<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>Some Argumentative Uses of the Rhetorical Figure of Antithesis in Environmental Science Policy Articles</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Nancy L. Green</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>University of North Carolina Greensboro</institution>
          ,
          <addr-line>Greensboro, NC</addr-line>
          ,
          <country country="US">USA</country>
        </aff>
      </contrib-group>
      <pub-date>
        <year>2021</year>
      </pub-date>
      <fpage>2</fpage>
      <lpage>3</lpage>
      <abstract>
        <p>This paper presents examples of the figure of antithesis in five environmental science policy journal articles and describes their argumentative functions. The examples include not just contrasting words (antonyms) but contrasting ideas. Many of the examples were found within instances of the RST relations of Contrast, Concession, and Antithesis. Other examples fit argument patterns such as argument from opposites. Collecting and analyzing such examples may contribute to future techniques for automatic rhetorical figure detection and argument mining.</p>
      </abstract>
      <kwd-group>
        <kwd>1 Rhetorical figure of antithesis</kwd>
        <kwd>Rhetorical Structure Theory</kwd>
        <kwd>Argumentation</kwd>
        <kwd>Environmental science policy</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>1. Introduction</title>
      <p>
        We are interested in exploring the argumentative uses of the rhetorical figure of antithesis in
environmental science policy writings. This paper presents examples of antithesis from five articles
written by experts on the subjects of food security, climate change, and water resource management in
three journals [
        <xref ref-type="bibr" rid="ref13 ref15 ref16 ref5">5, 13, 15, 16, 22</xref>
        ]. Each article involves a conflict between the perspectives of biologists
and engineers, i.e., the view that nature should be preserved versus the view that it can be engineered
to solve human problems. By studying examples of antithesis in the corpus, it is possible to better
understand the role of antithesis in argumentation in this genre. Eventually, this may contribute to
computational techniques for rhetorical figure detection and argument mining [
        <xref ref-type="bibr" rid="ref11 ref17">11, 17</xref>
        ].
      </p>
      <p>
        According to Fahnestock [4, p. 232], “A perfect antithesis takes pairs of terms opposed as contraries,
contradictories, or correlatives and puts them in parallel phrases.” Contraries, also known as polar or
scalar opposites, are gradable terms such as hot/cold. Contradictories, also known as binary opposites
or complementaries, are nongradable opposites such as dead/alive or predicates and their negation, such
as red/not red. Correlatives refer to opposite roles in a relationship such as buy/sell, father/son. Other
types of lexical opposites include reversives (reversed actions or events, such as build/destroy,
start/finish) and directional opposites, opposites referring to a horizontal axis (such as front/back), a
vertical axis (such as head/feet), a lateral axis (such as left/right), or a temporal axis (such as
past/present/future) [
        <xref ref-type="bibr" rid="ref14">14</xref>
        ]. Parallelism may consist of repeated words, similar grammatical structures,
and/or acoustic similarity [
        <xref ref-type="bibr" rid="ref3">3</xref>
        ].
      </p>
      <p>On the other hand, we have adopted a broader definition of antithesis, given in Silva Rhetorica
(http://rhetoric.byu.edu/) as the “juxtaposition of contrasting words or ideas (often, although not always,
in parallel structure).” Previous computational treatments of antithesis considered only contrasting
words, i.e. lexical opposites. By widening the scope to include contrasting ideas, not just lexical
opposites or negated predicates, it is possible to get a wider appreciation of how antithesis is used.</p>
      <p>
        Furthermore, we discovered that although some longer examples did not exhibit syntactic parallelism,
they did play a role in the discourse coherence relation of Antithesis in Rhetorical Structure Theory
(RST) [
        <xref ref-type="bibr" rid="ref18">18</xref>
        ]. This led to the realization that many of the other uses of antithesis in the corpus played a
role in two discourse coherence relations of RST related to Antithesis, Contrast and Concession. The
next section provides some background on RST and its definition of those discourse coherence
relations. Subsequent sections present uses of (the rhetorical figure of) antithesis in the corpus, a survey
of related work, and a discussion of implications and conclusions.
      </p>
    </sec>
    <sec id="sec-2">
      <title>2. RST: Contrast, Antithesis and Concession</title>
      <p>
        Rhetorical Structure Theory (RST) [
        <xref ref-type="bibr" rid="ref18">18</xref>
        ] models adjacent spans of coherent text in a hierarchical
structure of coherence relations. The leaves of the tree are elementary discourse units (EDUs), i.e., a
sentence of one clause, or each clause of a complex sentence, not including complement clauses or
relative clauses [21]. In relations such as Antithesis and Concession, the more important text span is
labeled as the nucleus (N), while the span labeled as satellite (S) provides support for the nucleus. In
certain relations such as Contrast, instead of S and N, there are two nuclei of equal importance. Note
that discourse connectives such as ‘although’ and ‘but’ are not sufficient for identification of coherence
relations. The following definitions and criteria for distinguishing these three relations are given in
[21].2
      </p>
      <p>In Contrast (p. 20), the contents of the two nuclei are “comparable yet not identical”. The effect of
Contrast is that the reader (R) “recognizes the comparability and the differences” between the contents
of the two nuclei. Although similar to the relations of Concession and Antithesis, “if neither segment
is deemed more important than the other, then Contrast is to be chosen (p. 23).”</p>
      <p>In Concession (p. 11), the writer (W) “concedes S and implicitly confirms that S and N are usually
not compatible; in the current instance, however, they are compatible, and N is being emphasized.” The
effect is that “R’s positive regard for N is increased.” (In RST definitions, the term ‘regard’
encompasses attitude and belief.)</p>
      <p>In Antithesis (p. 10), W “identifies with” the nucleus, and “the contents of N and S are not
compatible. Due to the incompatibility, one cannot have equal regard for N and S.” The effect is that
“R’s positive regard for N is increased”.</p>
      <p>On distinguishing Concession and Antithesis (p. 24), Concession “generally involves a violated or
failed expectation [in N] … One important difference between Antithesis and Concession is that the
claim which is represented by S is dismissed in Antithesis, but is acknowledged in Concession.”</p>
    </sec>
    <sec id="sec-3">
      <title>3. Antithesis Used in RST Concession Relation</title>
      <p>
        In the following examples of Concession from the corpus, one of the pair of opposites of a rhetorical
figure of antithesis occurs within S and the other within N. In the examples in this and later sections, S
and N are distinguished using square brackets, and underlining is added to identify opposing elements
of the rhetorical figure of antithesis. In examples containing more than one pair of opposites, e.g. in (2),
different styles of underlining are used to identify each pair. None of the following examples involve
lexical opposites; two of the examples involve negated predicates. N and S exhibit a high degree of
syntactic parallelism, and opposites tend to occur at the end of the phrase. Per the definition of
Concession, S conveys what is conceded by W, in order to increase R’s acceptance of N. The rhetorical
figure of antithesis in this context has the following role: the use of opposite concepts in N and S
emphasizes the contrast between N and S.
(1) [Moreover, certain marine species are vulnerable to acidification]N [whereas others are relatively
resilient.]S [
        <xref ref-type="bibr" rid="ref15">15</xref>
        ]
(2) [Most carbon dioxide remains in the air,]S [but as much as 25 percent is absorbed by the world’s
oceans…]N [
        <xref ref-type="bibr" rid="ref15">15</xref>
        ]
2 Complete definitions of all RST relations are given at http://www.sfu.ca/rst/01intro/definitions.html.
(3) [And although it might seem creepy …]S [doing so is, in fact, no big deal, proponents say]N [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ]
(4) [It is theoretically possible]S [but hugely improbable]N [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ]
(5) [Solar geoengineering is not a substitute…]N [It is –at best—a supplement.]S [
        <xref ref-type="bibr" rid="ref16">16</xref>
        ]
(6) [These concerns do suggest …]S; [they do not, individually or collectively, amount to …]N [
        <xref ref-type="bibr" rid="ref16">16</xref>
        ]
(7) [The insect research is a meaningful step toward sustaining the river for habitat as well as for
humans.]S [It also runs straight into a core conflict between science and Colorado River policy]N
3 [
        <xref ref-type="bibr" rid="ref13">13</xref>
        ]
      </p>
    </sec>
    <sec id="sec-4">
      <title>4. Antithesis Used in RST Antithesis Relation</title>
      <p>
        In the following examples of RST Antithesis from the corpus, one of the pair of opposites of a
rhetorical figure of antithesis occurs within S and the other within N. Note that, except in one case
-annual/perennial in (12) -- none of the examples involve lexical opposites or negated predicates. Also,
S and N do not exhibit syntactic parallelism. Per the definition of Antithesis, S and N are incompatible,
thereby increasing R’s acceptance of N. As was the role of the figure of antithesis in the examples of
Concession, the use of opposite concepts in N and S emphasizes the contrast between them.
(8) [But as climate change reached the top of the environmental agenda …, discussion of solar
geoengineering went quiet.]S [However, there are now signs of rapid change in the politics of solar
geoengineering.]N [
        <xref ref-type="bibr" rid="ref16">16</xref>
        ]
(9) [Yet opponents maintain that because the wholesale swapping or alteration of entire packages of
genes is a natural process …, it tends to produce few scary surprises today.]S [Changing a single
gene, on the other hand, might turn out to be a more subversive action, with unexplained ripple
effects, including the production of new proteins that might be toxins or allergens.]N [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ]
(10) [Yet here they are scrabbling in the dirt at a small organization privately funded by citizens’
donations.]N [If this work is so important, where are the powerful institutions, the high-tech
equipment, the labs full of students?]S [22]
(11) [Allocation of funds on this scale becomes dependent on an average of the opinions of numerous
bureaucrats, lawmakers, administrators and committees.]S [This is a far cry from the privately
wealthy gentlemen of science of the 17th to 19th centuries…]N [22]
(12) [Annuals die each year and must be replanted.]S [Perennials can stay green year-round
(evergreens), drop their leaves and go dormant during the winter (other trees) or die back to the
ground, surviving as underground stems (perennial herbs).]N [22]
      </p>
    </sec>
    <sec id="sec-5">
      <title>5. Antithesis Used in RST Contrast Relation</title>
      <p>In the following examples of Contrast from the corpus, one of the pair of opposites of a rhetorical
figure of antithesis occurs within each nucleus (N). Note that, except in a few cases (proponents/critics,
rise/fall, low/high), none of the other examples involve lexical opposites or negated predicates. Also,
the two N clauses exhibit a high degree of syntactic parallelism. Per the definition of Contrast, the effect
is that R recognizes “the comparability and differences” between the contents of the nuclei. However,
3 The pairs habitat/humans in S and science/policy in N play a different role than the role played by meaningful step/core conflict in
Concession. These two pairs represent two sides of a debate or conflict, similar to the use illustrated in (21) and (22).
unlike the case with Concession and Antithesis, the rhetorical figure of antithesis contributes to the
argumentative meaning in several ways as described below.</p>
      <p>
        In (13) and (14), antithesis characterizes opposing sides of a debate or conflict. In (15) and (16),
antithesis conveys a type of causal inference; the difference between two comparable situations is seen
to be responsible for their different outcomes.4 In (17), antithesis conveys another type of causal
inference; by comparing two extremes, each having negative consequences, it is suggested that a
situation between the two extremes is desirable.
(13) [Proponents of genetically modified crops say the technology is the only way to feed a warming,
increasingly populous world.]N [Critics say we tamper with nature at our peril.]N [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ]
(14) [scientists want the flexibility to experiment,]N [whereas power and water managers want
stability.]N [
        <xref ref-type="bibr" rid="ref13">13</xref>
        ]
(15) [Normally, … effectively sequestering those compounds from uptake by photosynthetic
organisms;]N [under acidified conditions, however, hydroxide and carbonate remain as free metals
that are bioavailable.]N [
        <xref ref-type="bibr" rid="ref15">15</xref>
        ]
(16) [Native prairies often remain productive even after decades of harvesting and removing hay.]N
[By contrast, adjacent prairies plowed up for wheat quickly degraded …]N [22]
(17) A balance of trace metals … is crucial. [If trace-metal concentrations fall too low, photosynthesis
falters;]N [if they rise too high, the excess of metal may prove toxic.]N [
        <xref ref-type="bibr" rid="ref15">15</xref>
        ]
      </p>
    </sec>
    <sec id="sec-6">
      <title>6. Other Argumentative Uses of Antithesis</title>
      <p>A number of examples of the rhetorical figure of antithesis in the corpus do not involve RST relations
since they occur within a single EDU. The following use of antithesis conveys a sort of argument from
negative consequences [23], i.e., that if X causes Y and Y is undesirable, then X is undesirable.
(Increasing influxes of CO2 have undesirable results, so increasing influxes of CO2 are undesirable;
changing something has undesirable results so changing something is undesirable.) In addition to the
argument from negative consequences, (18) conveys a causal argument from correlation [23]: the
correlation of the second pair of opposites, increase/decrease, implies a causal relationship between
them.
(18) Increasing influxes of CO2 cause a decrease in pH, which results in an increase in</p>
      <p>
        H+ and thus a decrease in hydroxide and carbonate ions in most surface waters. [
        <xref ref-type="bibr" rid="ref15">15</xref>
        ]
(19) “We change something we can control, and then two things we can’t control very quickly change,”
says geologist Ted Melis … [
        <xref ref-type="bibr" rid="ref13">13</xref>
        ]
The following use of antithesis conveys an argument from opposites [23]: if the opposite of X has a
property P, then X has the opposite of property P. In particular, if water released from Lake Powell
causes stress to endangered fish, then the natural state of the Colorado river does not cause stress to
those fish. Note that while clear/silty are lexical opposites, 48-degree water/80-degree flows are not.
(20) The clear, 48-degree water, released from the depths of Lake Powell, stressed endangered fish,
which were adapted to silty, 80-degree flows. [
        <xref ref-type="bibr" rid="ref13">13</xref>
        ]
The following use of antithesis conveys opposing sides of debate or conflict. World knowledge would
be necessary to recognize the opposition in the pair mainstream research/nonprofits.
4 This is similar to use of antithesis to convey “the logic of single-difference experimental design” [4, p. 233]. The difference is that these
examples describe naturally occurring situations rather than human-designed experiments.
(21) A second difference between mainstream research and nonprofits is that funding for universities
and experiment stations increasingly comes from competitive grants. [22]
(22) He argues that the benefits of GM crops greatly outweigh the health risks, which so far remain
theoretical. [
        <xref ref-type="bibr" rid="ref16">16</xref>
        ]
In the following example, use of three pairs of lexical opposites (small/large, short/long, high/low)
emphasizes the differences between lexically opposite types of crops (annual/perennial).
(23) Plant domestication has resulted in small, short-lived, high-yield annual crops and longer-lived,
larger [low-yield] perennial crops. [22]
Lastly, use of the opposites powerful/vulnerable lends emphasis to a prolepsis-as-presage [
        <xref ref-type="bibr" rid="ref19">19</xref>
        ] argument
against failing to protect people from the consequences of climate change.
(24) these changes would have their most powerful impact on the world’s most vulnerable people …
[
        <xref ref-type="bibr" rid="ref16">16</xref>
        ]
      </p>
    </sec>
    <sec id="sec-7">
      <title>7. Related Work</title>
      <p>
        On-line collections of literary uses of antithesis and other figures can be found on Silva Rhetorica
(http://rhetoric.byu.edu/) and RhetFig (https://artsresearch.uwaterloo.ca/chiastic/ display/). In [
        <xref ref-type="bibr" rid="ref3 ref4">3, 4</xref>
        ]
Fahnestock discussed how antithesis was used in historical scientific works to generate hypotheses
about opposites of known phenomena and as a rhetorical device to epitomize arguments based on
single-difference experimental design.
      </p>
      <p>Mitrović et al. [20] noted that, despite the different focus of RST and the rhetorical tradition, there
are some similarities between the figure of antithesis and the RST relation of Antithesis. In their
research on automatic detection of irony, simile, and oxymoron, rhetorical figures that like antithesis
involve meaning, they made use of a Serbian WordNet ontology as a knowledge source.</p>
      <p>
        Azar [
        <xref ref-type="bibr" rid="ref1">1</xref>
        ] described the use of several RST relations including Antithesis and Concession for
modeling arguments. Green [
        <xref ref-type="bibr" rid="ref6">6</xref>
        ] attempted to represent discourse structure and argument structure in a
single model combining RST relations, including Concession and Antithesis, with certain causal
argument schemes used in biology. However, Green later argued for modeling argument structure
independently of RST discourse structure [
        <xref ref-type="bibr" rid="ref7">7</xref>
        ]. In [
        <xref ref-type="bibr" rid="ref8">8</xref>
        ], Green analyzed the argument structure and
discourse structure of a scientific research paper and showed that they were not in one-to-one
correspondence. In [
        <xref ref-type="bibr" rid="ref9">9</xref>
        ], Green analyzed the argument schemes and a variety of rhetorical devices in
two of the papers from the corpus used in the current study, but did not specifically examine the
argumentative use of antithesis.
      </p>
      <p>
        Lawrence et al. [
        <xref ref-type="bibr" rid="ref17">17</xref>
        ] explored the role of the figure of antithesis in argumentation by developing an
algorithm for detecting antithesis in a dialogic corpus whose arguments had been annotated. The
algorithm defined antithesis solely in terms of antonyms given in WordNet (wordnet.princeton.edu).
They found that cases of antithesis detected by the algorithm were often used in turns annotated as
rebuttals in their corpus. That use is consistent with use of antithesis in the context of the RST relation
of Antithesis described in this paper, but does not cover other uses described in this paper.
      </p>
      <p>
        Green and Crotts [
        <xref ref-type="bibr" rid="ref10">10</xref>
        ] explored automatic detection of antithesis in a dataset of quotations that had
been annotated previously for chiasmus. The motivation for using that dataset was the hope that it
would contain many examples of antithesis since it had been noted that antithesis often occurs in
combination with chiasmus [
        <xref ref-type="bibr" rid="ref12">12</xref>
        ]. Green and Crotts developed an antithesis detection algorithm using a
broader definition of antithesis than Lawrence et al.’s. In addition to WordNet, the algorithm used
ConceptNet (conceptnet.io), and searched not only for antonyms but for antonyms of synonyms, as well
as for certain simple forms of negated predicates. However, there were many cases of antithesis in the
dataset that did not involve lexical opposites. In addition, although the dataset was indeed a rich source
of antithesis examples, it is not representative of the genre studied in this paper.
      </p>
    </sec>
    <sec id="sec-8">
      <title>8. Discussion</title>
      <p>The current study of contemporary environmental science policy articles has found that many cases
of the rhetorical figure of antithesis have argumentative force from the RST relations of Concession
and Antithesis. In addition, the study identified a number of other argumentative uses of antithesis.
Thus, detection of antithesis may play a future role in argument mining. However, many cases of
antithesis illustrated here do not involve lexical opposites or negated predicates. The problem of
automatic recognition of antithesis requires a deeper level of semantic/pragmatic interpretation than is
possible with available on-line lexical resources. A possible stopgap is to construct a domain ontology
from a corpus for a particular genre (e.g. science policy arguments) and topic (e.g. climate change) as
a resource for antithesis detection. Also, it may be helpful to examine features that have been used for
parsing RST structures in text.</p>
    </sec>
    <sec id="sec-9">
      <title>9. References</title>
      <p>[20] Mitrović J, O’Reilly C, Mladenović M, Handschuh S. Ontological Representations of Rhetorical</p>
      <p>Figures for Argument Mining’. Argument and Computation 2017; 8(3): 267-287.
[21] Stede M, Taboada M, Das D. Annotation Guidelines for Rhetorical Structure, Version 1.0, March
2017. Downloaded 4/20/21 from: https://www.sfu.ca/~mtaboada/docs/research/
RST_Annotation_Guidelines.pdf.
[22] Van Tassel DL, DeHaan, LR. Wild Plants to the Rescue. American Scientist 2013; 103(3):
21825.
[23] Walton D, Reed C, Macagno F. Argumentation Schemes. Cambridge University Press; 2008.</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          [1]
          <string-name>
            <surname>Azar</surname>
            <given-names>M. Argumentative</given-names>
          </string-name>
          <article-title>Text as Rhetorical Structure: An Application of Rhetorical Structure Theory</article-title>
          .
          <source>Argumentation</source>
          <volume>13</volume>
          :
          <fpage>97</fpage>
          -
          <lpage>114</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          [2]
          <string-name>
            <surname>Fahnestock</surname>
            <given-names>J. Rhetorical</given-names>
          </string-name>
          <article-title>Figures in Science</article-title>
          . Oxford University Press.
          <year>1999</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          [3]
          <string-name>
            <surname>Fahnestock</surname>
            <given-names>J</given-names>
          </string-name>
          .
          <source>Figures of Argument. Informal Logic</source>
          <year>2004</year>
          ;
          <volume>24</volume>
          (
          <issue>2</issue>
          ):
          <fpage>115</fpage>
          -
          <lpage>135</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          [4]
          <string-name>
            <surname>Fahnestock</surname>
            <given-names>J.</given-names>
          </string-name>
          <string-name>
            <surname>Rhetorical</surname>
          </string-name>
          <article-title>Style: The Uses of Language in Persuasion</article-title>
          . Oxford University Press.
          <year>2011</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          [5]
          <string-name>
            <surname>Freedman</surname>
          </string-name>
          , DH.
          <source>The Truth about Genetically Modified Food: Are Engineered Foods Evil? Scientific American</source>
          <volume>00368733</volume>
          ,
          <issue>Sep2013</issue>
          ;
          <volume>309</volume>
          (
          <issue>3</issue>
          ).
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          [6]
          <string-name>
            <surname>Green</surname>
            <given-names>NL</given-names>
          </string-name>
          .
          <article-title>Representation of Argumentation in Text with Rhetorical Structure Theory</article-title>
          .
          <source>Argumentation</source>
          <year>2010</year>
          ;
          <volume>24</volume>
          (
          <issue>2</issue>
          ):
          <fpage>181</fpage>
          -
          <lpage>196</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          [7]
          <string-name>
            <surname>Green</surname>
            <given-names>NL</given-names>
          </string-name>
          .
          <article-title>Annotating evidence-based argumentation in biomedical text</article-title>
          .
          <source>2015 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)</source>
          , Washington, DC, USA,
          <fpage>9</fpage>
          -
          <lpage>12</lpage>
          November
          <year>2015</year>
          . Piscataway, NJ: IEEE,
          <fpage>922</fpage>
          -
          <lpage>929</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          [8]
          <string-name>
            <surname>Green</surname>
            <given-names>NL</given-names>
          </string-name>
          .
          <article-title>Towards Mining Scientific Discourse Using Argumentation Schemes</article-title>
          .
          <source>Argument and Computation</source>
          <year>2018</year>
          ;
          <volume>9</volume>
          (
          <issue>2</issue>
          ):
          <fpage>121</fpage>
          -
          <lpage>135</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          [9]
          <string-name>
            <surname>Green</surname>
            <given-names>NL</given-names>
          </string-name>
          .
          <source>Recognizing Rhetoric in Science Policy Arguments. Argument and Computation</source>
          <year>2020</year>
          ;
          <volume>11</volume>
          (
          <issue>3</issue>
          ):
          <fpage>257</fpage>
          -
          <lpage>268</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref10">
        <mixed-citation>
          [10]
          <string-name>
            <surname>Green</surname>
            <given-names>NL</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Crotts</surname>
            <given-names>LJ</given-names>
          </string-name>
          .
          <source>Towards Automatic Detection of Antithesis. Computational Models of Natural Argument (CMNA</source>
          <year>2020</year>
          ),
          <fpage>69</fpage>
          -
          <lpage>73</lpage>
          . CEUR-WS.org/Vol-
          <volume>2669</volume>
          /paper9.pdf.
        </mixed-citation>
      </ref>
      <ref id="ref11">
        <mixed-citation>
          [11]
          <string-name>
            <surname>Harris</surname>
            <given-names>RA</given-names>
          </string-name>
          , Di Marco C. Introduction: Rhetorical Figures, Arguments, Computation.
          <source>Argument and Computation</source>
          <year>2017</year>
          ;
          <volume>8</volume>
          (
          <issue>3</issue>
          ):
          <fpage>1</fpage>
          -
          <lpage>21</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref12">
        <mixed-citation>
          [12]
          <string-name>
            <surname>Harris</surname>
            <given-names>RA</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Di Marco</surname>
            <given-names>C</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Ruan</surname>
            <given-names>S</given-names>
          </string-name>
          ,
          <string-name>
            <surname>O'Reilly</surname>
            <given-names>C.</given-names>
          </string-name>
          <article-title>An Annotation Scheme for Rhetorical Figures Argument</article-title>
          and
          <source>Computation</source>
          <year>2018</year>
          ;
          <volume>9</volume>
          (
          <issue>2</issue>
          ):
          <fpage>155</fpage>
          -
          <lpage>175</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref13">
        <mixed-citation>
          [13]
          <string-name>
            <surname>Hansman</surname>
            <given-names>H. Reengineering</given-names>
          </string-name>
          <article-title>the Colorado River</article-title>
          .
          <source>Scientific American</source>
          <volume>00368733</volume>
          ,
          <issue>2</issue>
          /1/2019;
          <volume>320</volume>
          (
          <issue>2</issue>
          ).
        </mixed-citation>
      </ref>
      <ref id="ref14">
        <mixed-citation>
          [14]
          <string-name>
            <surname>Ježek</surname>
            <given-names>E. The Lexicon: An</given-names>
          </string-name>
          <string-name>
            <surname>Introduction</surname>
          </string-name>
          . Oxford University Press,
          <year>2016</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref15">
        <mixed-citation>
          [15]
          <string-name>
            <surname>Johnson</surname>
            <given-names>A</given-names>
          </string-name>
          ,
          <string-name>
            <surname>White</surname>
            <given-names>ND</given-names>
          </string-name>
          .
          <source>Ocean Acidification: The Other Climate Change Issue. American Scientist</source>
          <year>2014</year>
          ;
          <volume>102</volume>
          (
          <string-name>
            <surname>January-February</surname>
          </string-name>
          ):
          <fpage>60</fpage>
          -
          <lpage>63</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref16">
        <mixed-citation>
          [16]
          <string-name>
            <surname>Keith</surname>
            <given-names>DW</given-names>
          </string-name>
          .
          <article-title>Toward a Responsible Solar Geoengineering Research Program</article-title>
          .
          <source>Issues in Science and Technology Spring</source>
          <year>2017</year>
          ;
          <volume>33</volume>
          (
          <issue>3</issue>
          ).
        </mixed-citation>
      </ref>
      <ref id="ref17">
        <mixed-citation>
          [17]
          <string-name>
            <surname>Lawrence</surname>
            <given-names>J</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Visser</surname>
            <given-names>J</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Reed</surname>
            <given-names>C</given-names>
          </string-name>
          .
          <article-title>Harnessing rhetorical figures for argument mining</article-title>
          .
          <source>Argument and Computation</source>
          <year>2017</year>
          ;
          <volume>8</volume>
          (
          <issue>3</issue>
          ):
          <fpage>289</fpage>
          -
          <lpage>310</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref18">
        <mixed-citation>
          [18]
          <string-name>
            <surname>Mann</surname>
            <given-names>WC</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Thompson</surname>
            <given-names>SA</given-names>
          </string-name>
          .
          <article-title>Rhetorical Structure Theory: Toward a Functional Theory of Text Organization</article-title>
          .
          <source>Text</source>
          <year>1988</year>
          ;
          <volume>8</volume>
          :
          <fpage>243</fpage>
          -
          <lpage>281</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref19">
        <mixed-citation>
          [19]
          <string-name>
            <surname>Mehlenbacher</surname>
            <given-names>AR</given-names>
          </string-name>
          .
          <article-title>Rhetorical figures as argument schemes - The proleptic suite</article-title>
          .
          <source>Argument and Computation</source>
          <year>2017</year>
          ;
          <volume>8</volume>
          (
          <issue>3</issue>
          ):
          <fpage>233</fpage>
          -
          <lpage>252</lpage>
          .
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>