Meredith Martin, Poetry’s Data: Digital Humanities and the History of Prosody (Princeton: Princeton University Press, 2025) 224 pp. £84 Hb. £30 Pb. ISBN: 9780691254678
Meredith Martin’s Poetry’s Data is a large-scale attempt to bring together literary studies, linguistics, digital humanities, and critical archives studies. It presents a compelling case on why we should critically examine in the digital age the way in which we read, quantify, interpret, and historicise poetry. Martin argues that her ‘aim is to show scholars of literary studies the urgency of theorising the practices that the adjacent disciplines of critical archival and critical data studies have understood as vital to our enterprise for quite some time’ (p. 3). The scope of Poetry’s Data is Martin’s resident collection, the Princeton Prosody Archive (PPA), which covers digital ‘materials related to the study of prosody, both versification and pronunciation, in English between 1532 and 1928, the PPA has at the time of this writing about seven thousand items’ (p. 7). It also historicizes digital humanities (DH) and archival collections that are collaborative, interdisciplinary, evolving, and financially precarious; today, more than ever. For a scholar of English literary studies, Martin’s book raises many questions rather than answers them: it made me reevaluate the relationship between text and interface and its influence on the way we read poems and the infrastructures and databases that hold poetry (and what poetry is). The chapters in this book are designed with the language of quantitative methodology in mind, and each include an ‘Exhibit’ as a case study.
Chapter 1, ‘How We Count [Literary]’, reviews how physical and digital archives and library infrastructures have contributed to the understanding of poetry as a literary subject, and disregarded ‘the history of versification, nor did it seem to acknowledge the historical conditions of poetry’s materiality––its formats, its circulation, its sounds––either written or spoken’ (p. 29). In this chapter, Martin accounts for the three decades of undertaking archival study and its relation to form, medium, and infrastructure. She names the first stage the ‘pre-Google Books’ age between 1999-2006, when the establishment of digital archives and infrastructures were built by individual textual scholars who ‘relied on their own authority as scholars in the field but did not necessarily involve the expertise of scholars in archival science, information management, or information retrieval’ (p. 35). Martin discusses the challenges of the transitional phase between 2007-2011, during which she grappled with searching in digitised materials and environments and ‘how information about a work, the page image of a work, and its underlying text are entirely different materials with different provenances and different histories’ (p. 49).
Chapter 2, ‘How We Read [Word Lists and Dictionaries]’, outlines how information structures concerning versification and pronunciation (both linked to the understanding of what poetry is) have changed the way we search and research. Martin is interested in the way in which ‘“Word Lists” and “Dictionaries” have been left out of the history of how we think about versification and poetry’ (p.62), explaining that these digitised books are often automatically left out from search interfaces, like PPA. Considering the context, particularly in digital searching, helps to turn us from searchers to researchers. For example, Martin notes that ‘prosody’ is not prosody in prosody-related books, but we need to search ‘prosody’ within ‘poetry’. She associates this decontextualised reading with I. A. Richards’s New Criticism: ‘the interface has rendered invisible the impracticalities, the need for theory or history’ (p.75). The chapter warns against turning everything into digital or digitised content without considering the historical context and conditions in which they have been conceived.
In Chapter 3, ‘How We Classify [Linguistic]’, Martin explores ‘poetry’s metadata’ (p. 80) and argues that the words that define data function as interpretative categories for, for example, poetry, that ultimately exclude other types of information. This chapter further employs PPA, demonstrating that classification (metadata) reveals that ‘scholars who mostly fall outside the boundaries of literary history [such as music, pedagogy and linguistics] can give new ways of thinking about the contours of that history’ (p. 87). Poetry’s data relies on information from other disciplines and histories that might consider data differently, and scholars need to inquire first about ‘[h]ow can we best classify prosodic discourse in texts for future work?’ (p. 92) to realise how might certain poems were read outside of the discourse of poetry and/or left out of the databases and records.
In Chapter 4, ‘How We Express [Typographically Unique]’, Martin looks closely at digitised materials in the Princeton Prosody Archive that ‘Google cannot read to think about what computers can and cannot know about poetry’ (p. 97), meaning they contain symbols and typographies beyond words. These materials are mostly related to scansion, elocution, and prosody that have pedagogical purposes, such as John Thelwall’s Illustration of English Rhythmus (1812). The chapter also discusses the theoretical challenges of expression in poetry, ‘whether the most literal sense of expression in performance, its metaphorical use in criticism (consolidated under the fiction of the ‘speaker’), the expressive possibilities of apprehended patterns, the computational operations of regular expressions’ (p. 111), and the distant reading (data reading?) in the age of digital interfaces. While there is a necessary reflection on generative AI, Martin also argues that the computational methods have their limitations since ‘[c]omputers can’t scan (yet) because the matter of meter and rhythm means entirely different things to different communities’ (p.128).
The last chapter, Chapter 5, ‘How We Argue [Original Bibliography]’, focuses on the conception, mechanism, and maintenance of T. V. F. Brogan’s 1981 publication of English Versification, 1570-1980: A Reference Guide with a Global Appendix (EVRG), which then became the foundation for The Princeton Encyclopedia of Poetry and Poetics (1993). The chapter reflects on the reimagination of EVRG in the digital age, in which the resource guide continues to evolve and is defined by hypertexts. Martin notes that these reflections ‘critique the largely invisible knowledge structures that uphold modern literary criticism, and the institutional infrastructures on which they are built, by proposing new ways of navigating scholarly materials and imagining different models for scholarship’ (p.135). This chapter not only analyses and reflects on the development of Brogan’s system, but Martin also shares more personal stories about career development and collaborative work, and interrogates how academic precarity and the funding models contribute to the challenges of maintaining digital resources and databases.
Poetry’s Data is a reminder for all that archival resources, online systems, databases, and even digital reading infrastructures constitute research, and they have a long and complex history. These systems are not ideologically neutral either. Martin argues resources like Brogan’s EVRG or PPA do ‘not count and [are] not legible to or properly valued by the very people who might use the PPA, people who assume that it is a service and not scholarship’ (p. 167). The book offers a wide-scale analysis of the Princeton Prosody Archives. It offers digital humanists a critical investigation into the conceptual understanding of poetry and poetry’s (meta)data, which are, perhaps today even more than ever before, shaped by systems and infrastructures.
Dorka Tamás, University of Groningen
