Digital Analysis Of The Literary Reception Of J.W. V. Goethe’s Die Leiden Des Jungen Werthers

paper, specified "short paper"
  1. 1. Sandra Murr

    Universität Stuttgart

  2. 2. Florian Barth

    Universität Stuttgart

Work text
This plain text was ingested for the purpose of full-text search, not to preserve original formatting or readability. For the most complete copy, refer to the original conference program.

The impact of Johann Wolfgang von Goethe's epistolary novel Die Leiden des jungen Werthers marks a singular moment in German literary history, as it sparked a remarkable critical and productive literary reception. After the publication in 1774, so-called Wertheriaden — literary adaptations that are formally, thematically and structurally guided by Goethe’s novel — appeared in various languages. These texts form our corpus for the Werther Project, which is part of the interdisciplinary DH Project CRETA of the University of Stuttgart.

Literary studies have mainly concentrated on individual texts of this literary reception and on the individual genres to which the adaptations belong. Our research approaches focus on a macro-perspective to recognize recurring structures, elements, motifs and character constellations in all of the Wertheriaden with the help of computer-assisted analysis. From this perspective, we want to determine how closely the literary adaptations are based on Goethe’s original text Werther.


Our Werther corpus contains about 150 German and 30 English texts from different genres and literary periods — mainly (epistolary) novels, dramas and poems. The rationale for our corpus selection is based on Werther's publication in 1774 and its translation into French (1776) and English (1779). In the last two decades of the 18th century no other work of German literature was ever been translated or adapted in English as often as Goethe’s novel.

To establish comparative parameters for such a heterogeneous corpus (which also allow an analysis across all literary genres), we concentrate on specific features which are present in Goethe's original work. Among these discernible and typical characteristics (Martens 1985) are the triangular relationship of the three main characters, the concept of the “Werther character” (emotional, artistic), the monoperspectival narration, the role of nature (with motifs such as "Herbst", i.e. autumn), the subject’s relation to society, the so-called "sickness unto death" (“Krankheit zum Tode”) with the protagonist’s subsequent suicide, as well as structural, stylistic and linguistic similarities (Horré 1997). These reference points will be applied leading to a large scale comparison.

One approach is to compare the different texts on the basis of network and lexical information. In this context, we are concentrating on the visualization of the typical character network of Goethe’s original text and whether this triangular constellation reoccurs in the Wertheriaden.

Character Networks
The analysis of character constellations is an essential part of literary studies. It shows the dynamic structure of the interactions of the central characters in the text and characterizes them in relation and contrast to each other (Pfister 1982). In the last few years, there has been a growing interest in graph-theoretic visualization and analysis of social networks in literary texts — especially in dramatic texts (Hettinger et al., 2015, Trilcke 2013, Moretti 2011). Central character constellations occur in all literary genres and form the basis for conflict constellations, particularly for the static counter-narrative between a protagonist and an antagonist. In Goethe’s Werther, this typical constellation is extended by another figure — Lotte. She mediates between the emotional protagonist Werther, who is immortally in love with her, and Albert, her fiancée and Werther’s adversary with opposite characteristics. A central aspect of the Werther Project is the identification of this triad by visualizing the character constellation in the original text and the literary adaptations, as well as detection of deviations.

For the determination of character relations, name lists were derived from Goethe’s epistolary novel and its adaptations in close reading. This lexicon-based approach leads to better results than machine learning techniques for named-entity recognition, since complex entities ("Graf C." or "Frau von S.") which exist in the original text have to be determined without doubt. For each character we paid attention to synonyms (e.g. Lotte, Lottgen, Lottchen) and in case of the Werther character we added the personal pronoun “I”, due to the first-person narration. Connections are established when two named entities are located at an adjustable distance of n tokens (normalization, stop word removal or sentence borders can be set as parameters). Based on this simple heuristic approach, the typical triad can be identified in both Goethe's Werther (fig. 1) and in Ulrich Plenzdorf's modern adaptation Die neuen Leiden desjungen W. (fig. 2) from 1972. Another constitutive element of the plot is the addressee of the messages (Wilhelm/Willi) of the Werther character (Werther/Edgar). With respect to the triangular relationship, we aim to compare the vectors of a single pair, like Werther-Lotte in Goethe's novel, with a corresponding pair, e.g. Edgar-Charlie in Plenzdorf's adaption.

Figure 1: Character network Die Leiden des jungen Werthers (1774); Distance for connections: 8 tokens

Figure 2: Weighted network of Die neuen Leiden des jungen W. (1972); Distance for connections: 8 tokens

Edgar, the Werther character in Plenzdorf's adaption, takes the central node with the highest degree (the maximum number of connections, cf. fig 3). We noticed a slightly different position of the beloved women in both texts: In Goethe's Werther Lotte appears to be very well connected with other nodes, compared to Charlie in Plenzdorf's novel, who is only in contact with Edgar and his antagonist Dieter (cf. fig. 2).

Die neuen Leiden des jungen W. produces a denser network based on fewer character nodes with more weighted edges. That means Edgar has relatively intense contact to other characters besides his message partner. In contrast, many more characters appear in Goethe's novel, but often only once in connection with Werther, which reduces the network's density.

Average Degree



Maximum Degree

21 (Werther)

8 (Edgar)




Figure 3: Network measures

Future prospects
The character constellation is determined by an opposition of the sensitive Werther to the rational opponent character Albert, with their love interest Lotte displaying character traits of both. Furthermore,

an examination of the relationship pairs with regard to their context is planned.

The usefulness of such an approach is illustrated by the sentence below, which defines the relationship of Lotte and Werther in context:

Trilcke, P. (2013): Social Network Analysis (SNA) als Methode einer textempirischen Literaturwissenschaft, in: Philip Ajouri, Katja Mellmann u. Christoph Rauen (Hgg.): Empirie in der Literaturwissenschaft, Münster 2013: 201-247.

Sie [Lotte] stand auf ihrem Ellenbogen gestützt

und ihr Blick durchdrang die Gegend,

sie sah gen Himmel und auf mich, ich [Werther]

sah ihr Auge tränenvoll, sie legte ihre

Hand auf die meinige und sagte — Klopstock!

The term "Klopstock" is distinctive for the “WertherLotte” pair. It reflects their spiritual kinship and characterizes their relationship. However, this expression is found exactly once in Goethe's epistolary novel. Overall, the "Werther-Lotte" pair appears 81 times in the text. Based on these instances we aim to characterize their relationship even with specific terms like "Tanze", "Porträt" or “Klopstock”. This will complete the network visualization of the characters with a description of their individual relations, both in Goethe’s novel and its adaptations.

As noted in the Introduction, the Werther Project is part of CRETA, a research center connecting both scholars from Humanities/Social Sciences and Computational Linguistics/Computer Science at the University of Stuttgart. We are grateful to the German Federal Ministry of Education and Research (BMBF) for its generous funding in the years 2016 until 2018 (funding reference number: 01UG1601)

Hettinger, L. et al. (2015): Genre Classification on German Novels, Proceedings of the 12th International Workshop on Text-based Information Retrieval, Valencia.

Horré, T. (1997): Werther-Roman und Werther-Figur in der deutschen Prosa des Wilhelminischen Zeitalters, St. Ingbert.

Martens, L. (1985): The Diary Novel, Cambridge.

Moretti, F. (2011): Network Theory, Plot Analysis, in; New Left Review 68: 80-102.

Pfister, M. (1982): Das Drama. Theorie und Analyse, München: 232-250.

If this content appears in violation of your intellectual property rights, or you see errors or omissions, please reach out to Scott B. Weingart to discuss removing or amending the materials.

Conference Info


ADHO - 2017

Hosted at McGill University, Université de Montréal

Montréal, Canada

Aug. 8, 2017 - Aug. 11, 2017

438 works by 962 authors indexed

Series: ADHO (12)

Organizers: ADHO