<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.2 20120330//EN" "http://jats.nlm.nih.gov/publishing/1.2/JATS-journalpublishing1.dtd">
<!--<?xml-stylesheet type="text/xsl" href="article.xsl"?>-->
<article article-type="research-article" dtd-version="1.2" xml:lang="en" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<front>
<journal-meta>
<journal-id journal-id-type="issn">2940-1348</journal-id>
<journal-title-group>
<journal-title>Journal of Computational Literary Studies</journal-title>
</journal-title-group>
<issn pub-type="epub">2940-1348</issn>
<publisher>
<publisher-name>Universit&#228;ts- und Landesbibliothek Darmstadt</publisher-name>
</publisher>
</journal-meta>
<article-meta>
<article-id pub-id-type="doi">10.48694/jcls.108</article-id>
<article-categories>
<subj-group>
<subject>Article</subject>
</subj-group>
</article-categories>
<title-group>
<article-title>Topic Modeling for the Identification of Gender-specific Discourse</article-title>
<subtitle>Virtues and Vices in French and Spanish 18<sup>th</sup> Century Periodicals</subtitle>
</title-group>
<contrib-group>
<contrib contrib-type="author" corresp="yes">
<contrib-id contrib-id-type="orcid">https://orcid.org/0000-0001-8625-3663</contrib-id>
<name>
<surname>V&#246;lkl</surname>
<given-names>Yvonne</given-names>
</name>
<xref ref-type="aff" rid="aff-1">1</xref>
</contrib>
<contrib contrib-type="author">
<contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-0802-6999</contrib-id>
<name>
<surname>Sari&#263;</surname>
<given-names>Sanja</given-names>
</name>
<xref ref-type="aff" rid="aff-2">2</xref>
</contrib>
<contrib contrib-type="author">
<contrib-id contrib-id-type="orcid">https://orcid.org/0000-0003-1438-3236</contrib-id>
<name>
<surname>Scholger</surname>
<given-names>Martina</given-names>
</name>
<xref ref-type="aff" rid="aff-2">2</xref>
</contrib>
</contrib-group>
<aff id="aff-1"><label>1</label>Institute of Interactive Systems and Data Science (ISDS), Graz University of Technology <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://ror.org/00d7xrm67">ROR</ext-link>, Graz, Austria.</aff>
<aff id="aff-2"><label>2</label>Institute Centre for Information Modelling - Austrian Centre for Digital Humanities, University of Graz <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://ror.org/01faaaf77">ROR</ext-link>, Graz, Austria.</aff>
<pub-date publication-format="electronic" date-type="pub" iso-8601-date="2022-11-24">
<day>24</day>
<month>11</month>
<year>2022</year>
</pub-date>
<pub-date pub-type="collection">
<year>2022</year>
</pub-date>
<volume>1</volume>
<issue>1</issue>
<fpage>1</fpage>
<lpage>29</lpage>
<history>
<date date-type="received" iso-8601-date="2021-12-22">
<day>22</day>
<month>12</month>
<year>2021</year>
</date>
<date date-type="accepted" iso-8601-date="2022-03-30">
<day>30</day>
<month>03</month>
<year>2022</year>
</date>
</history>
<permissions>
<copyright-statement>Copyright: &#x00A9; 2022 The Author(s)</copyright-statement>
<copyright-year>2022</copyright-year>
<license license-type="open-access" xlink:href="https://creativecommons.org/licenses/by/4.0/">
<license-p>The text of this work is released under the Creative Commons license CC BY 4.0 International. You can find the contract text of the license at <uri xlink:href="https://creativecommons.org/licenses/by/4.0/">https://creativecommons.org/licenses/by/4.0/</uri>. The illustrations are excluded from this license, here the copyright lies with the respective rights holder.</license-p>
</license>
</permissions>
<self-uri xlink:href="https://jcls.io/article/10.48694/jcls.108/"/>
<abstract>
<p>Gender-specific knowledge &#8211; just like knowledge in general &#8211; is generated through discourses that are disseminated through (mass) media. Among the first mass media is the Spectator press (<italic>Moralische Wochenschriften</italic>) which spread all over Europe throughout the 18<sup>th</sup> century. With their gender-specific discourses, analyzed in <italic>Spectatoriale Geschlechterkonstruktionen</italic> (<xref ref-type="bibr" rid="B58">V&#246;lkl 2022</xref>), they decisively promote the development of a (bourgeois) gender model, shaping the social perception of gender until today. Against this background, the present article examines the gender-specific discourses in the French and Spanish Spectator periodicals by means of topic modeling which detects semantically related words. The study, which originates from the project <italic>Distant Spectators. Distant Reading for Periodicals of the Enlightenment</italic> (<xref ref-type="bibr" rid="B52">Scholger et al. 2019&#8211;2021</xref>), shows that topic modeling reinforces previous findings on gender-specific discourses in the Spectator periodicals. Moreover, it offers new perspectives concerning this research corpus.</p>
</abstract>
<kwd-group>
<kwd>topic modeling</kwd>
<kwd>French</kwd>
<kwd>Spanish</kwd>
<kwd>Spectator press</kwd>
<kwd>18th century</kwd>
<kwd>Literary Gender Studies</kwd>
</kwd-group>
</article-meta>
</front>
<body>
<sec id="S1">
<title>1. Introduction</title>
<p>The Spectator periodicals are a popular journalistic genre of the 18<sup>th</sup> century which (co-)constructs and preserves the cultural knowledge of its time in general and gender-specific knowledge in particular, propagating a heteronormative society. As a broadly effective medium circulating from England throughout the Western world, the Spectator periodicals also promote the transcultural dissemination of a transforming understanding of gender,<xref ref-type="fn" rid="n1">1</xref> in conjunction with the changing values, norms, and practices among the constituting middle classes.</p>
<p>The quantitative-statistical as well as the discourse-analytical and interpretative study on gender-specific ways of worldmaking in the French- and Spanish-language Spectator periodicals (cf. <xref ref-type="bibr" rid="B58">V&#246;lkl 2022</xref>) reveals that the French-language periodicals of the first half of the 18<sup>th</sup> century contribute to the dissemination of the notion of a &#8216;natural&#8217; gender difference which primarily appears together with a discourse of character and/or physical differences. From the middle of the century onwards, in which the periodicals are also published in Spain,<xref ref-type="fn" rid="n2">2</xref> the discourse of difference is expanded to include the aspect of complementarity, finally recognizing woman and man as a mutually complementary entity. Due to her alleged closeness to nature, in this discourse of complementarity, the woman is hierarchically placed under the authority of man, whose assumed higher ability to reason is considered superior.</p>
<p>In order to disseminate the discourses of difference and complementarity, the French- and Spanish-language periodicals draw particularly on the notion of virtue (French: <italic>vertu</italic>; Spanish: <italic>virtud</italic>). According to research on the Enlightenment period, this term generally functions as a gender-specific key concept (&#8216;geschlechtsspezifischer Leitbegriff&#8217; according to <xref ref-type="bibr" rid="B45">Pabst 2007</xref>) and stands in opposition to the notion of vice (French: <italic>vice</italic>; Spanish: <italic>vicio</italic>) (cf. <xref ref-type="bibr" rid="B6">Bolufer Peruga 1998</xref>, <xref ref-type="bibr" rid="B33">Kilian 2002</xref>, <xref ref-type="bibr" rid="B48">Schaufler 2002</xref>, <xref ref-type="bibr" rid="B55">Steinbr&#252;gge 1987</xref>). Furthermore, the discourse on virtues and vices is combined with positive and negative (character) traits and behavioral patterns which are hierarchized and assessed as worthy or not worthy of emulation. Among the ignoble vices on the one hand, one can find, e.g., hypocrisy, idleness, vanity, or jealousy which should be avoided by women and men alike (and thus remain gender-<italic>un</italic>specific). The virtues worthy of emulation on the other hand, are constructed in a gender-specific way, with the &#8216;female&#8217; virtues revolving around concepts such as decency, modesty, kindness, shamefulness, beauty, or (a specific female) education, while the &#8216;male&#8217; virtues only include (a specific male) education, honesty, and reason. In order to make the large number of virtues and vices known to the Spectator audience &#8211; which decidedly also included women &#8211; they are incorporated into gender-stereotypical models, illustrating ideal images or warning examples. E.g., the characteristics of egoism and vanity which are considered vicious, are linked to the stereotypical models of the coquette or the fop and contrasted with virtuous models of women and men. The gender-stereotypical models with their manifold virtues and vices are enveloped into countless (exemplary) stories from everyday life and in (character) portraits which are narratively woven into the plot (cf. <xref ref-type="bibr" rid="B58">V&#246;lkl 2022, 282&#8211;286</xref>).</p>
<p>To quantitatively verify these observations on the (entire) Spectator corpus, a topic modeling analysis was carried out in the course of the project <italic>Distant Spectators. Distant Reading for Periodicals of the Enlightenment (DiSpecs)</italic> (<xref ref-type="bibr" rid="B52">Scholger et al. 2019&#8211;2021</xref>) after which special attention was given to the interpretation of those topics that stand out from a gender-theoretical perspective. The computed values and their visual representation were intended to provide a new perspective on the corpus and create new theories and questions. The following chapters first describe the related work, the research corpus, and the methodology before presenting the results and findings of the topic modeling analysis with regard to gender-specific discourse in the Spectator periodicals.</p>
</sec>
<sec id="S2">
<title>2. Related Work</title>
<p>Topic modeling has become an integral part of the range of methods used in digital humanities, and more specifically in computational literary studies. According to the survey of Du, it has been increasingly used since 2011 (cf. <xref ref-type="bibr" rid="B15">Du 2019</xref>). In the field of historical newspapers and periodicals, topic modeling was conducted for analyzing the social and political life of Civil War Richmond based on the <italic>Richmond Daily Dispatch</italic> (cf. <xref ref-type="bibr" rid="B43">Nelson 2020</xref>) and for investigating the discourse dynamics in historical newspapers published in Finland between 1854 and 1917 (cf. <xref ref-type="bibr" rid="B37">Marjanen et al. 2020</xref>). Regarding the Enlightenment period, Sch&#246;ch applied topic modeling on French Classical and Enlightenment drama for sub-genre classification (cf. <xref ref-type="bibr" rid="B50">Sch&#246;ch 2017</xref>), and Roe et al. analyzed the discursive structure in the <italic>Encyclop&#233;die</italic> of Denis Diderot and Jean le Rond d&#8217;Alembert (cf. <xref ref-type="bibr" rid="B47">Roe et al. 2016</xref>).</p>
<p>A persistent point of criticism in the application of topic modeling is the lack of explainability and comprehensibility of the results (cf. <xref ref-type="bibr" rid="B30">Hu et al. 2014, 424&#8211;425</xref>, <xref ref-type="bibr" rid="B36">Liu et al. 2017, 1&#8211;2</xref>). This is very much related to the lack of documentation of single working steps and parameters applied in the topic modeling process as Du pointed out (cf. <xref ref-type="bibr" rid="B15">Du 2019</xref>): In order to guarantee the reproducibility of the results, it is crucial to have details on the number of documents, the conducted pre-processing steps, the number of topics and iterations selected in the actual modeling process, etc. To address this criticism, this contribution aims to not only provide the results of our topic modeling analyses, but also to transparently document the workflow that led to them.</p>
</sec>
<sec id="S3">
<title>3. The Research Corpus</title>
<p>While the Spectators have been studied through close reading as a work-centered approach, there have been no previous activities that explore this genre from a distant reading perspective. For this reason, the project DiSpecs engaged in text mining of the collection of 3,863 periodical issues in six languages,<xref ref-type="fn" rid="n3">3</xref> assembled and edited during the digital scholarly edition project <italic>The &#8216;Spectators&#8217; in the International Context</italic> (<xref ref-type="bibr" rid="B17">Ertler et al. 2011&#8211;2021</xref>). In the DiSpecs project, topic modeling was used for investigating the semantic and stylistic structure.</p>
<p>What proved to be very useful for the analysis was the fact that the documents were already available in XML/TEI format (<xref ref-type="bibr" rid="B57">TEI Consortium 2021</xref>). This includes not only the annotation of metadata and structural elements (e.g., paragraphs and pagination), but also narrative forms (e.g., self-portrait, letter/letter to the editor, fable) and narrative levels of representation,<xref ref-type="fn" rid="n4">4</xref> as well as subjects (e.g., &#8216;Idea of man&#8217;, &#8216;Nature&#8217;, &#8216;Economy&#8217;, &#8216;Theatre Literature Arts&#8217;), mentioned places, person names, and intellectual works. The annotation format provided through the application of the Text Encoding Initiative (TEI) standard enables easier extraction of certain structures of the data for the analysis (e.g., metadata, headings, footnotes, editorial comments) with the possibility to separate issues into paragraphs and to exclude parts or whole issues during the pre-processing of the data which will be explained closer in <xref ref-type="sec" rid="S4.2">subsection 4.2</xref>.</p>
</sec>
<sec id="S4">
<title>4. Topic Modeling Workflow</title>
<p>The unsupervised probabilistic topic modeling method aims to identify hidden thematic structures in large text collections (cf. <xref ref-type="bibr" rid="B2">Blei 2020, 8</xref>) which means that the algorithm recognizes patterns in the data without having a training subset or a desired output (cf. <xref ref-type="bibr" rid="B1">Alloghani et al. 2020, 4</xref>). The resulting topics usually consist of thematically related words, i.e., tokens. However, some topics have structural rather than thematic significance. They can provide insight into the writing style of the author, terms typical for a genre, adjectives describing a matter, repeatedly mentioned places or persons. This is due to the fact that the method&#8217;s algorithms measure the co-occurrence of words, following statistical assumptions, meaning that if the same words often occur together in documents, they are most likely thematically related (cf. <xref ref-type="bibr" rid="B2">Blei 2020, 9</xref>).</p>
<p>Multiple algorithms were developed for topic modeling, but one of the most prominent, and the one we used in our analysis, is Latent Dirichlet Allocation (LDA). We owe this choice to the DARIAH-DE team who developed a Jupyter Notebook embedding the <italic>dariah_topics</italic> Python library for topic modeling (<xref ref-type="bibr" rid="B11">DARIAH-DE 2019</xref>) with MALLET (<xref ref-type="bibr" rid="B40">McCallum 2002&#8211;2018</xref>), a toolkit that builds on LDA. We adapted and expanded these notebooks to incorporate them into our topic modeling analysis workflow,<xref ref-type="fn" rid="n5">5</xref> which can be divided into four main parts: data evaluation, pre-processing of the data, topic modeling creation and post-processing of the results (<xref ref-type="fig" rid="F1">Figure 1</xref>). As we demonstrate in this chart, individual steps of the workflow have to be repeated to optimize the results. Further on in this chapter, we will describe how we conducted these steps and what decisions were important for quality results.</p>
<fig id="F1">
<label>Figure 1</label>
<caption>
<p>The topic modeling workflow.</p>
</caption>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="jcls-108_volkl-g1.png"/>
</fig>
<sec id="S4.1">
<title>4.1 Data Evaluation</title>
<p>To get an overview of the French- and Spanish-language research data, we conducted a number of exploratory data analysis steps. This included evaluating and visualizing the size of the corpus, the number of issues per periodical and per author, the number of tokens per issue as well as the distribution of manually assigned keywords and narrative forms. This simple statistical analysis allows insight into the corpus which can be relevant for interpreting and evaluating the results. For example, comparing manually assigned keywords with topics identified through topic modeling is used for cross-evaluation of these two approaches, i.e., finding out how similar the human- and the machine-assigned topics are. In addition, discrepancies in the metadata could be detected and corrected. Getting this insight was possible thanks to the TEI annotation of the Spectator corpus which allows extracting all the relevant data structures from the documents either with Python libraries like <italic>Beautiful Soup</italic> or with XSLT while transforming the XML/TEI to plain text files.</p>
</sec>
<sec id="S4.2">
<title>4.2 Pre-processing</title>
<p>Our workflow, building on DARIAH-DE, required plain text files as input for topic modeling. As part of this transformation, we extracted metadata from the TEI files and used it to build file names: publication year, periodical name, author of the periodical, volume, issue, and persistent identifier of the file. This way, we had easy access to specific parts of the files&#8217; metadata even when using plain text files. We also filtered the text material. On the one hand, we divided the collection into separate corpora according to their language and excluded files that did not contain manually assigned keywords (e.g., tables of contents). On the other hand, we removed titles and subtitles since they were repetitive and therefore had a disproportionate impact on the result.</p>
<p>These plain text files were already fulfilling formal requirements to proceed with topic modeling, meaning they were classified per language, in the desired format, and containing metadata in the file names. But LDA treats inflected forms (e.g., Span. <italic>mujer</italic> &#8211; &#8216;woman&#8217;, and <italic>mujeres</italic> &#8211; &#8216;women&#8217;, or <italic>muger/mugeres</italic> in 18<sup>th</sup> century orthography) as different concepts. A topic can therefore include multiple forms of the same concept which can result in semantically poor topics. To avoid this, we decided to lemmatize the Spectator texts before modeling the topics, using natural language processing with <italic>spaCy</italic> to replace each inflected word (e.g., <italic>mugeres</italic>) with its lexical base form (<italic>mujer</italic>), i.e., lemma. This step was one of the most challenging since <italic>spaCy</italic> was not trained on historical language. Wrongly lemmatized tokens had to be replaced with the correct lemma through a dictionary. Although still not without errors, the decision to lemmatize brought much cleaner and semantically richer results than the preliminary experiments with non-lemmatized texts.</p>
<p>Since topic modeling measures the co-occurrence frequency of tokens in the same document, another pre-processing step was to define what will be treated as a document. We decided to segment the issues in paragraphs with a minimum of 500 tokens, whereby longer paragraphs were avoided by cutting off a paragraph after the first following sentence&#8217;s end, if the number of included tokens had surpassed 600. Remaining paragraphs with less than 200 tokens were appended to the preceding paragraphs of the same issue, to avoid very short paragraphs. Although there were still a few outlier paragraphs left, this method resulted in a larger quantity of documents with a similar token number instead of a smaller quantity of documents with more strongly varying token numbers. Since there is no state-of-the-art consensus on the optimal number of tokens in a document, the selection was based on preliminary experiments with different values.</p>
<p>With this set of resized and lemmatized documents, we continued with the workflow as provided by DARIAH-DE, with some practical adjustments. From the imported and tokenized documents, we removed redundant tokens as a last pre-processing step since some tokens do not have semantic significance or are simply irrelevant and therefore not desired to be part of the final result. These tokens are a) the 100 most frequent words (MFW) because they tend to be functional words, like pronouns, articles, prepositions, etc., b) the hapax legomena (tokens occurring only once in the corpus), and c) a project specific stop word list. To create the stop word lists, we adjusted the <italic>Stopwords ISO</italic> (<xref ref-type="bibr" rid="B12">Diaz 2016</xref>) lists and expanded them after each of our topic modeling cycles with new resulting topic keywords we identified as irrelevant.<xref ref-type="fn" rid="n6">6</xref></p>
</sec>
<sec id="S4.3">
<title>4.3 Topic Model Creation</title>
<p>Since topic modeling is an unsupervised machine learning method, the researcher cannot impact the result by assigning categories in advance. There are, however, a couple of factors that do influence the results. One of them is, as previously mentioned, the pre-processing of the data. Another one is the choice of the input parameters: the number of topics, the number of iterations and the hyperparameter optimization interval. <xref ref-type="table" rid="T1">Table 1</xref> gives an overview of relevant parameters in our topic modeling analysis.</p>
<table-wrap id="T1">
<label>Table 1</label>
<caption>
<p>Parameters used in the topic modeling analysis of French- and Spanish-language periodicals.</p>
</caption>
<table>
<tr>
<td colspan="3"><hr/></td>
</tr>
<tr>
<td align="left" valign="top"></td>
<td align="left" valign="top">French</td>
<td align="left" valign="top">Spanish</td>
</tr>
<tr>
<td colspan="3"><hr/></td>
</tr>
<tr>
<td align="left" valign="top">Number of periodicals</td>
<td align="left" valign="top">25</td>
<td align="left" valign="top">18</td>
</tr>
<tr>
<td align="left" valign="top">Number of issues</td>
<td align="left" valign="top">1,658</td>
<td align="left" valign="top">690</td>
</tr>
<tr>
<td align="left" valign="top">Extracted segments</td>
<td align="left" valign="top">6,752</td>
<td align="left" valign="top">3,190</td>
</tr>
<tr>
<td align="left" valign="top">Lemmatization</td>
<td align="left" valign="top">yes</td>
<td align="left" valign="top">yes</td>
</tr>
<tr>
<td align="left" valign="top">Removed features</td>
<td align="left" valign="top">100 MFW, 801 stop words, hapax legomena</td>
<td align="left" valign="top">100 MFW, 823 stop words, hapax legomena</td>
</tr>
<tr>
<td align="left" valign="top">Number of topics</td>
<td align="left" valign="top">25</td>
<td align="left" valign="top">18</td>
</tr>
<tr>
<td align="left" valign="top">Iterations</td>
<td align="left" valign="top">2,000</td>
<td align="left" valign="top">2,000</td>
</tr>
<tr>
<td align="left" valign="top">Alfa hyperparameter</td>
<td align="left" valign="top">5.0 (MALLET default)</td>
<td align="left" valign="top">5.0 (MALLET default)</td>
</tr>
<tr>
<td align="left" valign="top">Beta hyperparameter</td>
<td align="left" valign="top">0.01 (MALLET default)</td>
<td align="left" valign="top">0.01 (MALLET default)</td>
</tr>
<tr>
<td align="left" valign="top">Hyperparameter optimization</td>
<td align="left" valign="top">20</td>
<td align="left" valign="top">20</td>
</tr>
<tr>
<td colspan="3"><hr/></td>
</tr>
</table>
</table-wrap>
<p>The number of topics is thus decided by the researcher. The reasonable number of topics in a text collection depends on the text scope, but also the genre and the thematic richness. Our approach was to experiment with different numbers of topics and evaluate the results to decide the optimal number of topics for each text corpus. Eventually, we determined 25 French and 18 Spanish topics. The number of topic keywords, on the other hand, is not a matter of the researcher&#8217;s decision: each topic consists of all tokens from the text collection, whereas the distribution of these tokens varies in the individual topics. So, each token from the treated text collection can be found in each resulting topic, but with a varying probability which is never equal to 0% (cf. S. <xref ref-type="bibr" rid="B5">Bock et al. 2016, 13</xref>). The researcher familiar with the analyzed content then decides on how many of the topic tokens, i.e., keywords, they see as significant to represent in the results. For each analyzed group, we chose to output the first 20 tokens.</p>
<p>The researcher also sets the number of iterations. More iterations lead to a longer processing time but can lead to more reliable and stable results until a limit is reached after which the quality stagnates (cf. <xref ref-type="bibr" rid="B32">Jockers 2014, 147</xref>). Choosing an optimization interval is optional and depends on the desire to observe the difference in topic weight, by &#8220;allowing some topics to be more prominent than others&#8221; (<xref ref-type="bibr" rid="B40">McCallum 2002&#8211;2018</xref>).<xref ref-type="fn" rid="n7">7</xref> In our analysis, we conducted 2,000 iterations with an optimization in every 20 iterations.</p>
<p>But even with the same data and the same parameters, the output of two modeling cycles is never exactly the same in terms of the topics per document and the words per topic distribution, due to the probabilistic and unsupervised nature of the method. Nevertheless, using our data and parameters, the comparison of multiple results showed a re-emergence of the same topics, with rather insubstantial differences in the sequence of the most frequent topic keywords as well as the probability of the topics which suggests a sufficient stability of the model.</p>
</sec>
<sec id="S4.4">
<title>4.4 Post-processing</title>
<p>The last step in the topic modeling workflow is the post-processing of the results. The probability of topics is being computed for each individual document (which, as explained in <xref ref-type="sec" rid="S4.2">subsection 4.2</xref>, is a segment of a periodical&#8217;s issue). Using these values, we computed the probability of topics per periodical and represented the results from different perspectives, utilizing multiple visualization techniques.<xref ref-type="fn" rid="n8">8</xref></p>
<p>A common way to represent topics are heat maps. The heat map (<xref ref-type="fig" rid="F2">Figure 2</xref>) is a visual representation of the data frame matrix (<xref ref-type="table" rid="T2">Table 2</xref>) resulting from the topic modeling and the computed results per periodical, consisting of periodicals (X axis), topics (Y axis), and the probabilities of each topic per periodical as values where darker color represents higher probability.</p>
<fig id="F2">
<label>Figure 2</label>
<caption>
<p>Heat map detail of 3 periodicals.</p>
</caption>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="jcls-108_volkl-g2.png"/>
</fig>
<table-wrap id="T2">
<label>Table 2</label>
<caption>
<p>Data frame matrix detail corresponding to the heat map detail in <xref ref-type="fig" rid="F2">Figure 2</xref>.</p>
</caption>
<table>
<tr>
<td colspan="4"><hr/></td>
</tr>
<tr>
<td align="left" valign="top"></td>
<td align="left" valign="top">La Bagatelle</td>
<td align="left" valign="top">La Bigarure</td>
<td align="left" valign="top">La Spectatrice</td>
</tr>
<tr>
<td colspan="4"><hr/></td>
</tr>
<tr>
<td align="left" valign="top">&#8230;</td>
<td align="left" valign="top">&#8230;</td>
<td align="left" valign="top">&#8230;</td>
<td align="left" valign="top">&#8230;</td>
</tr>
<tr>
<td align="left" valign="top">Topic 5</td>
<td align="left" valign="top">0.0196085</td>
<td align="left" valign="top">0.01663</td>
<td align="left" valign="top">0.126381</td>
</tr>
<tr>
<td align="left" valign="top">Topic 4</td>
<td align="left" valign="top">0.011347</td>
<td align="left" valign="top">0.0313603</td>
<td align="left" valign="top">0.03801</td>
</tr>
<tr>
<td align="left" valign="top">Topic 3</td>
<td align="left" valign="top">0.004227</td>
<td align="left" valign="top">0.00814</td>
<td align="left" valign="top">0.08031</td>
</tr>
<tr>
<td align="left" valign="top">Topic 2</td>
<td align="left" valign="top">0.03328</td>
<td align="left" valign="top">0.0153783</td>
<td align="left" valign="top">0.05184</td>
</tr>
<tr>
<td align="left" valign="top">Topic 1</td>
<td align="left" valign="top">0.02778</td>
<td align="left" valign="top">0.0434557</td>
<td align="left" valign="top">0.01843</td>
</tr>
<tr>
<td align="left" valign="top">Topic 0</td>
<td align="left" valign="top">0.064768</td>
<td align="left" valign="top">0.0685915</td>
<td align="left" valign="top">0.04833</td>
</tr>
<tr>
<td colspan="4"><hr/></td>
</tr>
</table>
</table-wrap>
<p>For each periodical as well as for each topic, we created a bar chart (e.g., <xref ref-type="fig" rid="F3">Figure 3</xref>). This technique offers a focus on one periodical or topic, whereas the heat map is better suited for getting an overview of the whole corpus. Both the heat map and the bar chart creation are part of the original DARIAH-DE Jupyter Notebook.</p>
<fig id="F3">
<label>Figure 3</label>
<caption>
<p>Distribution of topic 22 in French-language periodicals.</p>
</caption>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="jcls-108_volkl-g3.png"/>
</fig>
<p>Additionally, we decided to use word clouds to visualize the top 100 keywords of the respective topics. A larger font size indicates a keyword with a higher probability inside a topic (<xref ref-type="fig" rid="F4">Figure 4</xref>). We chose this visualization method despite some critics claiming that it is difficult for users to infer the relationship of the words from it (cf. <xref ref-type="bibr" rid="B14">Dobson 2021, &#167;20</xref>). While we do agree with this point, a visual overview of all topic keywords is beneficial next to a visualization of the topic distribution, especially when the topics are not labeled.</p>
<fig id="F4">
<label>Figure 4</label>
<caption>
<p>100 MFW in topic 24 in French-language periodicals.</p>
</caption>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="jcls-108_volkl-g4.png"/>
</fig>
<p>Another way we used Python libraries to represent topics in selected periodicals is with line diagrams which show the prevalence of topics over the issues of a single periodical, i.e., over time. This is only a relative prevalence over time since the time span between the issues was not always constant and is not explicitly available in the metadata. The interactive diagram (created using the library <italic>bokeh</italic>) can be viewed on our project website.<xref ref-type="fn" rid="n9">9</xref> Here, it is possible to zoom in, create sections, activate or deactivate the visibility of individual topics, and save the created versions of the diagram.</p>
<p>Finally, we used the software <italic>Gephi</italic> to create networks of topics, periodicals, and manually assigned keywords (<ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://gephi.org/">Gephi.org</ext-link>&#160;<xref ref-type="bibr" rid="B21">2008&#8211;2021</xref>). More precisely, we created a force-directed graph using the algorithms Fruchterman Reingold and Force Atlas 2 (<xref ref-type="fig" rid="F7">Figure 7</xref>).<xref ref-type="fn" rid="n10">10</xref> The periodical nodes are represented as pie charts showing the distribution of a certain manually assigned keyword throughout the periodical&#8217;s issues. The web presentations include color legends and numerical data for the pie charts. The size of a periodical node (pie chart) indicates whether the number of analyzed periodical issues from the topic modeling set is larger or smaller in comparison to other periodicals. Note that numerous issues do not mean the same as a large amount of text since some issues can be very long while others are quite short. The size of the topic nodes indicates whether a topic has a high or low representativity in the analyzed set of texts. The edges are higher weighted (thicker) if the likelihood of a topic in a periodical is higher. Nodes with the same color belong to the same community. This means that the densities of the edges between these nodes are higher than from these nodes towards the rest of the network. But, since this is a small network where all topics occur in all periodicals to some degree, the weighted modularity of this network is low, and the community structure is not perfectly clear. Nevertheless, it is possible to detect topics that often co-occurred in periodicals.</p>
<p>As shown by the visualizations, the topics are non-semantically labeled (Topic 0, Topic 1, Topic 2, &#8230;), and the numbers give no statement about the importance or frequency of the topic but are only used to distinguish the topics. This approach is contrary to the occasionally seen practice where researchers either label their topics by interpreting them (cf. <xref ref-type="bibr" rid="B7">Boyd-Graber et al. 2017, 40</xref>, or <xref ref-type="bibr" rid="B3">Blevins 2010</xref>) or by using a few of the most relevant keywords, as proposed by the DARIAH-DE Jupyter Notebook. In recent years, we noticed an increase in the non-semantic labeling approach (cf. <xref ref-type="bibr" rid="B29">Horstmann and Kleymann 2019</xref>, <xref ref-type="bibr" rid="B34">Krautter et al. 2020</xref>, or <xref ref-type="bibr" rid="B10">Chehal et al. 2021</xref>). We also decided to proceed without labels because the interpretation of a topic depends on the reception horizon of the researcher. This further impedes the obtrusion of a certain perspective and leaves room for different interpretations. We did, however, provide our interpretation in textual form. The gender-specific topics will be elaborated in <xref ref-type="sec" rid="S5">section 5</xref> and <xref ref-type="sec" rid="S6">section 6</xref>.</p>
<p>To ensure transparency and comprehensibility of the visualizations and interpretations, all the underlying raw data can be downloaded by the user, including the topic keywords list and the word weights. Nevertheless, it has to be pointed out that for understanding the results of distant reading, a certain familiarity with the source material through close reading expertise is always required to create meaning from the results and generate added value for related research. As Shadrova also suggests, &#8220;[i]t is of crucial importance to make the underlying contextualization, the model, explicit, both through hypothesis-based work and by tying results back to the theoretical and conceptual debates in the field&#8221; (<xref ref-type="bibr" rid="B53">Shadrova 2021, 16</xref>).</p>
</sec>
</sec>
<sec id="S5">
<title>5. Topic Modeling in the French-language Spectator Periodicals</title>
<p>Among the 25 topics of the Spectator periodicals published in French language, at least six topics stand out from a gender-specific perspective. Topics 4, 22, and 24 directly, topics 9, 18, and, 21 indirectly relate to character, behavior, and roles of women and men within the (emerging bourgeois) society in the 18<sup>th</sup> century (see <xref ref-type="table" rid="T3">Table 3</xref>).</p>
<table-wrap id="T3">
<label>Table 3</label>
<caption>
<p>Gender-specific topics in French-language Spectators.</p>
</caption>
<table>
<tr>
<td colspan="6"><hr/></td>
</tr>
<tr>
<td align="left" valign="top">Topic 4</td>
<td align="left" valign="top">Topic 9</td>
<td align="left" valign="top">Topic 18</td>
<td align="left" valign="top">Topic 21</td>
<td align="left" valign="top">Topic 22</td>
<td align="left" valign="top">Topic 24</td>
</tr>
<tr>
<td colspan="6"><hr/></td>
</tr>
<tr>
<td align="left" valign="top">fille</td>
<td align="left" valign="top">vertu</td>
<td align="left" valign="top">heureux</td>
<td align="left" valign="top">aimer</td>
<td align="left" valign="top">dame</td>
<td align="left" valign="top">air</td>
</tr>
<tr>
<td align="left" valign="top">jeune</td>
<td align="left" valign="top">m&#233;rite</td>
<td align="left" valign="top">dieu</td>
<td align="left" valign="top">sentir</td>
<td align="left" valign="top">quoiqu</td>
<td align="left" valign="top">bel</td>
</tr>
<tr>
<td align="left" valign="top">pere</td>
<td align="left" valign="top">vie</td>
<td align="left" valign="top">doux</td>
<td align="left" valign="top">bonheur</td>
<td align="left" valign="top">&#233;gard</td>
<td align="left" valign="top">sexe</td>
</tr>
<tr>
<td align="left" valign="top">mariage</td>
<td align="left" valign="top">nature</td>
<td align="left" valign="top">oeil</td>
<td align="left" valign="top">passion</td>
<td align="left" valign="top">mani&#232;re</td>
<td align="left" valign="top">dame</td>
</tr>
<tr>
<td align="left" valign="top">fils</td>
<td align="left" valign="top">ame</td>
<td align="left" valign="top">tendre</td>
<td align="left" valign="top">lettre</td>
<td align="left" valign="top">passion</td>
<td align="left" valign="top">beaut&#233;</td>
</tr>
<tr>
<td align="left" valign="top">mere</td>
<td align="left" valign="top">hommes</td>
<td align="left" valign="top">tendre</td>
<td align="left" valign="top">amant</td>
<td align="left" valign="top">tem</td>
<td align="left" valign="top">jeune</td>
</tr>
<tr>
<td align="left" valign="top">mari</td>
<td align="left" valign="top">propre</td>
<td align="left" valign="top">main</td>
<td align="left" valign="top">tendre</td>
<td align="left" valign="top">sexe</td>
<td align="left" valign="top">visage</td>
</tr>
<tr>
<td align="left" valign="top">famille</td>
<td align="left" valign="top">bonheur</td>
<td align="left" valign="top">ciel</td>
<td align="left" valign="top">moment</td>
<td align="left" valign="top">mauvais</td>
<td align="left" valign="top">oeil</td>
</tr>
<tr>
<td align="left" valign="top">p&#232;re</td>
<td align="left" valign="top">vice</td>
<td align="left" valign="top">ame</td>
<td align="left" valign="top">malheureux</td>
<td align="left" valign="top">propre</td>
<td align="left" valign="top">plaire</td>
</tr>
<tr>
<td align="left" valign="top">&#226;ge</td>
<td align="left" valign="top">passion</td>
<td align="left" valign="top">voix</td>
<td align="left" valign="top">douleur</td>
<td align="left" valign="top">convenir</td>
<td align="left" valign="top">aimable</td>
</tr>
<tr>
<td align="left" valign="top">enfan</td>
<td align="left" valign="top">heureux</td>
<td align="left" valign="top">feu</td>
<td align="left" valign="top">sentiment</td>
<td align="left" valign="top">peine</td>
<td align="left" valign="top">mode</td>
</tr>
<tr>
<td align="left" valign="top">marier</td>
<td align="left" valign="top">conduite</td>
<td align="left" valign="top">aimable</td>
<td align="left" valign="top">perdre</td>
<td align="left" valign="top">conduite</td>
<td align="left" valign="top">habit</td>
</tr>
<tr>
<td align="left" valign="top">chevalier</td>
<td align="left" valign="top">action</td>
<td align="left" valign="top">charme</td>
<td align="left" valign="top">malheur</td>
<td align="left" valign="top">obliger</td>
<td align="left" valign="top">joli</td>
</tr>
<tr>
<td align="left" valign="top">&#233;pouser</td>
<td align="left" valign="top">sage</td>
<td align="left" valign="top">gloire</td>
<td align="left" valign="top">heureux</td>
<td align="left" valign="top">penser</td>
<td align="left" valign="top">figure</td>
</tr>
<tr>
<td align="left" valign="top">tendresse</td>
<td align="left" valign="top">honneur</td>
<td align="left" valign="top">peine</td>
<td align="left" valign="top">tendresse</td>
<td align="left" valign="top">montrer</td>
<td align="left" valign="top">femmes</td>
</tr>
<tr>
<td align="left" valign="top">demoiselle</td>
<td align="left" valign="top">digne</td>
<td align="left" valign="top">objet</td>
<td align="left" valign="top">devenir</td>
<td align="left" valign="top">dessein</td>
<td align="left" valign="top">go&#251;t</td>
</tr>
<tr>
<td align="left" valign="top">devenir</td>
<td align="left" valign="top">noble</td>
<td align="left" valign="top">bel</td>
<td align="left" valign="top">ame</td>
<td align="left" valign="top">affection</td>
<td align="left" valign="top">conversation</td>
</tr>
<tr>
<td align="left" valign="top">maison</td>
<td align="left" valign="top">mal</td>
<td align="left" valign="top">terrebeaut&#233;</td>
<td align="left" valign="top">oeil</td>
<td align="left" valign="top">devenir</td>
<td align="left" valign="top">grace</td>
</tr>
<tr>
<td align="left" valign="top">soin</td>
<td align="left" valign="top">fortune</td>
<td align="left" valign="top">sage</td>
<td align="left" valign="top">objet</td>
<td align="left" valign="top">libert&#233;</td>
<td align="left" valign="top">rire</td>
</tr>
<tr>
<td align="left" valign="top">amant</td>
<td align="left" valign="top">estime</td>
<td align="left" valign="top">brillant</td>
<td align="left" valign="top">&#233;tois</td>
<td align="left" valign="top">avis</td>
<td align="left" valign="top">compagnie</td>
</tr>
<tr>
<td colspan="6"><hr/></td>
</tr>
</table>
</table-wrap>
<p><bold>Topic 4</bold> lists the various French terms for &#8216;marriage&#8217; and &#8216;getting married&#8217; (<italic>mariage, marier, &#233;pouser</italic>), &#8216;family&#8217; (<italic>famille</italic>), &#8216;child&#8217; (<italic>enfan</italic>), or &#8216;house&#8217; (<italic>maison</italic>) which are terms that construct the destiny of young (!) women (<italic>fille, demoiselle</italic>) within the domestic sphere (in contrast to the public sphere which is attributed to men).<xref ref-type="fn" rid="n11">11</xref> In this private sphere, her main duty is to take tender (<italic>tendresse</italic>) care (<italic>soin</italic>) of her husband (<italic>mari</italic>) and children.</p>
<p><bold>Topic 22</bold> refers to the vocabulary used in the translation of the <italic>Female Spectator</italic> (1749&#8211;51), <italic>La Spectatrice, traduite de l&#8217;anglais</italic> (1750&#8211;51) as indicated in the bar chart with a probability of over 0.3 within this periodical (<xref ref-type="fig" rid="F3">Figure 3</xref>) which is much higher in relation to other periodicals. <italic>La Spectatrice</italic> is one of the few spectatorial titles specifically directed to (bourgeois) women. This focus on the female readership also reverberates in the first term of the topic with &#8216;lady&#8217; (<italic>dame</italic>). The subsequent terms used, such as &#8216;passion&#8217; (<italic>passion</italic>), &#8216;bad&#8217; (<italic>mauvais</italic>), &#8216;suitable&#8217; (<italic>propre</italic>), &#8216;corresponding&#8217; (<italic>convenir</italic>), &#8216;conduct&#8217; (<italic>conduite</italic>), or &#8216;affection&#8217; (<italic>affection</italic>), indicate that this topic is concerned with the behavior of women in public, especially in the company of men.</p>
<p><bold>Topic 24</bold>, visualized as word cloud (<xref ref-type="fig" rid="F4">Figure 4</xref>), also lists attributes associated with the &#8216;fair sex&#8217; (<italic>beau sexe</italic>).<xref ref-type="fn" rid="n12">12</xref> On the one hand, a woman has to &#8216;please&#8217; (<italic>plaire</italic>) through her inner beauty &#8211; expressed by terms such as &#8216;beautiful&#8217; (<italic>bel</italic>), &#8216;beauty&#8217; (<italic>beaut&#233;</italic>), &#8216;amiable&#8217; (<italic>aimable</italic>), and &#8216;grace&#8217; (<italic>gr&#226;ce</italic>) &#8211; and on the other hand through her outer beauty &#8211; expressed as well by the terms &#8216;beautiful&#8217; (<italic>bel</italic>) and &#8216;beauty&#8217; (<italic>beaut&#233;</italic>), but also by &#8216;pretty&#8217; (<italic>joli</italic>) or &#8216;taste&#8217; (<italic>go&#251;t</italic>). Both inner and outer beauty are accentuated by appropriate &#8216;clothing&#8217; (<italic>habit, mode</italic>), good &#8216;taste&#8217; (<italic>go&#251;t</italic>), and &#8216;conversation practices&#8217; (<italic>conversation</italic>) that are understood as suitable for a woman. Her &#8216;appearance&#8217; (<italic>air</italic>), i.e., her outward appearance, has the highest priority here, as can be seen from the prominent position of the term in the first place, and is represented in all periodicals (see also topic 17 of the Spanish periodicals where the orientation on outward appearances manifests through terms such as <italic>moda</italic> &#8211; &#8216;fashion&#8217;, <italic>adornar</italic> &#8211;&#8216;to adorn&#8217;, <italic>gustar</italic> &#8211; &#8216;to please&#8217;, <italic>hermoso</italic> &#8211; &#8216;beautiful&#8217;, <italic>hermosura</italic> &#8211; &#8216;beauty&#8217;).</p>
<p>The discourse on women within the French-language periodicals is further supported by topics 9, 18, and 21. While the first three topics mentioned above explicitly evoke terms for women (e.g., <italic>beau sexe, femme</italic>), and also use self-explaining terms alluding to their status (e.g., <italic>dame</italic> &#8211; &#8216;lady&#8217;, <italic>demoiselle</italic> &#8211; &#8216;unmarried young woman&#8217;, <italic>m&#232;re</italic> &#8211; &#8216;mother&#8217;) as well as gender-specific, heteronormative practices (e.g., <italic>marier, &#233;pouser</italic> &#8211; the act of getting married), the terms used in topics 9, 18, and 21 are more implicit to the extent that they only indirectly allude to the gender-specific discourse and roles of women and men in the (bourgeois) society within the Spectator periodicals.</p>
<p>The terms occurring in <bold>topic 9</bold> describe virtuous behavior and practices. The gender-specific key concept of virtue (cf. <xref ref-type="bibr" rid="B45">Pabst 2007</xref>) stands at the very beginning of the word sequence. The following terms refer to the fact that virtue leads to (individual and collective) &#8216;happiness&#8217; (<italic>bonheur</italic>). In general, 18<sup>th</sup> century philosophers equate &#8216;virtue&#8217; with &#8216;happiness&#8217;, for only those who lead a virtuous life can contribute to their own happiness and to the happiness of the community. Virtue is thus seen as a means to achieve the individual and collective goal of happiness (cf. <xref ref-type="bibr" rid="B58">V&#246;lkl 2022, 121&#8211;122</xref>).<xref ref-type="fn" rid="n13">13</xref></p>
<p>Moreover, as explained at the outset of this article, the concept of virtue becomes a gender-specific key concept in the course of the 18<sup>th</sup> century insofar as not only different virtues emerge for women and men, but also considering that women have to follow a larger amount of virtuous (i.e., morally acceptable) behavioral norms than men. On the discursive level, this appears through a quantitatively higher number of terms attributed to &#8216;female&#8217; behavior than those attributed to &#8216;male&#8217; behavior. The reason for this is that in the 18<sup>th</sup> century, the virtuousness of the whole (bourgeois) society becomes more and more associated with the virtuous behavior of women. In the Spectator press, precisely this virtue is constantly promoted (cf. <xref ref-type="bibr" rid="B58">V&#246;lkl 2022, 120-124</xref>).</p>
<p>Topic 9 can be found in all Spectator periodicals at a median rate of 0.62 (<xref ref-type="fig" rid="F5">Figure 5</xref>) which makes it the most probable in the corpus. This is not surprising because most periodicals explicitly state their goal in their first lines which is to turn all people into useful members of the society &#8211; a society which is becoming increasingly complex and integrated into a nation (cf. <xref ref-type="bibr" rid="B16">Ertler 2010, 100</xref>). In terms of women as useful members of society, the role of the (bourgeois) woman is conceptualized in three ways: as spouse, housewife, and mother. Outside the domestic sphere, she has no right to exist, which is the reason why, for example, the image of the learned woman was defamed in the Spectator periodicals at the beginning of the 18<sup>th</sup> century and has subsequently been omitted altogether &#8211; according to the motto &#8216;out of sight, out of mind&#8217; (cf. <xref ref-type="bibr" rid="B58">V&#246;lkl 2022, 309&#8211;310</xref>).</p>
<fig id="F5">
<label>Figure 5</label>
<caption>
<p>Distribution of topic 9 in French-language periodicals.</p>
</caption>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="jcls-108_volkl-g5.png"/>
</fig>
<p><bold>Topic 18</bold> results in terms referring to the virtuous ideal image of both women and men. The terms &#8216;tender&#8217; (<italic>doux, tendre</italic>), &#8216;amiable&#8217; (<italic>aimable</italic>), &#8216;grace&#8217; (<italic>charme</italic>), &#8216;prudent&#8217; (<italic>sage</italic>), &#8216;witty&#8217; (<italic>brillant</italic>) here refer to inner virtues while the terms &#8216;beautiful&#8217; (<italic>bel</italic>) and &#8216;beauty&#8217; (<italic>beaut&#233;</italic>) can refer to inner and outer virtues at the same time as explained above. Although this is not a frequent topic, it is consistently present in all Spectator periodicals.</p>
<p><bold>Topic 21</bold> exhibits terms that can be assigned to the discourse field of love. They are associated positively or negatively with love. For example, next to the approbatives &#8216;to love&#8217; (<italic>aimer</italic>), &#8216;happiness&#8217; (<italic>bonheur</italic>), or &#8216;tender&#8217; (<italic>tendre</italic>), one can find the pejoratives such as &#8216;unhappy&#8217; (<italic>malheureux</italic>), &#8216;pain&#8217; (<italic>douleur</italic>), or &#8216;to lose&#8217; (<italic>perdre</italic>). The frequency of individual terms will be discussed below.</p>
<p>A look at the distribution of topic 21 within the French-language periodicals (<xref ref-type="fig" rid="F6">Figure 6</xref>) reveals that the three successive periodicals <italic>Le Nouveau Spectateur</italic> (1758&#8211;60), <italic>Le Monde comme il est</italic> (1760), and <italic>Le Monde</italic> (1760&#8211;61) of Jean-Fran&#231;ois de Bastide (1724&#8211;1788) are particularly endowed with this topic. The literary and cultural studies research carried out by Fischer-Pernkopf et al. and V&#246;lkl support the finding that Bastide continuously narrates exemplary stories of happy and unhappy (heterosexual) lovers (cf. <xref ref-type="bibr" rid="B20">Fischer-Pernkopf et al. 2018</xref>, <xref ref-type="bibr" rid="B58">V&#246;lkl 2022</xref>).</p>
<fig id="F6">
<label>Figure 6</label>
<caption>
<p>Distribution of topic 21 in French-language periodicals.</p>
</caption>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="jcls-108_volkl-g6.png"/>
</fig>
<p>The proximity of the nodes and the edges weight in the network analysis graph in <xref ref-type="fig" rid="F7">Figure 7</xref> also illustrates the prevalence of topic 21 in all of Bastide&#8217;s periodicals. It further shows that <bold>topics 6<xref ref-type="fn" rid="n14">14</xref></bold> and <bold>23<xref ref-type="fn" rid="n15">15</xref></bold> are also very common in Bastide&#8217;s periodicals. They include typical narrative vocabulary (e.g., <italic>demander</italic> &#8211; &#8216;to ask&#8217;, <italic>r&#233;pondre</italic> &#8211; &#8216;to answer&#8217;, <italic>entrer</italic> &#8211; &#8216;to enter&#8217;, <italic>entendre</italic> &#8211; &#8216;to listen&#8217;, <italic>lire</italic> &#8211; &#8216;to read&#8217;) and typical narrative elements (e.g., <italic>ami</italic> &#8211; &#8216;friend&#8217;, <italic>maison</italic> &#8211; &#8216;house&#8217;, <italic>chambre</italic> &#8211; &#8216;room&#8217;, <italic>porte</italic> &#8211; &#8216;door&#8217;). Based on the accumulation of narrative terms, it can be concluded that in Bastide&#8217;s periodicals, the discourse of love is primarily conveyed through stories and storytelling. This interpretation of the topic modeling results is supported by previous literary and cultural research in this field which also stresses the strong narrative design of Bastide&#8217;s periodicals (cf. <xref ref-type="bibr" rid="B20">Fischer-Pernkopf et al. 2018</xref>, <xref ref-type="bibr" rid="B42">Mussner 2016</xref>, <xref ref-type="bibr" rid="B58">V&#246;lkl 2022</xref>).</p>
<fig id="F7">
<label>Figure 7</label>
<caption>
<p>Detail from the network graph of periodicals and topics, showing the prevalence of topic 21 in the three periodicals of Jean-Fran&#231;ois de Bastide.</p>
</caption>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="jcls-108_volkl-g7.png"/>
</fig>
<p>Additionally, the analysis of the issues manually annotated with the subjects/keywords &#8216;Image of women&#8217; and &#8216;Image of men&#8217;<xref ref-type="fn" rid="n16">16</xref> of the <italic>Nouveau Spectateur<xref ref-type="fn" rid="n17">17</xref></italic> and the <italic>Monde comme il est</italic>&#160;<xref ref-type="fn" rid="n18">18</xref> identified that the following five narrative forms (<italic>Erz&#228;hlformen</italic>) are predominately used to discuss family life (in particular education) and couple relationships (with a focus on the romantic tender love relationship): general account (<italic>allgemeine Erz&#228;hlung, AE</italic>), heteroportrait (<italic>Fremdportr&#228;t, FP</italic>), metatextuality (<italic>metatextueller Kommentar, MT</italic>), dialogue (<italic>Dialog, D</italic>) and letter/letter to the editor (<italic>Leser*innenbriefe, LB</italic>) (cf. <xref ref-type="bibr" rid="B58">V&#246;lkl 2022, 209</xref>). Concerning the distribution and arrangement of these text types, it has to be emphasized that they also repeatedly appear intertwined within each other which leads to the &#8211; for the Spectator periodicals &#8211; typical multi-layered system of communication (cf. <xref ref-type="bibr" rid="B19">Fischer 2014, 74&#8211;83</xref>).</p>
<p><xref ref-type="fig" rid="F8">Figure 8</xref>, which shows a statistical examination of all issues of Bastide&#8217;s periodicals, further supports the above-mentioned results. It shows that Bastide uses the following narrative forms as predominant communication strategy: metatextual commentaries (MT), letters/letters to the editor (LB), dialogues (D), and general accounts (AE). While the heteroportrait (FD) only stands on fifth position after citation/motto (ZM).</p>
<fig id="F8">
<label>Figure 8</label>
<caption>
<p>Narrative forms in Bastide&#8217;s periodicals.</p>
</caption>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="jcls-108_volkl-g8.png"/>
</fig>
<p>Furthermore, the three bar charts of the topic distribution in Bastide&#8217;s periodicals (<xref ref-type="fig" rid="F9">Figure 9</xref>, <xref ref-type="fig" rid="F10">Figure 10</xref>, and <xref ref-type="fig" rid="F11">Figure 11</xref>) indicate a wide distribution of topic 9 (virtuous behavior and action) and topic 6 (describing the postulate of enlightened philosophers: &#8216;(Self)reflection&#8217; (<italic>r&#233;flexion</italic>) leading to &#8216;truth&#8217; (<italic>v&#233;rit&#233;</italic>) and knowledge). This focus on virtue and vice is not surprising at all, considering that the Spectator periodicals aim at the moral education of their female and male audience. The readers of the periodicals in general and of Bastide&#8217;s periodicals in particular are repeatedly exposed to vicious behavior and actions by means of shorter and longer stories in order to guide them to virtuous behavior and actions. A detailed definition or specification of the social norm designated by the term &#8216;virtue&#8217;, however, is lacking and thus remains undefined; rather, &#8216;being virtuous&#8217; is illustrated indirectly through the depiction of its opposite: &#8216;being vicious&#8217;. Via the detour of numerous love and relationship stories as well as character portraits which clearly highlight vicious behavior and vicious character traits, the readers are thus led to the desired social norm (cf. <xref ref-type="bibr" rid="B58">V&#246;lkl 2022, 291&#8211;292</xref>).</p>
<fig id="F9">
<label>Figure 9</label>
<caption>
<p>Topics in <italic>Le Monde</italic>.</p>
</caption>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="jcls-108_volkl-g9.png"/>
</fig>
<fig id="F10">
<label>Figure 10</label>
<caption>
<p>Topics in <italic>Le Monde comme il est</italic>.</p>
</caption>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="jcls-108_volkl-g10.png"/>
</fig>
<fig id="F11">
<label>Figure 11</label>
<caption>
<p>Topics in <italic>Le Nouveau Spectateur</italic>.</p>
</caption>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="jcls-108_volkl-g11.png"/>
</fig>
</sec>
<sec id="S6">
<title>6. Topic Modeling in the Spanish-language Spectator Periodicals</title>
<p>Regarding gender-specific topics, the topic modeling results for the Spanish-language periodicals were similar to those of the French-language Spectators. Within the 18 Spanish topics, topics 8, 9, 11 and 17 can be identified as referring to women and men (see <xref ref-type="table" rid="T4">Table 4</xref>).</p>
<table-wrap id="T4">
<label>Table 4</label>
<caption>
<p>Gender-specific topics in Spanish-language Spectators.</p>
</caption>
<table>
<tr>
<td colspan="4"><hr/></td>
</tr>
<tr>
<td align="left" valign="top">Topic 8</td>
<td align="left" valign="top">Topic 9</td>
<td align="left" valign="top">Topic 11</td>
<td align="left" valign="top">Topic 17</td>
</tr>
<tr>
<td colspan="4"><hr/></td>
</tr>
<tr>
<td align="left" valign="top">virtud</td>
<td align="left" valign="top">espa&#241;a</td>
<td align="left" valign="top">hijo</td>
<td align="left" valign="top">mujer</td>
</tr>
<tr>
<td align="left" valign="top">vida</td>
<td align="left" valign="top">siglo</td>
<td align="left" valign="top">mujer</td>
<td align="left" valign="top">moda</td>
</tr>
<tr>
<td align="left" valign="top">amor</td>
<td align="left" valign="top">lengua</td>
<td align="left" valign="top">padre</td>
<td align="left" valign="top">dama</td>
</tr>
<tr>
<td align="left" valign="top">corazon</td>
<td align="left" valign="top">ciencia</td>
<td align="left" valign="top">madre</td>
<td align="left" valign="top">gustar</td>
</tr>
<tr>
<td align="left" valign="top">honor</td>
<td align="left" valign="top">escribir</td>
<td align="left" valign="top">criar</td>
<td align="left" valign="top">hermoso</td>
</tr>
<tr>
<td align="left" valign="top">vivir</td>
<td align="left" valign="top">nacion</td>
<td align="left" valign="top">maridar</td>
<td align="left" valign="top">adornar</td>
</tr>
<tr>
<td align="left" valign="top">placer</td>
<td align="left" valign="top">mundo</td>
<td align="left" valign="top">ni&#241;o</td>
<td align="left" valign="top">hermosura</td>
</tr>
<tr>
<td align="left" valign="top">alma</td>
<td align="left" valign="top">estudiar</td>
<td align="left" valign="top">familia</td>
<td align="left" valign="top">personar</td>
</tr>
<tr>
<td align="left" valign="top">amar</td>
<td align="left" valign="top">historia</td>
<td align="left" valign="top">amor</td>
<td align="left" valign="top">gracia</td>
</tr>
<tr>
<td align="left" valign="top">mirar</td>
<td align="left" valign="top">leer</td>
<td align="left" valign="top">amar</td>
<td align="left" valign="top">sexo</td>
</tr>
<tr>
<td align="left" valign="top">mundo</td>
<td align="left" valign="top">libro</td>
<td align="left" valign="top">edad</td>
<td align="left" valign="top">figurar</td>
</tr>
<tr>
<td align="left" valign="top">viciar</td>
<td align="left" valign="top">arte</td>
<td align="left" valign="top">tratar</td>
<td align="left" valign="top">se&#241;ora</td>
</tr>
<tr>
<td align="left" valign="top">honrar</td>
<td align="left" valign="top">letra</td>
<td align="left" valign="top">a&#241;o</td>
<td align="left" valign="top">bayle</td>
</tr>
<tr>
<td align="left" valign="top">noble</td>
<td align="left" valign="top">sabio</td>
<td align="left" valign="top">marido</td>
<td align="left" valign="top">cortejo</td>
</tr>
<tr>
<td align="left" valign="top">despreciar</td>
<td align="left" valign="top">idioma</td>
<td align="left" valign="top">cuidar</td>
<td align="left" valign="top">mirar</td>
</tr>
<tr>
<td align="left" valign="top">felicidad</td>
<td align="left" valign="top">espa&#241;oles</td>
<td align="left" valign="top">matrimoniar</td>
<td align="left" valign="top">cabeza</td>
</tr>
<tr>
<td align="left" valign="top">desear</td>
<td align="left" valign="top">ciencias</td>
<td align="left" valign="top">se&#241;ora</td>
<td align="left" valign="top">naturaleza</td>
</tr>
<tr>
<td align="left" valign="top">efecto</td>
<td align="left" valign="top">llamar</td>
<td align="left" valign="top">esposo</td>
<td align="left" valign="top">rostro</td>
</tr>
<tr>
<td align="left" valign="top">naturaleza</td>
<td align="left" valign="top">antiguo</td>
<td align="left" valign="top">hermano</td>
<td align="left" valign="top">arte</td>
</tr>
<tr>
<td align="left" valign="top">ojo</td>
<td align="left" valign="top">naciones</td>
<td align="left" valign="top">cari&#241;o</td>
<td align="left" valign="top">visitar</td>
</tr>
<tr>
<td colspan="4"><hr/></td>
</tr>
</table>
</table-wrap>
<p><bold>Topic 8</bold> is headed by the gender-specific key concept of &#8216;virtue&#8217; (<italic>virtud</italic>) followed by terms describing elements of a virtuous lifestyle (<italic>vida</italic> &#8211; &#8216;life&#8217;, <italic>amor/amar</italic> &#8211; &#8216;(to) love&#8217;, <italic>honor/honrar</italic> &#8211; &#8216;(to) honor&#8217;), thereby showing considerable similarities to topic 9 of the French-language periodicals.<xref ref-type="fn" rid="n19">19</xref> This topic similarity is not surprising at all since the contemporary gender discourse within the French-language periodicals enters the Spanish periodicals &#8211; that first appear in Spain from mid-century onward &#8211; through numerous translations, imitations, and cultural adaptations. Here, as in France, women and their behavior are held more responsible for the virtuousness of the whole (bourgeois) society than men. More than in other European countries, however, women in Spain are excluded from public life and confined to the private sphere which centers on home, family, and motherhood (cf. <xref ref-type="bibr" rid="B58">V&#246;lkl 2022, 229</xref>).<xref ref-type="fn" rid="n20">20</xref> An abundant presence of topic 8 in all Spanish periodicals is thus an expectable development (<xref ref-type="fig" rid="F12">Figure 12</xref>).</p>
<fig id="F12">
<label>Figure 12</label>
<caption>
<p>Distribution of topic 8 in Spanish-language periodicals.</p>
</caption>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="jcls-108_volkl-g12.png"/>
</fig>
<p><bold>Topic 9</bold>, with terms such as &#8216;writing&#8217; (<italic>escribir</italic>), &#8216;studying&#8217; (<italic>estudiar</italic>), and &#8216;reading&#8217; (<italic>leer</italic>), alludes to educational activities; the terms &#8216;science&#8217; (<italic>ciencia</italic>), &#8216;art&#8217; (<italic>arte</italic>), or &#8216;history&#8217; (<italic>historia</italic>) of &#8216;ancient&#8217; (<italic>antiguo</italic>) time to specific study objects. The convergence of these terms suggests that this topic describes the education of a bourgeois man, even though no term referring to a male subject (such as <italic>hombre</italic>) &#8211; nor to a female subject (such as <italic>mujer</italic>) &#8211; can be found. In fact, although the Spanish periodicals grant the female gender a certain capacity for education as well, the terms of topic 9 refer to male formation only. Education for young women is conceived differently to education for young men because (as the French-language periodicals) the Spanish Spectators also propagate a complementary gender model, implying that women and men need to be educated specifically for the correct fulfillment of their gender-specific role in society. The Spanish ideal of the virtuous (bourgeois) woman is also praised in her threefold role as spouse, housewife, and mother, through the fulfillment of which she contributes to the common good of society. This image of woman is conceived in a &#8216;natural complementarity&#8217; to man, whose ideal image is embodied by the &#8216;hombre de bien&#8217;. The latter is characterized by the training of his intellect and subsequently proving useful for his fatherland and the common good. The &#8216;hombre de bien&#8217; of the 18<sup>th</sup> century is thus not to be confused with the preceding aristocratic &#8216;hombre de bien&#8217; of the 17<sup>th</sup> century, whose idleness causes his reputation to fall below that of an active citizen &#8211; regardless of his status (cf. <xref ref-type="bibr" rid="B25">He&#223;e 2008, 113&#8211;130</xref>).</p>
<p>Very similar to topic 4 in the French Spectators, the content of <bold>topic 11</bold> supports the construction of the heteronormative society, suggesting the ideal role of &#8216;woman&#8217; (<italic>mujer</italic>) and man in &#8216;marriage&#8217; (<italic>maridar, matrimoniar</italic>) where they become &#8216;mother&#8217; (<italic>madre</italic>) and &#8216;father&#8217; (<italic>padre</italic>) of &#8216;children&#8217; (<italic>hijo, ni&#241;o</italic>). The role of the woman is thus conceived by her &#8216;husband&#8217;s&#8217; (<italic>marido, esposo</italic>) side to whom she is supposed to be a good spouse and housewife. Within the domestic sphere (<italic>familia</italic>), she also receives the role of the &#8216;caring&#8217; (<italic>cuidar, cari&#241;o</italic>) mother who &#8216;loves&#8217; (<italic>amor, amar</italic>) and &#8216;raises&#8217; (<italic>criar</italic>) her &#8216;children&#8217; (<italic>hijo, ni&#241;o</italic>). Although a rather infrequent topic (<xref ref-type="fig" rid="F13">Figure 13</xref>), it exists throughout all Spanish Spectators.</p>
<fig id="F13">
<label>Figure 13</label>
<caption>
<p>Distribution of topic 11 in Spanish-language periodicals.</p>
</caption>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="jcls-108_volkl-g13.png"/>
</fig>
<p><bold>Topic 17</bold>, represented in <xref ref-type="fig" rid="F14">Figure 14</xref>, points to two discourses associated with the female gender: on the one hand the subject of beauty, on the other hand the then vicious trend of having a relationship with a younger man (<italic>cortejo</italic>). The first eight terms of this topic (<italic>mujer</italic> &#8211; &#8216;woman&#8217;, <italic>moda</italic> &#8211; &#8216;fashion&#8217;, <italic>dama</italic> &#8211; &#8216;lady&#8217;, <italic>gustar</italic> &#8211; &#8216;to please&#8217;, <italic>hermoso</italic> &#8211; &#8216;beautiful&#8217;, <italic>adornar</italic> &#8211; &#8216;to adorn&#8217;, <italic>hermosura</italic> &#8211; &#8216;beauty&#8217;) refer to the semantic field of beauty which pervades the spectatorial gender discourses throughout the century and clearly reflects topic 24 of the French-language periodicals (see <xref ref-type="fig" rid="F4">Figure 4</xref>). In fact, the Spectator periodicals by and large constantly instruct their readers to cultivate external and, increasingly, internal beauty because female beauty is perceived as a pledge for marriage (cf. <xref ref-type="bibr" rid="B48">Schaufler 2002, 190</xref>) which is seen as the &#8216;natural&#8217; destiny of the (bourgeois) woman and is thus considered her ultimate goal. At the same time, however, the periodicals warn against falling prey to a cult of beauty that goes hand in hand with the vices of vanity and jealousy. One of these vices, also represented in topic 17, is the gender-stereotypical model of the <italic>cortejo</italic>, i.e., a (younger) man maintaining a very close relationship with a married woman or widow who he &#8216;visits&#8217; (<italic>visitar</italic>) regularly.<xref ref-type="fn" rid="n21">21</xref> While this (mostly platonic) form of relationship is not a moral offense in aristocratic tradition, it is criticized and stigmatized in the Spectator periodicals.</p>
<fig id="F14">
<label>Figure 14</label>
<caption>
<p>Distribution of topic 17 in Spanish-language periodicals.</p>
</caption>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="jcls-108_volkl-g14.png"/>
</fig>
<p>Regarding the dissemination of the gender-related topics, the Spanish periodicals pursue a similar strategy to their French-language precursors. Likewise, in the Spanish Spectators virtuous and vicious gender-specific values, norms and practices are mostly conveyed through stories and storytelling. Similar to the French topics 6 and 23, the topic modeling process for the Spanish Spectators revealed topics with a high concentration of narrative vocabulary, such as in topic 2.<xref ref-type="fn" rid="n22">22</xref> Therein, narrative vocabulary revolves around the semantic fields of movement (e.g., <italic>venir</italic> &#8211; &#8216;to come&#8217;, <italic>salir</italic> &#8211; &#8216;to leave&#8217;, <italic>llegar</italic> &#8211; &#8216;to arrive&#8217;), speech (e.g., <italic>contar</italic> &#8211; &#8216;to narrate&#8217;, <italic>entender</italic> &#8211; &#8216;to listen&#8217;, <italic>palabra</italic> &#8211; &#8216;word&#8217;), and time (e.g., <italic>a&#241;o</italic> &#8211; &#8216;year&#8217;, <italic>hora</italic> &#8211; &#8216;hour&#8217;, <italic>noche</italic> &#8211; &#8216;night&#8217;), all of which are important components in a story. As can be discerned in <xref ref-type="fig" rid="F15">Figure 15</xref>, topic 2 occurs in all Spanish periodicals.</p>
<fig id="F15">
<label>Figure 15</label>
<caption>
<p>Distribution of topic 2 in Spanish-language periodicals.</p>
</caption>
<graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="jcls-108_volkl-g15.png"/>
</fig>
</sec>
<sec id="S7">
<title>7. Conclusion</title>
<p>With their gender-specific discourses, the Spectator press (co-)constructed, preserved, and propagated a bourgeois gender model which is still valid in socio-cultural perception today. This contribution investigates 1,658 French- and 690 Spanish-language issues which were analyzed with topic modeling using LDA. The findings with a focus on gender-specific discourse match and reinforce the results from V&#246;lkl&#8217;s study on narrative and media-specific gender construction within the Spectator periodicals (<xref ref-type="bibr" rid="B58">V&#246;lkl 2022</xref>).</p>
<p>Using topic modeling, gender-specific topics were identified in the Spectator corpus. Additionally, the application of topic modeling also showed that the Spectator press employed a certain narrative vocabulary (French Spectators: topic 6 and 23; Spanish Spectators: topic 2). Moreover, the comparison between the French- and Spanish-language periodicals rendered similar results: The Spectator corpus of both languages manifested several topics pertaining to a gender-specific discourse. This discourse can be discerned explicitly in topics which exhibit terms referring to female or male stereotypical models, or implicitly in topics which exhibit terms referring to virtuous and vicious gender-specific values, norms, and practices. The study also showed that within the gender-specific topics more terms refer to &#8216;female&#8217; than to &#8216;male&#8217; (character) traits and behavioral patterns. This result is not surprising neither since the Spectator press supports the creation of a (bourgeois) gender model, based on the complementarity of women and men, and in which the &#8216;female&#8217; virtues are assigned a much greater importance for the stability and moral functioning of the society (cf. <xref ref-type="bibr" rid="B45">Pabst 2007</xref>; <xref ref-type="bibr" rid="B55">Steinbr&#252;gge 1987</xref>; <xref ref-type="bibr" rid="B58">V&#246;lkl 2022, 120-124</xref>). These concordances ascertain that topic modeling as the method used for the present analysis can be successfully employed to question and confirm hypotheses gained through close reading.</p>
<p>In addition to our findings on the gender-specific topics in the Spectators, we described the topic modeling workflow used in DiSpecs in <xref ref-type="sec" rid="S4">section 4</xref>. We aim to make our analysis process transparent for other researchers interested in this method. The research community can also benefit from the primary data available in TEI, and the code which all are publicly available online (<xref ref-type="bibr" rid="B17">Ertler et al. 2011&#8211;2021</xref>, <xref ref-type="bibr" rid="B52">Scholger et al. 2019&#8211;2021</xref>, <xref ref-type="bibr" rid="B51">Scholger et al. 2022</xref>).</p>
<p>The DARIAH-DE Notebooks that implement LDA topic modeling proved to be very useful as a basis in our analysis workflow. With some adaptations and additional pre-processing (especially segmentation and lemmatization) and post-processing steps (e.g., results categorization and additional visualizations), we were able to produce comprehensible and insightful results. Our own experience and the comparison with other topic modeling projects allow us to conclude that pre-processing is a crucial part of the analysis since it strongly impacts the quality of the results. The decisions on the respective steps depend on the research material and the specific project goals.</p>
<p>An advantage of topic modeling is the possibility to analyze more content than with close reading, to illustrate the hypothesis on a broader level than through individual examples, and to present the findings using different types of visualizations. Our topic modeling analysis resulted in measurable data of a large text collection&#8217;s semantic structure which we were able to interpret and comprehensively demonstrate to the Spectators research community. Furthermore, the analysis invoked some new insights into the corpus. Concerning the gender-specific discourse in the Spectators, we saw, e.g., with topic 22 that the French translation of the <italic>Female Spectator</italic> is equipped with a specific semantic vocabulary that can almost exclusively be found in this specific periodical. This result can be attributed to the fact that in this case, we are dealing with a translation and not with a genuine French periodical.</p>
<p>The primary data and the digital scholarly edition also benefit from the topic modeling analysis. With the resulting topics, it is now possible to revise the keywords manually assigned to the individual issues and to further differentiate them. The list of 37 keywords was determined at the beginning of the digital edition project around 2011 and was only minimally expanded in the course of the project. Consequently, the list seems somewhat arbitrary: culture- and language-specific topics &#8211; such as &#8216;Apologetic of Spain&#8217; which only apply to a few issues &#8211; are on the same level as very broad topics such as &#8216;Theatre, Literature, Arts&#8217; which combine three areas in one topic. Therefore, the results from topic modeling help to expand and adjust the list of keywords for thematic indexing, thus improving the analysis capabilities within the digital edition as demonstrated in <italic>LdoD Visual</italic> by <xref ref-type="bibr" rid="B46">Portela and Rito Silva 2017</xref>. The identified terms in the topics can be incorporated into the TEI metadata header and subsequently used for a more precise and sophisticated search not only at the document level but also on specific text fragments.</p>
<p>Nevertheless, it is necessary to mention certain challenges in using topic modeling. Critics like Dobson argue that the variability of the output depending on the algorithms and set parameters of the method is problematic (cf. <xref ref-type="bibr" rid="B14">Dobson 2021, &#167;20</xref>), while Roe, Gladstone, and Morrisey also refer to the probabilistic nature of LDA causing variability in individual runs even with the same parameters (cf. <xref ref-type="bibr" rid="B47">Roe et al. 2016, 4</xref>). While we did not compare our LDA results with other algorithms, we agree with Sch&#246;ch that these variations manifest themselves &#8220;in the details of word ranks rather than in the general topics obtained&#8221; (<xref ref-type="bibr" rid="B50">Sch&#246;ch 2017</xref>). Parameters have to be tested for individual projects, but once optimized, the method provides relatively stable results.</p>
<p>Furthermore, Murakami et al. as well as Shadrova are skeptical towards methods based on the bag-of-words approach because it ignores the grammatical structures and semantic relations between words (cf. <xref ref-type="bibr" rid="B41">Murakami et al. 2017, 246</xref>, <xref ref-type="bibr" rid="B53">Shadrova 2021, 13-14</xref>). While we do agree with this statement and believe that every scientific method should be questioned, we also argue that digital methods are not supposed to take on our tasks as humanities experts, but to facilitate research and help us to interpret our data. For these reasons, using a combined approach of topic modeling (and text mining methods in general) and close reading is essential as well as the understanding of the material itself. As Fechner and Wei&#223; point out, it is not the topics that answer research questions themselves, but the researchers through the interpretations of the topics (cf. <xref ref-type="bibr" rid="B18">Fechner and Wei&#223; 2017, 20</xref>).</p>
<p>Besides the contribution to the current state of Spectators research and to practical applications of topic modeling, our work also lays the foundations for future work on 18<sup>th</sup> century literature. The presented results can be compared with similar research on other genres of that time. In addition to the probabilistic topic modeling approach, we intend to integrate transformer-based models to investigate a new corpus of Spanish epistolary novels which are considered to have continued propagating gender-specific values, norms, and practices from the Spectators while also representing an intermediate step towards the 19<sup>th</sup> century novel.</p>
</sec>
<sec id="S8">
<title>8. Data Availability</title>
<p>Data can be found here: <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://gams.uni-graz.at/dispecs">https://gams.uni-graz.at/dispecs</ext-link> and <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://gams.uni-graz.at/spectators">https://gams.uni-graz.at/spectators</ext-link></p>
</sec>
<sec id="S9">
<title>9. Software Availability</title>
<p>Software can be found here: <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://github.com/distantspectators">https://github.com/distantspectators</ext-link></p>
</sec>
</body>
<back>
<sec id="S10">
<title>10. Acknowledgements</title>
<p>We thank the Austrian Academy of Sciences who funded the project DiSpecs, as well as our project team co-members Bernhard Geiger, Christina Glatz, Elisabeth Hobisch, and Philipp Koncar for their cooperation, contribution, and support.</p>
</sec>
<sec id="S11">
<title>11. Author Contributions</title>
<p><bold>Yvonne V&#246;lkl:</bold> Results Interpretation, Conceptualization, Writing, Editing</p>
<p><bold>Sanja Sari&#263;:</bold> Topic Modeling, Conceptualization, Writing, Editing</p>
<p><bold>Martina Scholger:</bold> Project Coordination, Conceptualization, Writing, Editing</p>
</sec>
<fn-group>
<fn id="n1"><p>In the course of the 18<sup>th</sup> century, the perception of female and male bodies and their genitalia changed. The so far gradually assumed difference between women and men is increasingly interpreted qualitatively and a complementary understanding of two genders can assert itself (cf. <xref ref-type="bibr" rid="B35">Laqueur 2003[1990]</xref>). The new perception of women and men also leads to a cultural redefinition of their gender relations (e.g., woman as the &#8216;moral gender&#8217;, cf. <xref ref-type="bibr" rid="B55">Steinbr&#252;egge 1987</xref>) and to a major shift in the conception of virtue which was originally connotated to the meritorious properties and qualities of men (Latin: <italic>vir</italic>) and was feminized as of the end of the 17<sup>th</sup> century only (cf. <xref ref-type="bibr" rid="B45">Pabst 2007, 25ff.</xref>).</p></fn>
<fn id="n2"><p>For a description of the Spanish Spectator periodicals and an in-depth analysis of the use and function of the letter as mode of communication with the public, see <xref ref-type="bibr" rid="B26">Hobisch 2017</xref>.</p></fn>
<fn id="n3"><p>The corpus contains periodicals in English, French, German, Italian, Portuguese, and Spanish, but due to the rather small corpus size of English, German, and Portuguese, these languages were not considered in the topic modeling analysis. Therefore, we analyzed 1,658 French-, 1,344 Italian-, and 690 Spanish-language issues.</p></fn>
<fn id="n4"><p>The Spectator periodicals stand out for their multi-layered system of communication consisting of various narrative levels of representation which are embedded in various narrative forms. Further, narrative forms are also intertwined within each other when, e.g., the fictitious editor includes a supposedly authentic reader&#8217;s letter in the periodical which, in turn, narrates a story about a woman who then enters into a dialogue with another woman about a letter to a man (cf. <xref ref-type="bibr" rid="B19">Fischer 2014</xref>, 81&#8211;83). This epistolary correspondence between editor and readers has been considered a major element for the success of the Spectator periodicals in the course of the 18<sup>th</sup> century (cf. <xref ref-type="bibr" rid="B27">Hobisch 2018</xref>).</p></fn>
<fn id="n5"><p>The Jupyter Notebooks with the Python code are provided via GitHub: <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://github.com/distantspectators/DiSpecs">https://github.com/distantspectators/DiSpecs</ext-link>.</p></fn>
<fn id="n6"><p>Besides functional words, such as the Spanish <italic>inmediatamente</italic> (immediately) or <italic>entonces</italic> (therefore), or some frequently used adjectives and modal verbs like the French <italic>grand</italic> (great) and <italic>devoit</italic> (should), we also excluded some nouns from the analysis, e.g., the Spanish <italic>n&#250;mero</italic> as it is often used as an issue title.</p></fn>
<fn id="n7"><p>Sch&#246;ch gave a more detailed reflection on hyperparameter optimization in his scientific blog (cf. <xref ref-type="bibr" rid="B49">Sch&#246;ch 2016</xref>).</p></fn>
<fn id="n8"><p>The heat map, bar charts, and word clouds of the French periodicals can be viewed under the URL <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://gams.uni-graz.at/o:dispecs.result.tm.fr">https://gams.uni-graz.at/o:dispecs.result.tm.fr</ext-link>. Spanish results are accessible under the URL <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://gams.uni-graz.at/o:dispecs.result.tm.es">https://gams.uni-graz.at/o:dispecs.result.tm.es</ext-link>.</p></fn>
<fn id="n9"><p>Line diagrams: <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://gams.uni-graz.at/archive/objects/o:dispecs.result.tm.fr/methods/sdef:TEI/get?mode=diachronic">https://gams.uni-graz.at/archive/objects/o:dispecs.result.tm.fr/methods/sdef:TEI/get?mode=diachronic</ext-link> and <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://gams.uni-graz.at/archive/objects/o:dispecs.result.tm.es/methods/sdef:TEI/get?mode=diachronic">https://gams.uni-graz.at/archive/objects/o:dispecs.result.tm.es/methods/sdef:TEI/get?mode=diachronic</ext-link>.</p></fn>
<fn id="n10"><p>The full visualizations can be viewed on our web page: <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://gams.uni-graz.at/archive/objects/o:dispecs.result.tm.fr/methods/sdef:TEI/get?mode=topic-network">https://gams.uni-graz.at/archive/objects/o:dispecs.result.tm.fr/methods/sdef:TEI/get?mode=topic-network</ext-link> and <ext-link ext-link-type="uri" xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="https://gams.uni-graz.at/archive/objects/o:dispecs.result.tm.es/methods/sdef:TEI/get?mode=topic-network">https://gams.uni-graz.at/archive/objects/o:dispecs.result.tm.es/methods/sdef:TEI/get?mode=topic-network</ext-link>.</p></fn>
<fn id="n11"><p>Regarding gender discourse in 18<sup>th</sup> century France, see the articles of G. <xref ref-type="bibr" rid="B4">Bock and Zimmermann 1997</xref>, <xref ref-type="bibr" rid="B8">Brink 2008</xref>, <xref ref-type="bibr" rid="B28">Honegger 2011</xref>, or <xref ref-type="bibr" rid="B54">Sieuzac 2009</xref>. As to the presence of women in society and literature, see the essay collection edited by <xref ref-type="bibr" rid="B31">Jacobs et al. 1979</xref>. Concerning the theoretical and literary discourse on the woman as the &#8216;moral gender&#8217; in the 18<sup>th</sup> century, see Steinbr&#252;gge&#8217;s monograph (<xref ref-type="bibr" rid="B55">Steinbr&#252;egge 1987</xref>). As to the representation of women in the French Enlightenment press, see <xref ref-type="bibr" rid="B13">Dijk 1988</xref> and to the history of the &#8216;presse f&#233;minine&#8217; in France, see <xref ref-type="bibr" rid="B56">Sullerot 1966</xref>.</p></fn>
<fn id="n12"><p>The French term <italic>beau sexe</italic> for the female part of the population is a compound and has been separated during the topic modeling process. Further, the term <italic>beau</italic> has been lemmatized into <italic>bel</italic>. This is the reason why the terms <italic>bel</italic> and <italic>sexe</italic> appear separately in topic 24. Nonetheless, their immediate position next to each other indicates their connection.</p></fn>
<fn id="n13"><p>On the discourse of happiness in the 18<sup>th</sup> century, cf. <xref ref-type="bibr" rid="B39">Mauzi 1969</xref>, on the concept of &#8216;happiness&#8217; in <italic>The Spectator</italic>, cf. <xref ref-type="bibr" rid="B44">Norton 2015</xref>.</p></fn>
<fn id="n14"><p>Topic 6: penser v&#233;rit&#233; mal vrai honneur ami r&#233;flexion juger mauvais d&#233;faut r&#233;pondre caractere sentir droit convenir lorsqu conno&#238;tre entendre lire quelquefois</p></fn>
<fn id="n15"><p>Topic 23: maison demander heure chambre para&#238;tre jeune tem main peine passer entrer revenir air ami vouloit porte arriver alloit entendre sortir</p></fn>
<fn id="n16"><p>In total, the list of subjects comprises 37 keywords which were determined at the beginning of the digital scholarly edition project (cf. <xref ref-type="bibr" rid="B17">Ertler et al. 2011&#8211;2021</xref>) and which was slightly expanded in the course of the project.</p></fn>
<fn id="n17"><p>From the 108 issues within the <italic>Nouveau Spectateur</italic> 58 issues (44%) are indexed by the subject &#8216;Image of women&#8217; and 16 issues (14,8%) by the subject &#8216;Image of men&#8217; (cf. <xref ref-type="bibr" rid="B58">V&#246;lkl 2022, 206</xref>).</p></fn>
<fn id="n18"><p>Within the 60 issues of Bastide&#8217;s <italic>Monde comme il est</italic>, 38 issues (64%) are indexed by the subject &#8216;Image of women&#8217; and 12 issues (20%) by the subject &#8216;Image of men&#8217; (cf. <xref ref-type="bibr" rid="B58">V&#246;lkl 2022, 206</xref>).</p></fn>
<fn id="n19"><p>The Spanish topic 8 and the French topic 9 show the following equivalent terms: <italic>virtud</italic> &#8211; <italic>vertu, vida</italic> &#8211; <italic>vie, honor/honrar</italic> &#8211; <italic>honneur, alma</italic> &#8211; <italic>ame, viciar</italic> &#8211; <italic>vice, noble</italic> &#8211; <italic>noble, felicidad</italic> &#8211; <italic>bonheur</italic>.</p></fn>
<fn id="n20"><p>Regarding gender discourse in 18<sup>th</sup> century Spain, see e.g., the monographs and articles by <xref ref-type="bibr" rid="B38">Mart&#237;n Gaite 1972</xref>, <xref ref-type="bibr" rid="B23">Hassauer 1997</xref>, <xref ref-type="bibr" rid="B6">Bolufer Peruga 1998</xref>, <xref ref-type="bibr" rid="B8">Brink 2008</xref>, <xref ref-type="bibr" rid="B9">Capel Mart&#237;nez 2010</xref>, or <xref ref-type="bibr" rid="B22">Gronemann 2013</xref>; on the gender discourses in the Spanish novels of the &#8218;siglo de las luces&#8217;, see <xref ref-type="bibr" rid="B24">Hertel-Mesenh&#246;ller 2001</xref> or <xref ref-type="bibr" rid="B33">Kilian 2002</xref>.</p></fn>
<fn id="n21"><p>The term &#8216;cortejo&#8217;, which only exists in the masculine form, is not only used to designate the man in this special relationship with a married woman, but also for the woman who allows herself to be courted, and furthermore even to paraphrase the liaison itself (cf. <xref ref-type="bibr" rid="B25">He&#223;e 2008</xref>, 135&#8211;136).</p></fn>
<fn id="n22"><p>Topic 2: venir salir pasar tomar mano a&#241;o llegar llamar mil quedar mundo volver entender contar hora amigar o&#237;r acabar noche palabra.</p></fn>
</fn-group>
<ref-list>
<ref id="B1"><label>1</label><mixed-citation publication-type="book"><string-name><surname>Alloghani</surname>, <given-names>Mohamed</given-names></string-name>, <string-name><given-names>Dhiya</given-names> <surname>Al-Jumeily</surname></string-name>, <string-name><given-names>Jamila</given-names> <surname>Mustafina</surname></string-name>, <string-name><given-names>Abir</given-names> <surname>Hussain</surname></string-name>, and <string-name><given-names>Ahmed J.</given-names> <surname>Aljaaf</surname></string-name> (<year>2020</year>). <chapter-title>&#8220;Chapter 1: A Systematic Review on Supervised and Unsupervised Machine Learning Algorithms for Data Science&#8221;</chapter-title>. In: <source>Supervised and Unsupervised Learning for Data Science</source>. Ed. by <string-name><given-names>Michael W.</given-names> <surname>Berry</surname></string-name>, <string-name><given-names>Azlinah</given-names> <surname>Mohamed</surname></string-name>, and <string-name><given-names>Bee Wah</given-names> <surname>Yap</surname></string-name>. Unsupervised and Semi-Supervised Learning. <publisher-name>Springer</publisher-name>, pp. <fpage>3</fpage>&#8211;<lpage>22</lpage>.</mixed-citation></ref>
<ref id="B2"><label>2</label><mixed-citation publication-type="webpage"><string-name><surname>Blei</surname>, <given-names>David M</given-names></string-name>. (<year>2020</year>). <article-title>&#8220;Topic Modeling and Digital Humanities&#8221;</article-title>. In: <source>Journal of Digital Humanities</source> <volume>1</volume> (<issue>2</issue>), pp. <fpage>8</fpage>&#8211;<lpage>11</lpage>. URL: <uri>http://journalofdigitalhumanities.org/2-1/topic-modeling-and-digital-humanities-by-david-m-blei/</uri> (visited on 12/20/2021).</mixed-citation></ref>
<ref id="B3"><label>3</label><mixed-citation publication-type="webpage"><string-name><surname>Blevins</surname>, <given-names>Cameron</given-names></string-name> (<year>2010</year>). <chapter-title>&#8220;Topic Modeling Martha Ballard&#8217;s Diary&#8221;</chapter-title>. In: URL: <uri>https://www.cameronblevins.org/posts/topic-modeling-martha-ballards-diary/</uri> (visited on 12/20/2021).</mixed-citation></ref>
<ref id="B4"><label>4</label><mixed-citation publication-type="book"><string-name><surname>Bock</surname>, <given-names>Gisela</given-names></string-name> and <string-name><given-names>Margarete</given-names> <surname>Zimmermann</surname></string-name>, eds. (<year>1997</year>). <source>Jahrbuch f&#252;r Frauenforschung. Band 2. Die europ&#228;ische Querelle des Femmes. Geschlechterdebatten seit dem 15. Jahrhundert</source>. <publisher-name>Metzler</publisher-name>.</mixed-citation></ref>
<ref id="B5"><label>5</label><mixed-citation publication-type="journal"><string-name><surname>Bock</surname>, <given-names>Sina</given-names></string-name>, <string-name><given-names>Keli</given-names> <surname>Du</surname></string-name>, <string-name><given-names>Michael</given-names> <surname>Huber</surname></string-name>, <string-name><given-names>Stefan</given-names> <surname>Pernes</surname></string-name>, and <string-name><given-names>Steffen</given-names> <surname>Pielstr&#246;m</surname></string-name> (<year>2016</year>). <article-title>&#8220;Der Einsatz quantitativer Textanalyse in den Geisteswissenschaften. Bericht &#252;ber den Stand der Forschung&#8221;</article-title>. In: <source>DARIAH-DE Working Papers (18)</source>. Ed. by <string-name><given-names>Mirjam</given-names> <surname>Bl&#252;mm</surname></string-name>, <string-name><given-names>Thomas</given-names> <surname>Kollatz</surname></string-name>, <string-name><given-names>Stefan</given-names> <surname>Schmunk</surname></string-name>, and <string-name><given-names>Christof</given-names> <surname>Sch&#246;ch</surname></string-name>. URL: <uri>http://webdoc.sub.gwdg.de/pub/mon/dariah-de/dwp-2016-18.pdf</uri> (visited on 04/22/2022).</mixed-citation></ref>
<ref id="B6"><label>6</label><mixed-citation publication-type="book"><string-name><surname>Bolufer Peruga</surname>, <given-names>M&#243;nica</given-names></string-name> (<year>1998</year>). <source>Mujeres e Ilustraci&#243;n. La construcci&#243;n de la feminidad en la Ilustraci&#243;n espa&#241;ola</source>. <publisher-name>Inst. Alfons el Magn&#224;nim</publisher-name>.</mixed-citation></ref>
<ref id="B7"><label>7</label><mixed-citation publication-type="webpage"><string-name><surname>Boyd-Graber</surname>, <given-names>Jordan</given-names></string-name>, <string-name><given-names>Yuening</given-names> <surname>Hu</surname></string-name>, and <string-name><given-names>David</given-names> <surname>Mimno</surname></string-name> (<year>2017</year>). <chapter-title>&#8220;Applications of Topic Models&#8221;</chapter-title>. In: <source>Foundations and Trends in Information Retrieval</source>. URL: <uri>https://mimno.infosci.cornell.edu/papers/2017_fntir_tm_applications.pdf</uri> (visited on 12/20/2021).</mixed-citation></ref>
<ref id="B8"><label>8</label><mixed-citation publication-type="book"><string-name><surname>Brink</surname>, <given-names>Margot</given-names></string-name> (<year>2008</year>). <chapter-title>&#8220;Geschlechterstreit und Dialektik der Aufkl&#228;rung in Spanien und Frankreich. Die ambivalente Rolle von Vernunft und Natur in Egalit&#228;ts- und Komplementarit&#228;tstheorien des 18. Jahrhunderts&#8221;</chapter-title>. In: <source>Hei&#223;er Streit und kalte Ordnung. Epochen der &#8216;Querelle des femmes&#8217; zwischen Mittelalter und Gegenwart</source>. Ed. by <string-name><given-names>Friederike</given-names> <surname>Hassauer</surname></string-name>. <publisher-name>Wallstein Verlag</publisher-name>, pp. <fpage>344</fpage>&#8211;<lpage>364</lpage>.</mixed-citation></ref>
<ref id="B9"><label>9</label><mixed-citation publication-type="webpage"><string-name><surname>Capel Mart&#237;nez</surname>, <given-names>Rosa Mar&#237;a</given-names></string-name> (<year>2010</year>). <chapter-title>&#8220;Prensa y Escritura Femenina en la Espa&#241;a Ilustrada&#8221;</chapter-title>. In: <source>El Argonauta espa&#241;ol</source> (<volume>7</volume>). URL: <uri>http://journals.openedition.org/argonauta/431</uri> (visited on 12/20/2021).</mixed-citation></ref>
<ref id="B10"><label>10</label><mixed-citation publication-type="journal"><string-name><surname>Chehal</surname>, <given-names>Dimple</given-names></string-name>, <string-name><given-names>Perul</given-names> <surname>Gupta</surname></string-name>, and <string-name><given-names>Payal</given-names> <surname>Gulati</surname></string-name> (<year>2021</year>). <article-title>&#8220;Implementation and comparison of topic modeling techniques based on user reviews in e-commerce recommendations&#8221;</article-title>. In: <source>Journal of Ambient Intelligence and Humanized Computing</source> (<volume>12</volume>), pp. <fpage>5055</fpage>&#8211;<lpage>5070</lpage>. DOI: <pub-id pub-id-type="doi">10.1007/s12652-020-01956-6</pub-id>.</mixed-citation></ref>
<ref id="B11"><label>11</label><mixed-citation publication-type="webpage"><collab>DARIAH-DE</collab> (<year>2019</year>). <source>Notebook IntroducingMallet.ipynb</source>. URL: <uri>https://github.com/DARIAH-DE/Topics</uri> (visited on 04/01/2019).</mixed-citation></ref>
<ref id="B12"><label>12</label><mixed-citation publication-type="webpage"><string-name><surname>Diaz</surname>, <given-names>Gene</given-names></string-name> (<year>2016</year>). <source>Stopwords ISO</source>. URL: <uri>https://github.com/stopwords-iso</uri> (visited on 09/20/2019).</mixed-citation></ref>
<ref id="B13"><label>13</label><mixed-citation publication-type="book"><string-name><surname>Dijk</surname>, <given-names>Suzanna van</given-names></string-name> (<year>1988</year>). <source>Traces de femmes. Pr&#233;sence f&#233;minine dans le journalisme fran&#231;ais du XVIIIe si&#232;cle</source>. <publisher-name>APA, Holland University Press</publisher-name>.</mixed-citation></ref>
<ref id="B14"><label>14</label><mixed-citation publication-type="webpage"><string-name><surname>Dobson</surname>, <given-names>James</given-names></string-name> (<year>2021</year>). <chapter-title>&#8220;Interpretable Outputs: Criteria for Machine Learning in the Humanities&#8221;</chapter-title>. In: <source>Digital Humanities Quarterly</source> <volume>2</volume> (<issue>15</issue>). URL: <uri>http://digitalhumanities.org:8081/dhq/vol/15/2/000555/000555.html</uri> (visited on 12/20/2021).</mixed-citation></ref>
<ref id="B15"><label>15</label><mixed-citation publication-type="journal"><string-name><surname>Du</surname>, <given-names>Keli</given-names></string-name> (<year>2019</year>). <article-title>&#8220;A Survey On LDA Topic Modeling In Digital Humanities&#8221;</article-title>. In: <source>Book of Abstracts DH2019</source>. DOI: <pub-id pub-id-type="doi">10.34894/H9UYPI</pub-id>.</mixed-citation></ref>
<ref id="B16"><label>16</label><mixed-citation publication-type="book"><string-name><surname>Ertler</surname>, <given-names>Klaus-Dieter</given-names></string-name> (<year>2010</year>). <chapter-title>&#8220;Die Moralischen Wochenschriften als Vehikel zur diskursiven Ausdifferenzierung der Nation in Spanien&#8221;</chapter-title>. In: <source>Beitr&#228;ge zur Nationalisierung der Kultur im Spanien des aufgekl&#228;rten Absolutismus</source>. Ed. by <string-name><given-names>Jan-Henrik</given-names> <surname>Witthaus</surname></string-name>. <publisher-name>Peter Lang</publisher-name>, pp. <fpage>93</fpage>&#8211;<lpage>107</lpage>.</mixed-citation></ref>
<ref id="B17"><label>17</label><mixed-citation publication-type="webpage"><string-name><surname>Ertler</surname>, <given-names>Klaus-Dieter</given-names></string-name>, <string-name><given-names>Alexandra</given-names> <surname>Fuchs</surname></string-name>, <string-name><given-names>Michaela</given-names> <surname>Fischer-Pernkopf</surname></string-name>, <string-name><given-names>Elisabeth</given-names> <surname>Hobisch</surname></string-name>, <string-name><given-names>Martina</given-names> <surname>Scholger</surname></string-name>, and <string-name><given-names>Yvonne</given-names> <surname>V&#246;lkl</surname></string-name> (<year>2011&#8211;2021</year>). <source>The Spectators in the International Context</source>. URL: <uri>https://gams.uni-graz.at/spectators</uri> (visited on 12/20/2021).</mixed-citation></ref>
<ref id="B18"><label>18</label><mixed-citation publication-type="journal"><string-name><surname>Fechner</surname>, <given-names>Martin</given-names></string-name> and <string-name><given-names>Andreas</given-names> <surname>Wei&#223;</surname></string-name> (<year>2017</year>). <article-title>&#8220;Einsatz von Topic Modeling in den Geschichtswissenschaften: Wissensbest&#228;nde des 19. Jahrhunderts&#8221;</article-title>. In: <source>Zeitschrift f&#252;r digitale Geisteswissenschaften</source>. DOI: <pub-id pub-id-type="doi">10.17175/2017_005</pub-id>.</mixed-citation></ref>
<ref id="B19"><label>19</label><mixed-citation publication-type="book"><string-name><surname>Fischer</surname>, <given-names>Michaela</given-names></string-name> (<year>2014</year>). <source>Die Figur des Lesers im Kommunikationssystem der Spectateurs</source>. <publisher-name>Peter Lang</publisher-name>.</mixed-citation></ref>
<ref id="B20"><label>20</label><mixed-citation publication-type="book"><string-name><surname>Fischer-Pernkopf</surname>, <given-names>Michaela</given-names></string-name>, <string-name><given-names>Veronika</given-names> <surname>Mussner</surname></string-name>, and <string-name><given-names>Klaus-Dieter</given-names> <surname>Ertler</surname></string-name> (<year>2018</year>). <source>Die &#171;Spectators&#187; in Frankreich. &#171;Le Nouveau Spectateur&#187; und &#171;Le Monde comme il est&#187; von Jean-Fran&#231;ois de Bastide</source>. <publisher-name>Peter Lang</publisher-name>.</mixed-citation></ref>
<ref id="B21"><label>21</label><mixed-citation publication-type="webpage"><collab>Gephi.org</collab> (<year>2008&#8211;2021</year>). <source>Gephi V.0.9.2</source>. URL: <uri>https://gephi.org/</uri> (visited on 12/20/2021).</mixed-citation></ref>
<ref id="B22"><label>22</label><mixed-citation publication-type="book"><string-name><surname>Gronemann</surname>, <given-names>Claudia</given-names></string-name> (<year>2013</year>). <source>Polyphone Aufkl&#228;rung. Zur Textualit&#228;t und Performativit&#228;t der spanischen Geschlechterdebatten im 18. Jahrhundert</source>. <publisher-name>Vervuert</publisher-name>.</mixed-citation></ref>
<ref id="B23"><label>23</label><mixed-citation publication-type="book"><string-name><surname>Hassauer</surname>, <given-names>Friederike</given-names></string-name> (<year>1997</year>). <chapter-title>&#8220;Die Seele ist nicht Mann, nicht Weib. Stationen der Querelles des Femmes in Spanien und Lateinamerika vom 16. zum 18. Jahrhundert&#8221;</chapter-title>. In: <source>Die europ&#228;ische Querelle des Femmes. Geschlechterdebatten seit dem 15. Jahrhundert</source>. Ed. by <string-name><given-names>Gisela</given-names> <surname>Bock</surname></string-name> and <string-name><given-names>Margarete</given-names> <surname>Zimmermann</surname></string-name>. Jahrbuch f&#252;r Frauenforschung. <publisher-name>Metzler</publisher-name>, pp. <fpage>203</fpage>&#8211;<lpage>238</lpage>.</mixed-citation></ref>
<ref id="B24"><label>24</label><mixed-citation publication-type="book"><string-name><surname>Hertel-Mesenh&#246;ller</surname>, <given-names>Heike</given-names></string-name> (<year>2001</year>). <source>Das Bild der Frau im spanischen Roman des 18. Jahrhunderts. Im Spannungsfeld von Lebenswirklichkeit und Fiktion</source>. <publisher-name>Vervuert</publisher-name>.</mixed-citation></ref>
<ref id="B25"><label>25</label><mixed-citation publication-type="book"><string-name><surname>He&#223;e</surname>, <given-names>Kristina</given-names></string-name> (<year>2008</year>). <source>M&#228;nnlichkeiten im Spanien der Aufkl&#228;rung. Der Diskurs der moralischen Wochenschriften &#8217;El Pensador&#8217;, &#8217;La Pensadora gaditana&#8217; und &#8217;El Censor&#8217;</source>. <publisher-name>Logos</publisher-name>.</mixed-citation></ref>
<ref id="B26"><label>26</label><mixed-citation publication-type="book"><string-name><surname>Hobisch</surname>, <given-names>Elisabeth</given-names></string-name> (<year>2017</year>). <source>La forma epistolar en los espectadores espa&#241;oles: Caracter&#237;sticas y tipolog&#237;a de las cartas</source>. <publisher-name>Peter Lang</publisher-name>.</mixed-citation></ref>
<ref id="B27"><label>27</label><mixed-citation publication-type="book"><string-name><surname>Hobisch</surname>, <given-names>Elisabeth</given-names></string-name> (<year>2018</year>). <chapter-title>&#8220;Les strat&#233;gies publicitaires dans les lettres des &#8217;spectateurs&#8217; espagnols&#8221;</chapter-title>. In: <source>Discourses on Economy in the Spectators / Discours sur l&#8217;&#233;conomie dans les spectateurs</source>. Ed. by <string-name><given-names>Klaus-Dieter</given-names> <surname>Ertler</surname></string-name>, <string-name><given-names>Samuel</given-names> <surname>Baudry</surname></string-name>, and <string-name><given-names>Yvonne</given-names> <surname>V&#246;lkl</surname></string-name>. <publisher-name>Dr. Kova&#269;</publisher-name>, pp. <fpage>199</fpage>&#8211;<lpage>214</lpage>.</mixed-citation></ref>
<ref id="B28"><label>28</label><mixed-citation publication-type="book"><string-name><surname>Honegger</surname>, <given-names>Claudia</given-names></string-name> (<year>2011</year>). <chapter-title>&#8220;Die kognitiven Prinzipien der neuen Wissenschaften vom Menschen und die Genese einer weiblichen Sonderanthropologie in Frankreich&#8221;</chapter-title>. In: <source>Die gesellschaftliche Verortung des Geschlechts. Diskurse der Differenz in der deutschen und franz&#246;sischen Soziologie um 1900</source>. Ed. by <string-name><given-names>Theresa</given-names> <surname>Wobbe</surname></string-name>. <publisher-name>Campus</publisher-name>, pp. <fpage>93</fpage>&#8211;<lpage>113</lpage>.</mixed-citation></ref>
<ref id="B29"><label>29</label><mixed-citation publication-type="journal"><string-name><surname>Horstmann</surname>, <given-names>Jan</given-names></string-name> and <string-name><given-names>Rabea</given-names> <surname>Kleymann</surname></string-name> (<year>2019</year>). <article-title>&#8220;Alte Fragen, neue Methoden &#8211; Philologische und digitale Verfahren im Dialog. Ein Beitrag zum Forschungsdiskurs um Entsagung und Ironie bei Goethe&#8221;</article-title>. In: <source>Zeitschrift f&#252;r digitale Geisteswissenschaften</source>. DOI: <pub-id pub-id-type="doi">10.17175/2019_007</pub-id>.</mixed-citation></ref>
<ref id="B30"><label>30</label><mixed-citation publication-type="journal"><string-name><surname>Hu</surname>, <given-names>Yuening</given-names></string-name>, <string-name><given-names>Jordan</given-names> <surname>Boyd-Graber</surname></string-name>, <string-name><given-names>Brianna</given-names> <surname>Satinoff</surname></string-name>, and <string-name><given-names>Alison</given-names> <surname>Smith</surname></string-name> (<year>2014</year>). <article-title>&#8220;Interactive topic modeling&#8221;</article-title>. In: <source>Mach Learn</source> (<volume>95</volume>), pp. <fpage>423</fpage>&#8211;<lpage>469</lpage>. DOI: <pub-id pub-id-type="doi">10.1007/s10994-013-5413-0</pub-id>.</mixed-citation></ref>
<ref id="B31"><label>31</label><mixed-citation publication-type="book"><string-name><surname>Jacobs</surname>, <given-names>Eva</given-names></string-name>, <string-name><given-names>W. H.</given-names> <surname>Barber</surname></string-name>, <string-name><given-names>Jean H.</given-names> <surname>Bloch</surname></string-name>, <string-name><given-names>F. W.</given-names> <surname>Leakey</surname></string-name>, and <string-name><given-names>Eileen Le</given-names> <surname>Breton</surname></string-name> (<year>1979</year>). <source>Woman and society in eighteenth-century France : essays in honour of John Stephenson Spink</source>. <publisher-name>Athlone Press</publisher-name>.</mixed-citation></ref>
<ref id="B32"><label>32</label><mixed-citation publication-type="book"><string-name><surname>Jockers</surname>, <given-names>Matthew Lee</given-names></string-name> (<year>2014</year>). <source>Text analysis with R for students of literature</source>. Quantitative Methods in the Humanities and Social Sciences. <publisher-name>Springer, Cham</publisher-name>.</mixed-citation></ref>
<ref id="B33"><label>33</label><mixed-citation publication-type="book"><string-name><surname>Kilian</surname>, <given-names>Elena</given-names></string-name> (<year>2002</year>). <source>Bildung, Tugend, N&#252;tzlichkeit. Geschlechterentw&#252;rfe im spanischen Aufkl&#228;rungsroman des sp&#228;ten 18. Jahrhunderts</source>. <publisher-name>K&#246;nigshausen &amp; Neumann</publisher-name>.</mixed-citation></ref>
<ref id="B34"><label>34</label><mixed-citation publication-type="journal"><string-name><surname>Krautter</surname>, <given-names>Benjamin</given-names></string-name>, <string-name><given-names>Janis</given-names> <surname>Pagel</surname></string-name>, <string-name><given-names>Nils</given-names> <surname>Reiter</surname></string-name>, and <string-name><given-names>Marcus</given-names> <surname>Willand</surname></string-name> (<year>2020</year>). <article-title>&#8220;&#187;[E]in Vater, d&#228;chte ich, ist doch immer ein Vater&#171;. Figurentypen im Drama und ihre Operationalisierung.&#8221;</article-title> In: <source>Zeitschrift f&#252;r digitale Geisteswissenschaften</source>. DOI: <pub-id pub-id-type="doi">10.17175/2020_007</pub-id>.</mixed-citation></ref>
<ref id="B35"><label>35</label><mixed-citation publication-type="book"><string-name><surname>Laqueur</surname>, <given-names>Thomas</given-names></string-name> (<year>2003[1990]</year>). <source>Making Sex. Body and Gender from the Greeks to Freud</source>. <publisher-name>Harvard University Press</publisher-name>.</mixed-citation></ref>
<ref id="B36"><label>36</label><mixed-citation publication-type="webpage"><string-name><surname>Liu</surname>, <given-names>Alan</given-names></string-name>, <string-name><given-names>Scott</given-names> <surname>Kleinman</surname></string-name>, <string-name><given-names>Jeremy</given-names> <surname>Douglass</surname></string-name>, <string-name><given-names>Thomas</given-names> <surname>Lindsay</surname></string-name>, <string-name><given-names>Ashley</given-names> <surname>Champagne</surname></string-name>, and <string-name><given-names>Jamal</given-names> <surname>Russell</surname></string-name> (<year>2017</year>). <chapter-title>&#8220;Open, Shareable, Reproducible Workflows for the Digital Humanities: The Case of the 4Humanities.org &#8217;WhatEvery1Says&#8217; Project&#8221;</chapter-title>. In: <source>Digital Humanities 2017. Conference Abstracts</source>. <publisher-name>McGill University and Universit&#233; de Montr&#233;al</publisher-name>, pp. <fpage>95</fpage>&#8211;<lpage>98</lpage>. URL: <uri>https://dh2017.adho.org/abstracts/DH2017-abstracts.pdf</uri> (visited on 11/20/2021).</mixed-citation></ref>
<ref id="B37"><label>37</label><mixed-citation publication-type="webpage"><string-name><surname>Marjanen</surname>, <given-names>Jani</given-names></string-name>, <string-name><given-names>Elaine</given-names> <surname>Zosa</surname></string-name>, <string-name><given-names>Simon</given-names> <surname>Hengchen</surname></string-name>, <string-name><given-names>Lidia</given-names> <surname>Pivovarova</surname></string-name>, and <string-name><given-names>Mikko</given-names> <surname>Tolonen</surname></string-name> (<year>2020</year>). <chapter-title>&#8220;Topic modelling discourse dynamics in historical newspapers&#8221;</chapter-title>. In: <source>arXiv preprint</source>. DOI: <pub-id pub-id-type="doi">10.48550/arXiv.2011.10428</pub-id>.</mixed-citation></ref>
<ref id="B38"><label>38</label><mixed-citation publication-type="book"><string-name><surname>Mart&#237;n Gaite</surname>, <given-names>Carmen</given-names></string-name> (<year>1972</year>). <source>Usos amorosos del dieciocho en Espa&#241;a</source>. <publisher-name>Siglo XXI de Espa&#241;a Editores</publisher-name>.</mixed-citation></ref>
<ref id="B39"><label>39</label><mixed-citation publication-type="book"><string-name><surname>Mauzi</surname>, <given-names>Robert</given-names></string-name> (<year>1969</year>). <source>L&#8217;idee du bonheur dans la litt&#233;rature et la pens&#233;e fran&#231;aises au XVIIIe si&#232;cle</source>. <publisher-name>Colin</publisher-name>.</mixed-citation></ref>
<ref id="B40"><label>40</label><mixed-citation publication-type="webpage"><string-name><surname>McCallum</surname>, <given-names>Andrew Kachites</given-names></string-name> (<year>2002&#8211;2018</year>). <source>MALLET. A Machine Learning for Language Toolkit. V. 2.0.8</source>. URL: <uri>https://mimno.github.io/Mallet/index</uri> (visited on 11/28/2021).</mixed-citation></ref>
<ref id="B41"><label>41</label><mixed-citation publication-type="journal"><string-name><surname>Murakami</surname>, <given-names>Akira</given-names></string-name>, <string-name><given-names>Paul</given-names> <surname>Thompson</surname></string-name>, <string-name><given-names>Susan</given-names> <surname>Hunston</surname></string-name>, and <string-name><given-names>Dominik</given-names> <surname>Vajn</surname></string-name> (<year>2017</year>). <article-title>&#8220;&#8216;What is this corpus about?&#8217;: using topic modelling to explore a specialised corpus&#8221;</article-title>. In: <source>Corpora</source> <volume>2</volume> (<issue>12</issue>), pp. <fpage>243</fpage>&#8211;<lpage>277</lpage>. DOI: <pub-id pub-id-type="doi">10.3366/cor.2017.0118</pub-id>.</mixed-citation></ref>
<ref id="B42"><label>42</label><mixed-citation publication-type="thesis"><string-name><surname>Mussner</surname>, <given-names>Veronika</given-names></string-name> (<year>2016</year>). <chapter-title>&#8220;Die Moralischen Wochenschriften in Frankreich. &#8217;Le Monde comme il est&#8217; von Jean-Fran&#231;ois de Bastide als Spiegel seiner Zeit&#8221;</chapter-title>. MA thesis. <publisher-name>Universit&#228;t Graz</publisher-name>.</mixed-citation></ref>
<ref id="B43"><label>43</label><mixed-citation publication-type="webpage"><string-name><surname>Nelson</surname>, <given-names>Robert K</given-names></string-name>. (<year>2020</year>). <source>Mining the Dispatch</source>. URL: <uri>https://dsl.richmond.edu/dispatch/</uri> (visited on 12/20/2021).</mixed-citation></ref>
<ref id="B44"><label>44</label><mixed-citation publication-type="journal"><string-name><surname>Norton</surname>, <given-names>Brian Michael</given-names></string-name> (<year>2015</year>). <article-title>&#8220;The Spectator, Aesthetic Experience and the Modern Idea of Happiness&#8221;</article-title>. In: <source>English Literature</source> <volume>2</volume> (<issue>1</issue>), pp. <fpage>87</fpage>&#8211;<lpage>104</lpage>.</mixed-citation></ref>
<ref id="B45"><label>45</label><mixed-citation publication-type="book"><string-name><surname>Pabst</surname>, <given-names>Esther S</given-names></string-name>. (<year>2007</year>). <source>Die Erfindung der weiblichen Tugend. Kulturelle Sinngebung und Selbstreflexion im franz&#246;sischen Briefroman des 18. Jahrhunderts</source>. <publisher-name>Wallstein</publisher-name>.</mixed-citation></ref>
<ref id="B46"><label>46</label><mixed-citation publication-type="webpage"><string-name><surname>Portela</surname>, <given-names>Manuel</given-names></string-name> and <string-name><given-names>Ant&#243;nio Rito</given-names> <surname>Silva</surname></string-name> (<year>2017</year>). <source>Arquivo LdoD: Arquivo Digital Colaborativo do Livro do Desassossego</source>. URL: <uri>https://ldod.uc.pt/</uri>.</mixed-citation></ref>
<ref id="B47"><label>47</label><mixed-citation publication-type="journal"><string-name><surname>Roe</surname>, <given-names>Glenn</given-names></string-name>, <string-name><given-names>Clovis</given-names> <surname>Gladstone</surname></string-name>, and <string-name><given-names>Robert</given-names> <surname>Morrissey</surname></string-name> (<year>2016</year>). <article-title>&#8220;Discourses and Disciplines in the Enlightenment: Topic Modeling the French Encyclop&#233;die&#8221;</article-title>. In: <source>Frontiers in Digital Humanities</source> (<volume>2</volume>). DOI: <pub-id pub-id-type="doi">10.3389/fdigh.2015.00008</pub-id>.</mixed-citation></ref>
<ref id="B48"><label>48</label><mixed-citation publication-type="book"><string-name><surname>Schaufler</surname>, <given-names>Birgit</given-names></string-name> (<year>2002</year>). <source>&#8217;Sch&#246;ne Frauen &#8211; starke M&#228;nner&#8217;. Zur Konstruktion von Leib, K&#246;rper und Geschlecht</source>. <publisher-name>Leske + Budrich</publisher-name>.</mixed-citation></ref>
<ref id="B49"><label>49</label><mixed-citation publication-type="webpage"><string-name><surname>Sch&#246;ch</surname>, <given-names>Christof</given-names></string-name> (<year>2016</year>). <chapter-title>&#8220;Topic Modeling with MALLET. Hyperparameter Optimization&#8221;</chapter-title>. In: URL: <uri>https://dragonfly.hypotheses.org/1051</uri> (visited on 12/20/2021).</mixed-citation></ref>
<ref id="B50"><label>50</label><mixed-citation publication-type="webpage"><string-name><surname>Sch&#246;ch</surname>, <given-names>Christof</given-names></string-name> (<year>2017</year>). <chapter-title>&#8220;Topic Modeling Genre. An Exploration of French Classical and Enlightenment Drama&#8221;</chapter-title>. In: <source>Digital Humanities Quarterly</source> <volume>2</volume> (<issue>11</issue>). URL: <uri>http://www.digitalhumanities.org/dhq/vol/11/2/000291/000291.html</uri> (visited on 12/20/2021).</mixed-citation></ref>
<ref id="B51"><label>51</label><mixed-citation publication-type="webpage"><string-name><surname>Scholger</surname>, <given-names>Martina</given-names></string-name>, <string-name><given-names>Bernhard</given-names> <surname>Geiger</surname></string-name>, <string-name><given-names>Elisabeth</given-names> <surname>Hobisch</surname></string-name>, <string-name><given-names>Philipp</given-names> <surname>Koncar</surname></string-name>, <string-name><given-names>Sanja</given-names> <surname>Sari&#263;</surname></string-name>, <string-name><given-names>Yvonne</given-names> <surname>V&#246;lkl</surname></string-name>, and <string-name><given-names>Christina</given-names> <surname>Glatz</surname></string-name> (<year>2022</year>). <source>Distant Spectators. Distant Reading for Periodicals of the Enlightenment (DiSpecs). GitHub repository</source>. URL: <uri>https://github.com/distantspectators/DiSpecs</uri>.</mixed-citation></ref>
<ref id="B52"><label>52</label><mixed-citation publication-type="webpage"><string-name><surname>Scholger</surname>, <given-names>Martina</given-names></string-name>, <string-name><given-names>Bernhard</given-names> <surname>Geiger</surname></string-name>, <string-name><given-names>Elisabeth</given-names> <surname>Hobisch</surname></string-name>, <string-name><given-names>Philipp</given-names> <surname>Koncar</surname></string-name>, <string-name><given-names>Sanja</given-names> <surname>Sari&#263;</surname></string-name>, <string-name><given-names>Yvonne</given-names> <surname>V&#246;lkl</surname></string-name>, and <string-name><given-names>Christina</given-names> <surname>Glatz</surname></string-name> (<year>2019&#8211;2021</year>). <source>Distant Spectators. Distant Reading for Periodicals of the Enlightenment (DiSpecs)</source>. URL: <uri>https://gams.uni-graz.at/dispecs</uri> (visited on 12/20/2021).</mixed-citation></ref>
<ref id="B53"><label>53</label><mixed-citation publication-type="journal"><string-name><surname>Shadrova</surname>, <given-names>Anna</given-names></string-name> (<year>2021</year>). <article-title>&#8220;Topic models do not model topics: epistemological remarks and steps towards best practices&#8221;</article-title>. In: <source>Journal of Data Mining &amp; Digital Humanities</source> (2021). DOI: <pub-id pub-id-type="doi">10.46298/jdmdh.7595</pub-id>.</mixed-citation></ref>
<ref id="B54"><label>54</label><mixed-citation publication-type="book"><string-name><surname>Sieuzac</surname>, <given-names>Laurence</given-names></string-name> (<year>2009</year>). <chapter-title>&#8220;&#201;ducation et vocation de la femme au Si&#232;cle des lumi&#232;res&#8221;</chapter-title>. In: <source>Genre &amp; &#201;ducation. Former, se former, &#234;tre form&#233;e au f&#233;minin</source>. Ed. by <string-name><given-names>Paul</given-names> <surname>Pasteur</surname></string-name> et al. <publisher-name>Presses universitaires de Rouen et du Havre</publisher-name>, pp. <fpage>271</fpage>&#8211;<lpage>287</lpage>.</mixed-citation></ref>
<ref id="B55"><label>55</label><mixed-citation publication-type="book"><string-name><surname>Steinbr&#252;gge</surname>, <given-names>Lieselotte</given-names></string-name> (<year>1987</year>). <source>Das moralische Geschlecht. Theorien und literarische Entw&#252;rfe &#252;ber die Natur der Frau in der franz&#246;sischen Aufkl&#228;rung</source>. <publisher-name>Beltz</publisher-name>.</mixed-citation></ref>
<ref id="B56"><label>56</label><mixed-citation publication-type="book"><string-name><surname>Sullerot</surname>, <given-names>Evelyne</given-names></string-name> (<year>1966</year>). <source>Histoire de la Presse f&#233;minine en France des origines &#224; 1848</source>. <publisher-name>Armand Colin</publisher-name>.</mixed-citation></ref>
<ref id="B57"><label>57</label><mixed-citation publication-type="webpage"><collab>TEI Consortium</collab> (<year>2021</year>). <source>Guidelines for Electronic Text Encoding and Interchange</source>. URL: <uri>http://www.tei-c.org/P5</uri> (visited on 12/20/2021).</mixed-citation></ref>
<ref id="B58"><label>58</label><mixed-citation publication-type="book"><string-name><surname>V&#246;lkl</surname>, <given-names>Yvonne</given-names></string-name> (<year>2022</year>). <source>Spectatoriale Geschlechterkonstruktionen: Geschlechtsspezifische Wissensund Welterzeugung in den franz&#246;sisch- und spanischsprachigen Moralischen Wochenschriften des 18. Jahrhunderts</source>. <publisher-name>transcript</publisher-name>.</mixed-citation></ref>
</ref-list>
</back>
</article>
