influence (direct or indirect) of Marshall McLuhan’s The Medium is the Massage, as well as Jack Goody’s 1977 The Domestication of the Savage Mind. Both of these earlier works—and the article under review—reflect a stronger view of technological determinism than this reviewer finds compelling, but in light of the idea advanced here that processes of abstraction create new possibilities of knowledge, it is vital to engage the key earlier scholarship that has advanced these claims. The claims in this paragraph on p. 3 (the last before the section break) are in some tension with the arguments advanced on p. 2 (and throughout) that present Priscillian’s approach simply as the written form of existing mnemonic technologies, rather than a textualised transformation of them.
• p. 5: To continue the previous point, the author discusses how paratextual systems ‘translate’ technologies from mind to manuscript. But translation implies difference as well as similarity. I found myself repeatedly wishing that the author would address the other side of the coin: In what ways do Priscillian and others adapt or change existing technologies in order to work with the affordances of the manuscript? Most of the other suggestions in my revisions are to some extent optional. This one is vital.
• Note that (e.g. on p. 3 n. 7), non-linear access is not as such ‘random’. The varied information technologies discussed here (and in the other scholarship referenced) are about making non-linear access possible in a ‘mapped’ (Coogan 2017) and ‘ordered’ (König and Whitmarsh 2007) way. The language of ‘random’ (here probably borrowed indirectly from modern informatics) is misleading. There are, of course, truly random modes of non-linear access in antiquity: various sorts of sortilege would qualify.
• The discussion (esp. on p. 5) of the loci method and alphabetic-numerical organisation omits discussion of alphabetisation as a textual and mnemonic strategy, yet this is fundamental to the cognitive practices envisioned here. It would strengthen the author’s argument to note how the mnemonic practices here discussed are themselves already textualised. This disrupts the tidy teleology of memory-to-writing that undergirds much of the scholarship on both ancient memory and ancient textuality. On alphabetisation, see especially Riggsby’s 2019 monograph (which draws upon Daly, L. W. 1967. Contributions to a History of Alphabetization in Antiquity and the Middle Ages. Bruxelles: Latomus). For explicit discussion of alphabetisation and early Christian paratexts, see Coogan, J. 2021. “Gospel as Recipe Book: Nonlinear Reading and Practical Texts in Late Antiquity.” Early Christianity 12 (1): 40–60.
• p. 5: Many modern readers find it much harder to move backward in the alphabet than to move forward. The spooling forward-and-back in the loci method here discussed is itself — at least to some extent — dependent on a substructure in ancient Mediterranean education in which letters and syllables were manipulated in both directions. On this aspect of ancient education, see Morgan, T. 1998. Literate Education in the Hellenistic and Roman Worlds. Cambridge: Cambridge University Press. In short (and a point that emerges again vis-a-vis the third paragraph on p. 6): forward and back as a cognitive practice is cultivated.
• The dynamics of division and recombination that the author discusses on p. 6 are addressed in the context of early Christian textuality (and grammatical education particularly) by Chin, C. M. 2008. Grammar and Christianity in the Late Roman World. Philadelphia: University of Pennsylvania Press. Note further Chin’s discussion of Priscillian in: Chin, C. M. 2015. “Cosmos.” In Late Ancient Knowing: Explorations in Intellectual History, edited by C. M. Chin, and M. Vidas, 99–116. Berkeley: University of California Press.

Third, the reading of Priscilian might be located in the context of late ancient biblical scholarship and might be refined to make the argument more persuasive:

• In light of the argument advanced on the top half of p. 4, the author should engage Gérard Genette more extensively. In particular, Genette’s distinction between ‘mute’ and ‘voiced’ textual divisions invites exploration. Eusebius’ system of Gospel cross-references was not organised under thematic headings like the textual content of Priscillian’s Canons, and exploring this contrast (as both a technological and theological decision) might illuminate the distinctive uses of Priscillian’s project.
• The example using Isa 53:7 on p. 4 works against the author’s argument. After all, such granular reference systems for the Greek Bible were not in common use in the period under discussion. Note, however, the recent work of Charles Hill: Hill, C. E. 2022. The First Chapters: Dividing the Text of Scripture in Codex Vaticanus and Its Predecessors. Oxford: Oxford University Press.
• p. 6: The author asserts that Priscillian’s system was ‘the most sophisticated and powerful’ such system in the ancient world. This reviewer finds the claim hyperbolic in comparison with the Gospel cross-reference system devised by Eusebius (on which Priscillian indubitably depends, as his choice of the name ‘canon’ for his project — itself not a ‘canon’ in any prior sense of that word — and the language of his preface both attest). Eusebius’ system of mute intertitles is both substantially more intricate (organising many more kinds of textual information), more extensive, and more granular in segmentation and cross-reference. I don’t expect I will change the author’s mind on whether Eusebius’ system or Priscillian’s is more powerful and sophisticated. Rather, I ask, what’s the value of making the contestable claim here at all? 
• Likewise on p. 7, the discussion fails to fully acknowledge the influence of the Eusebian apparatus. Also missing is acknowledgment of the reasonably parallel and roughly contemporaneous Euthalian system.
• p. 6 n. 14: The author rightly notes that the use of architectural (and more broadly, spatial) imagery as a metaphor and as a mental device for structuring the mind is standard. What the author does not observe is that this logic is explicitly adopted in the conceptualisation and mise-en-page of numerous paratexts in late antiquity (including a number before Priscillian). Note the rich recent scholarship, especially:

• • Coogan, J. 2017. “Mapping the Fourfold Gospel: Textual Geography in the Eusebian Apparatus.” Journal of Early Christian Studies 25 (3): 337–57.
  • Strøm-Olsen, R. 2018. “The Propylaic Function of the Eusebian Canon Tables in Late Antiquity.” Journal of Early Christian Studies 26 (3): 403–31.
  • Elsner, J. 2020. “Beyond Eusebius: Prefatory Images and the Early Book.” In Canones: The Art of Harmony: The Canon Tables of the Four Gospels, edited by A. Bausi, B. Reudenbach, and H. Wimmer, 99–132. Berlin: de Gruyter.
  • Crawford, M. R. forthcoming. “Reconsidering the Tholos Image in the Eusebian Canon Tables: Symbols, Space, and Books in the Late Antique Christian Imagination.” In The Intellectual World of Late Antiquity, edited by L. Ayres, M. R. Crawford, and M. W. Champion, forthcoming. Cambridge: Cambridge University Press.
• p. 7: Priscillian’s execution was for magic and not simply for heresy (see the 2015 Chin essay cited above). This matters in the present context insofar as bookish ritual or ‘magical’ practices in late antiquity often involved manipulation and reconfiguration of written signs. One should not separate the kinds of textual manipulation and organisation involved in Priscillian’s canons from the accusations that led to his death.
• p. 8: The discussion of the term ‘testimonia’ here warrants another sentence or two. The footnote is relevant, but is insufficient. The statement in the main text that ‘testimonia’ is equivalent to ‘verses’ is an oversimplification; the semantic content is quite different even if the function is analogous. There’s a hermeneutic posture encoded into this language.
• p. 8: The comparison in granularity between Priscillian’s system and the Stephanus system is incomplete without comparison to the Euthalian divisions.
• p. 8: A contrast between the sense units of Priscillian’s sections and the cross-referential segmentation logic of the Eusebian apparatus would be worthwhile, at least in a footnote. See Coogan 2017 in addition to Crawford 2019 (full references above).
• p. 8: To call Priscillian’s system a ‘grid’ is overstated. It is not a tabular system in the same technologically powerful sense that Eusebius’ system is. To juxtapose this statement the subsequent checksum table is likely to mislead a reader who doesn’t already have Priscillian’s system in front of them.
• p. 9: On the use of contrasting colours of ink, see the Eusebian and Euthalian systems. This is not an innovation of Priscillian’s but a shared late ancient strategy for information organisation through colour-coding. Recent scholarship on the Eusebian and Euthalian systems has discussed and contextualised this practice.
• p. 9ff: The central contribution of the article is to show the possibilities for creative reading afforded by Priscillian’s system. Here the author might compare the examples and the terminology of ‘creative juxtaposition’ in Coogan, J. 2021. “Transforming Textuality: Porphyry, Eusebius, and Late Ancient Tables of Contents.” Studies in Late Antiquity 5 (1): 6–27.
• p. 10: It would be helpful to clarify (either for the example of canon 79 or for Priscillian’s system broadly): What proportion of the words in the canon titles are derived from the wording of Pauline texts included under those headings?
• p. 11: Why does the author cite the NRSV but supply a parenthetical Latin text? It would be preferable to provide a translation actually derived from a late ancient Latin text.
• p. 12 n. 31: Does the different approach to the division of Hebrews suggest that Priscillian had a different hermeneutical posture toward Hebrews? Or does it suggest the possibility that Hebrews was incorporated into the system subsequently (by either Priscillian or a different reviser of the system)? Or, yet again, is this a reflection of the difficulties that arise in reading Hebrews as a Pauline? This is, perhaps, a question larger than can be answered in the present essay, but the possibilities are tantalising and they might be acknowledged here.
• p. 13: This reviewer agrees that ‘Priscillian anticipates the technologies of modern biblical citation for constructive theological purposes’. Yet Priscillian is participating in a culture of late ancient biblical scholarship in which several different projects deploy similar strategies using numerically-structured paratexts. In particular (as argued in Coogan 2017 and followed in Crawford 2019), the Eusebian apparatus similarly juxtaposes Gospel material in order to invite theologically laden reading.
• pp. 12–13: To what extent are the particular theological and textual connections that Priscillian offers in canon 79 anticipated or paralleled by other late ancient readers? To what extent is the reading that the author offers here one that is already developing in other modes of late ancient exegesis (whether connected to Priscillian or not) and to what extent is Priscillian’s paratextual project offering new theological possibilities for reading the Pauline corpus?
• p. 14: The author indulges in hyperbole. With respect to what set of criteria is Priscillian’s project ‘best’, and why are those relevant here? After all, Priscillian’s theologically rich and technologically sophisticated project did not secure for itself a particularly robust reception. Over the course of the Latin Middle Ages, other tools shaped the reading of Paul much more. What’s the basis for this claim and what are the stakes of its truth?

Note a number of minor errata:

• p. 4: ‘earliest ancient’ is redundant; ‘earliest’ would suffice.
• p. 4 n. 10 should read ‘Aristotle’. I also suggest omitting ‘To be clear’.
• p. 9: I assume that ‘I, II, II’ should be a rising sequence of ‘I, II, III’.
• p. 11: There is an unnecessary tab space in the middle of a paragraph. Should this be a simple space or a paragraph break?
• p. 13: For ‘newly order relations’ correct to ‘newly ordered relations’.

Author Response

Excellent feedback. Many thanks. Much has been addressed. 

Reviewer 2 Report

The article is both formally and materially of great interest for biblical and theological study in relation to the Greco-Roman context.

Author Response

Many thanks.