Andrew Piper

Rethinking the Print Object: Goethe and the Book of Everything

This essay combines a consideration of the two-decades-long publishing strategy of Goethe’s last major prose work, Wilhelm Meisters Wanderjahre (1808–29), with a reading of specific formal features...

Vanishing Points: The Heterotopia of the Romantic Book

This article aims to establish a set of interdisciplinary practices to identify the uniqueness of Romantic books in particular and bibliographic change more generally. It argues that when we take into account the book at a synaesthetic level, we see a new relationship to the book emerging during the Romantic period that converges around the production of “vanishing points.” Reading is no longer framed as something utopian, as in a medieval or early-modern tradition (from Dante to More), but instead as heterotopian, as the production of some outside within. Heterotopia, I argue, becomes the condition through which new kinds of knowledge formation as well new kinds of mediation become possible, up to and including the non-book. The Romantic book marks the beginning of the theorization of the books end. Things liberate one from surveillance and … observation frees one from suspicion. Sciences that are not acquainted with objects can only rely on sleuthing and policing. (Michel Serres, The Five Senses 42)

Feature Learning for Footnote-Based Document Image Classification.

Classifying document images is a challenging problem that is confronted by many obstacles; specifically, the pivotal need of hand-designed features and the scarcity of labeled data. In this paper, a new approach for classifying document images, based on the availability of footnotes in them, is presented. Our proposed approach depends mainly on a Deep Belief Network (DBN) that consists of two phases, unsupervised pre-training and supervised fine-tuning. The main advantage of using this approach is its capability to automatically engineer the best features to be extracted from a raw document image for the sake of generating an efficient representation of it. This feature learning approach takes advantage of the vast amount of available unlabeled data and employs it with the limited number of labeled data. The obtained results show that the proposed approach provides an effective document image classification framework with a highly reliable performance.

Footnote-Based Document Image Classification

Analyzing historical document images is considered a challenging task due to the complex and unusual structures of these images. It is even more challenging to automatically find the footnotes in them. In fact, detecting footnotes is one of the essential elements for scholars to analyze and answer key questions in the historical documents. In this work, we present a new framework for footnote detection in historical documents. To this aim, we used the most salient feature of the footnotes, which is their smaller font size compared to the rest of the page content. We proposed three types of features to track the font size changes and fed them to two classifiers: SVM and AdaBoost. The framework shows promising results over 80% for both classifiers using our dataset.

The Werther Effect: An Interview with Andrew Piper and Mark Algee-Hewitt

Andrew Piper (AP): It began as a collaboration. Mark was at McGill as a postdoctoral student. He had been thinking for some time about quantitative and computational approaches to literary study. The methods he could bring to bear, and the questions he was able to ask, aligned with things that I was interested in learning. So we started a pilot project that focused on exploring the legacy of Werther and, specifically, the ways in which influence fans out culturally. There are a number of commonplaces about Werther in cultural history: that it was the first European bestseller, that it had a major impact on European culture. We felt this made it a good subject for a study of textual influence—and there was an opportunity to reexamine those commonplaces. We started with Goethe’s own corpus. His personal trajectory is also one of those scholarly commonplaces: he writes a sensational novel, he’s well known for it, and then, as an author, he has disdain for it and wants to disown it. Many authors have gone through the same arc: they wish to move on stylistically, and in order to do so they feel they have to renegotiate their public identity. So our first major step was to explore Werther’s implication in the rest of Goethe’s writing. We looked at different kinds of vectoral representations of the novel—ways of modelling the novel as a semantic distribution, then how that distribution brings other clusters of texts together. Throughout, the research was probing an intuitive literary question: to what extent is there a latent Werther-ity within Goethe’s other writing? Ultimately, when we looked at the topologies that Mark developed, we saw a push against the clean periodisation that’s typical of descriptions of Goethe’s writing: not a clear division between early, middle, and late, but other kinds of groupings.

The More Antecedents, the Merrier: Resolving Multi-Antecedent Anaphors

Anaphor resolution is an important task in NLP with many applications. Despite much research effort, it remains an open problem. The difficulty of the problem varies substantially across different sub-problems. One sub-problem, in particular, has been largely untouched by prior work despite occurring frequently throughout corpora: the anaphor that has multiple antecedents, which here we call multi-antecedent anaphors or manaphors. Current coreference resolvers restrict anaphors to at most a single antecedent. As we show in this paper, relaxing this constraint poses serious problems in coreference chain-building, where each chain is intended to refer to a single entity. This work provides a formalization of the new task with preliminary insights into multi-antecedent noun-phrase anaphors, and offers a method for resolving such cases that outperforms a number of baseline methods by a significant margin. Our system uses local agglomerative clustering on candidate antecedents and an existing coreference system to score clusters to determine which cluster of mentions is antecedent for a given anaphor. When we augment an existing coreference system with our proposed method, we observe a substantial increase in performance (0.6 absolute CoNLL F1) on an annotated

Mr. Bennet, his coachman, and the Archbishop walk into a bar but only one of them gets recognized: On The Difficulty of Detecting Characters in Literary Texts

Characters are fundamental to literary analysis. Current approaches are heavily reliant on NER to identify characters, causing many to be overlooked. We propose a novel technique for character detection, achieving significant improvements over state of the art on multiple datasets.

Of Automatics: Sophie von La Roche and the Life of the Writer's Desk

This essay explores the desk as an important figure through which to understand the genre of life-writing and the life of writing at the close of the eighteenth century. Through a discussion of Sophie von La Roches My Writing Desk (1799), the desk and its biography emerge as key instruments for working out the automaticity and the machinality of writing—the way writing and its instruments could produce more writing. In so doing, the biography of the desk rewrites one of the founding tropes of autobiographical narrative—that of conversion. Instead of being understood in the sense of a personal crisis, conversion assumes a new meaning in the nineteenth century as a means of exploring the discursive and technological transformations that allow for writings movement between different types of material and epistemic spaces.

The Art of Sharing: Reading in the Romantic Miscellany

‘Sharing is more difficult than you think.’1 This was the advice offered to the Major by his friend in Goethe’s novella, ‘The Man of Fifty’, and it concerned the difficulties of transmitting the Verjngungskunst, the art of rejuvenation, that the Major required in order to remain vital and youthful for his niece who, in a typical Goethean fantasy, had fallen in love with him. ‘The Man of Fifty’ had initially appeared in part in 1817 in Cotta’sLadies’ Pocket-Book, and it was a story that was in fact largely concerned with the problem of the part – with the parting, imparting, and parting with things. It would later be included in Goethe’s last novel,Wilhelm Meister’s Travels (1821/29), where it would achieve its fame as one of his most important prose works, and yet its initial placement within Cotta’s miscellany disclosed an important fact about the culture of nineteenth-century miscellanies in which it first appeared: that the question of the part, imparting and parting with — in a word, sharing — was integral to the miscellanies’ success as a literary format in the nineteenth century.

The Making of Transnational Textual Communities: German Women Translators, 1800-1850

By portraying the work of female translators from the first half of the nineteenth century, this essay explores how women created new textual communities that capitalized on the increasing diffusion of print networks. In doing so, women used translation to negotiate new relationships to print and publishing that facilitated their emergence as a professional writing class. Translations by women also promoted the creation of new, increasingly international cultural geographies in which translation functioned not as a force for homogenization, but as a means of identifying cultural differences. (AP)