<?xml version="1.0" encoding="UTF-8"?>
<TEI xmlns="http://www.tei-c.org/ns/1.0" xml:lang="en">
  <teiHeader>
    <fileDesc>
      <titleStmt>
        <title>A Network Analysis of Figurative Topic Classification: The Case Study of Timon of Athens</title>
        <author>Gilad Gutman</author>
      </titleStmt>
      <publicationStmt>
        <p>Tag layer of the DHSS Meta-Tagging Project. The article text is reproduced from the
           staged copy named in sourceDesc and is not altered; the tags are held separately in
           standOff so that overlapping and nested spans are preserved.</p>
      </publicationStmt>
      <sourceDesc>
        <bibl>
          <title>A Network Analysis of Figurative Topic Classification: The Case Study of Timon of Athens</title>
          <date>2024</date>
          <publisher>Digital Humanities Quarterly 18(3)</publisher>
          
          
          <idno type="record">gutman2024-figurative-topics</idno>
          <note type="staged-file">gutman2024-figurative-topics.txt, the project's staged copy</note>
          <note type="text-sha1">ff3e73244c50810930e1ea48e377d8ef52aba19c</note>
          <note type="rights">open-access</note>
        </bibl>
      </sourceDesc>
    </fileDesc>
    <profileDesc>
      <textClass><keywords><term>digital-humanities</term></keywords></textClass>
      <particDesc>
        <person xml:id="paper"><p>the paper's own words</p></person>
        <person xml:id="tagger"><p>the tag layer; a sentence written by the tagger about the
          paper, always anchored to a quotation but not itself a quotation</p></person>
      </particDesc>
    </profileDesc>
    <revisionDesc>
      <change when="2026-10-06"
              who="build_paper_edition.py">0 spans located by searching the stored text for the quote's exact words, allowing any whitespace between them</change>
    </revisionDesc>
  </teiHeader>
  <standOff>
    <spanGrp type="tags">
      <span xml:id="t0" type="claim_type" ana="#claim_type" resp="#paper">
        <label>method</label>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t1" type="phenomenon" ana="#phenomenon" resp="#paper">
        <label>figurative language as a network of topics</label>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t2" type="method" ana="#method" resp="#paper">
        <label>system-or-tool-building</label>
        <term>system-or-tool-building</term>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t3" type="mechanisms" ana="#mechanisms[0]" resp="#tagger" cert="0.95">
        <label>mechanisms</label>
        <term>THE DIFFICULTY IS ATTRIBUTED TO THE STRUCTURE OF THE OBJECT, NOT TO THE METHOD: analysing figurative RATHER THAN LITERAL language POSES CERTAIN DIFFICULTIES DUE TO THE VERY STRUCTURE OF FIGURATIVE EXPRESSIONS - so the obstacle is located in what is being measured before any tool is named</term>
        <note>about this quote: The span contains the author&#x27;s typo &quot;in this articles&quot; for &quot;in this article&quot;. Not extraction damage - it is in the published DHQ text - and quoted verbatim.</note>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t4" type="mechanisms" ana="#mechanisms[1]" resp="#tagger" cert="0.95">
        <label>mechanisms</label>
        <term>A PROHIBITION IS SHOWN VIOLATING ITSELF IN ITS OWN WORDING: the Areopagite judges CONDEMN the use of figurative language AS DECEPTIVE, and THEY THEMSELVES USE A METAPHOR OF A CARPENTER IN THEIR CONDEMNATION. The rule against a class of language is stated in that class of language</term>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t5" type="mechanisms" ana="#mechanisms[2]" resp="#tagger" cert="0.95">
        <label>mechanisms</label>
        <term>AND THE SHARPER POINT IS THAT THE HOMONYMY DOES THE WORK, NOT THE METAPHOR: the influence is NOT ONLY evident through the carpenter - RATHER IT IS THE MORE SUBTLE HOMONYM SENSES OF THE WORDS UPRIGHT AND RULE THAT TRULY REVEAL THE PERVASIVE POWER OF THE FIGURATIVE. Two words carrying two senses each, unnoticed by the speakers, are what the argument turns on</term>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t6" type="mechanisms" ana="#mechanisms[3]" resp="#tagger" cert="0.95">
        <label>mechanisms</label>
        <term>A CLASS OF METAPHOR IS DEFINED BY WHAT WOULD BE IMPOSSIBLE WITHOUT IT: the word is catalogued as part of the ORIENTATIONAL class of metaphors, which RELY ON OUR MOST BASIC SENSE OF SPACE AND PHYSICAL CONCEPTS, so much so that WITHOUT WHICH WE COULD NOT FUNCTION IN THE WORLD - COULD NOT REASON OR COMMUNICATE. The category&#x27;s members are identified by the cost of their absence</term>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t7" type="mechanisms" ana="#mechanisms[4]" resp="#tagger" cert="0.85">
        <label>mechanisms</label>
        <term>A METHOD&#x27;S OUTPUT AND THE CONCEPT IT IS MEANT TO MEASURE ARE REPORTED NOT TO CORRESPOND: after comparing a computational reading with a thematic one, the conclusion quoted is WE CANNOT FIND A CLEAR OR CONSTANT RELATIONSHIP BETWEEN AN LDA-TOPIC AND A THEME. The instrument produces something, and it is not the thing the word names</term>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t8" type="mechanisms" ana="#mechanisms[5]" resp="#tagger" cert="0.95">
        <label>mechanisms</label>
        <term>THE METHOD IS SAID TO WORK ON ONE TEXT TYPE AND FAIL ON ANOTHER, AND THE REASON GIVEN IS CONTEXT-EXCLUSIVITY OF VOCABULARY: topic modelling applied to poetry DOES NOT EXTRACT CLEAR AND DEFINED THEMES AS IT DOES WHEN APPLIED TO SCIENTIFIC TEXTS, because IN SCIENTIFIC TEXTS THERE ARE WORDS THAT ARE USED EXCLUSIVELY IN VERY SPECIFIC CONTEXTS - and THE LINGUISTIC USES IN POETRY WORK IN THE EXACT OPPOSITE WAY. The property that makes the method work is a property of the vocabulary, not of the algorithm</term>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t9" type="mechanisms" ana="#mechanisms[6]" resp="#tagger" cert="0.95">
        <label>mechanisms</label>
        <term>THE NAMING STEP IS IDENTIFIED AS THE MANUAL AND SUBJECTIVE ONE: in topic modelling, DISCERNING THE TITLES of the groups of words REMAINS A MANUAL TASK THAT REQUIRES ACTIVE AND OFTEN SUBJECTIVE INTERPRETATION. The algorithm delivers groups; a person supplies the words that say what the groups are</term>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t10" type="mechanisms" ana="#mechanisms[7]" resp="#tagger" cert="0.85">
        <label>mechanisms</label>
        <term>A NAMED DEFECT OF AUTOMATIC GROUPING IS CITED, AND IT IS A DISAGREEMENT BETWEEN MACHINE AND HUMAN CATEGORIES: a list of words treated as one topic MAY INCLUDE WORDS THAT HUMAN INTERPRETERS RECOGNIZE AS BELONGING TO DISTINCT TOPICS - a problem with its own name, WORD INTRUSION. The defect is defined by the mismatch, and the mismatch is what the name records</term>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t11" type="mechanisms" ana="#mechanisms[8]" resp="#tagger" cert="0.85">
        <label>mechanisms</label>
        <term>THE RIGHT RELIABILITY MEASURE IS NAMED, ITS PURPOSE STATED, AND IT IS EXPLICITLY NOT DONE, WITH THE REASON: given the SAME set of topics and instances, TWO INTERPRETERS MIGHT STILL CHOOSE DIFFERENT TOPICS for a word - INTER-ANNOTATORS AGREEMENT is the method by which that agreement can be measured, PROVIDING A STATUS OF CONSENSUS, and the procedure REQUIRES CERTAIN RESOURCES AND HENCE WILL BE IMPLEMENTED later. The measure that would qualify the result is named and deferred in the same breath</term>
        <note>about this quote: EXTENDED to the full stop, and the span contains the author&#x27;s typo &quot;Tthis procedure&quot; with a doubled T. Not extraction damage; it is in the published text. The extension matters because &quot;will be implemented in a future study&quot; is what dates the deferral - the reliability measure is named, its purpose stated, and its absence scheduled, all in one passage.</note>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t12" type="mechanisms" ana="#mechanisms[9]" resp="#tagger" cert="0.95">
        <label>mechanisms</label>
        <term>THE IMPROVEMENTS PROPOSED ARE PROPERTIES OF THE SETUP RATHER THAN OF THE ANALYSIS: the recommendations begin with REDUCING THE NUMBER OF TOPICS AND ENLARGING THE DATASET, which will allow the algorithm to have MORE DATA AND LESS AMBIGUITY REGARDING TOPICS - so the ambiguity is treated as a quantity the design can lower rather than a feature of the object</term>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t13" type="limitations" ana="#limitations[0]" resp="#tagger">
        <label>Figurative instances still have to be extracted from the plays by hand; no automatic procedure isolates them.</label>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t14" type="limitations" ana="#limitations[1]" resp="#tagger">
        <label>The trained topic classifier reaches only 72.94% accuracy and is presented as needing further fine-tuning.</label>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t15" type="limitations" ana="#limitations[2]" resp="#tagger">
        <label>The number of topics and their titles are partly subjective researcher decisions.</label>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t16" type="limitations" ana="#limitations[3]" resp="#tagger">
        <label>Inter-annotator agreement on the manual classification was not measured; it is deferred to a future study.</label>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t17" type="limitations" ana="#limitations[4]" resp="#tagger">
        <label>The automatic classification is described as still needing improvement (fewer topics, larger dataset, other language models).</label>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t18" type="limitations" ana="#limitations[5]" resp="#tagger">
        <label>Extending the network to a directed one (tenor to vehicle) is blocked by cases where the tenor is only implied.</label>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t19" type="limitations" ana="#limitations[6]" resp="#tagger">
        <label>Low centrality in the network does not establish that a topic is unimportant, so the method cannot by itself dismiss an interpretation.</label>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t20" type="limitations" ana="#limitations[7]" resp="#tagger" cert="1.0">
        <label>A modelled topic may group words that human readers see as belonging to different topics.</label>
        <note>what this limits: any topic produced by the modelling step here: the authors note that discerning the titles is a manual task requiring subjective interpretation, and that a topic&#x27;s word list may include words human interpreters recognise as belonging to distinct topics, a problem named word intrusion</note>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t21" type="transfers" ana="#transfers[0]" resp="#tagger" cert="0.85">
        <label>the computational apparatus of natural language processing - tokenisation, distributional similarity, network construction - applied to figurative language in a corpus of liturgical poetry</label>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t22" type="controls" ana="#controls[0]" resp="#tagger" cert="0.85">
        <label>randomly splitting the dataset into a 66% training set and a 33% test set</label>
        <note>eliminates: that the classifier&#x27;s apparent accuracy comes from memorising the data it was trained on · outcome: held</note>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t23" type="concepts" ana="#concepts[5]" resp="#paper">
        <label>tenor</label>
        <term>tenor</term>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t24" type="concepts" ana="#concepts[6]" resp="#paper">
        <label>vehicle</label>
        <term>vehicle</term>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t25" type="concepts" ana="#concepts[7]" resp="#paper">
        <label>word intrusion</label>
        <term>word intrusion</term>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t26" type="concepts" ana="#concepts[8]" resp="#paper">
        <label>topic modeling</label>
        <term>topic modeling</term>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t27" type="senses" ana="#senses[0]" resp="#tagger" cert="1.0">
        <label>a topic title - SUPPLIED BY A READER, AND DEFENDED AS SUCH: the paper states that titling the groups of words is a manual task requiring active and often subjective interpretation, and then argues that this subjectivity does not undermine the method but shows that various classifications are possible depending on the researcher&#x27;s orientation - a position, not an oversight</label>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t28" type="senses" ana="#senses[1]" resp="#tagger" cert="1.0">
        <label>tenor and vehicle - GLOSSED IN ONE PARENTHESIS: the underlying subject and the transported meaning, taken from I. A. Richards; the paper states the connection between them is not a necessary one but rests on substitution, which is why the network it builds is undirected</label>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t29" type="senses" ana="#senses[2]" resp="#tagger" cert="1.0">
        <label>word intrusion - THE NAMED FAILURE MODE, ATTRIBUTED: a list of words the model calls one topic may contain words that human interpreters place in distinct topics, which the paper attributes to Chang et al. rather than presenting as its own discovery</label>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t30" type="claim_relations" ana="#claim_relations[0]" resp="#tagger">
        <label>builds-on → Moretti&#x27;s notion of &quot;operationalization&quot; (Moretti 2013), after Bridgman&#x27;s The Logic of Modern Physics</label>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t31" type="claim_relations" ana="#claim_relations[1]" resp="#tagger">
        <label>builds-on → Lakoff and Johnson (2003), Metaphors We Live By — metaphorical entailment</label>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t32" type="claim_relations" ana="#claim_relations[2]" resp="#tagger">
        <label>contradicts → Rhody (2012), topic modeling of figurative language</label>
        <note>differs in: BYPASSING AGAINST SOLVING. &#x27;Quotation withheld&#x27; The method is granted as usable; what is contested is whether it addresses the problem.</note>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t33" type="claim_relations" ana="#claim_relations[3]" resp="#tagger">
        <label>contradicts → Finkelstein (2020), &quot;Amicitia and beneficia in Timon of Athens&quot;</label>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t34" type="concept_edges" ana="#concept_edges[0]" resp="#tagger">
        <label>tenor —distinct-from→ vehicle</label>
        <term>tenor</term>
        <term>vehicle</term>
        <note>differs in: which side of the metaphor each names: the underlying subject, or the transported meaning · strength of the relation: 1</note>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t35" type="content_tags" ana="#content_tags.findings[0]" resp="#tagger">
        <label>Figurative language is harder to analyse than literal language because of the very structure of figurative expressions.</label>
        <note>about this quote: QUOTE RECOVERED FROM THE PAPER&#x27;S OWN CHARACTERS. The recorded quote was LETTERS-ONLY IDENTICAL to this stretch of the staged text but not an exact substring of it - spaces, hyphens, ligature glyphs or mathematical operators differed, so it could never anchor. This is the file&#x27;s own text between the same two letters: not produced by a rule about the damage, copied out of the file. Round-trip checked (the recovered span is letters-only equal to the original) and length-checked. The text as originally recorded is kept in evidence as extracted.</note>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t36" type="content_tags" ana="#content_tags.findings[1]" resp="#tagger">
        <label>The method combines two techniques rather than one: topic classification to label the figures, network analysis to relate them.</label>
        <note>about this quote: QUOTE RECOVERED FROM THE PAPER&#x27;S OWN CHARACTERS. The recorded quote was LETTERS-ONLY IDENTICAL to this stretch of the staged text but not an exact substring of it - spaces, hyphens, ligature glyphs or mathematical operators differed, so it could never anchor. This is the file&#x27;s own text between the same two letters: not produced by a rule about the damage, copied out of the file. Round-trip checked (the recovered span is letters-only equal to the original) and length-checked. The text as originally recorded is kept in evidence as extracted.</note>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t37" type="content_tags" ana="#content_tags.findings[2]" resp="#tagger">
        <label>Figurative analysis is not confined to literary criticism; it matters wherever the object of study is text.</label>
        <note>about this quote: QUOTE RECOVERED FROM THE PAPER&#x27;S OWN CHARACTERS. The recorded quote was LETTERS-ONLY IDENTICAL to this stretch of the staged text but not an exact substring of it - spaces, hyphens, ligature glyphs or mathematical operators differed, so it could never anchor. This is the file&#x27;s own text between the same two letters: not produced by a rule about the damage, copied out of the file. Round-trip checked (the recovered span is letters-only equal to the original) and length-checked. The text as originally recorded is kept in evidence as extracted.</note>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
      <span xml:id="t38" type="content_tags" ana="#content_tags.findings[3]" resp="#tagger">
        <label>The method is judged by whether it can support a close reading, not only by whether it runs.</label>
        <note>about this quote: QUOTE RECOVERED FROM THE PAPER&#x27;S OWN CHARACTERS. The recorded quote was LETTERS-ONLY IDENTICAL to this stretch of the staged text but not an exact substring of it - spaces, hyphens, ligature glyphs or mathematical operators differed, so it could never anchor. This is the file&#x27;s own text between the same two letters: not produced by a rule about the damage, copied out of the file. Round-trip checked (the recovered span is letters-only equal to the original) and length-checked. The text as originally recorded is kept in evidence as extracted.</note>
        <note type="quotation-withheld">Quotation withheld because this record is not cleared for redistribution.</note>
      </span>
    </spanGrp>
  </standOff>
  <text>
    <body>
      <p><note type="rights">The full text of this article is not redistributable, so it is
      not included here. The evidence sentences are withheld with it; the tag layer and its
      grounding status remain. The article itself is at .</note></p>
    </body>
  </text>
</TEI>
