An introduction to argumentation mining for PhD students. This was presented in the SwissUniversities Doctoral Programme, Language & Cognition, in the Module: Linguistic and corpus perspectives on argumentative discourse. The presentation largely follows Chapters 1-4 and Chapter 10 of my book, Argumentation Mining, co-authored with Manfred Stede in the Synthesis Lectures on Human Language Technologies from Morgan & Claypool: https://doi.org/10.2200/S00883ED1V01Y201811HLT040
Topics:
My book w/computational linguist Manfred Stede: Argumentation Mining
What is argumentation?
Argumentation mining: a first look
Argumentative language
Challenges for argumentation mining
Argumentation structures
Corpus annotation
Why study argumentation mining?
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdf
Argumentation mining--an introduction for linguists--Fribourg--2019-09-02
1. Argumentation Mining: An
Introduction for Linguists
Linguistic and corpus perspectives on argumentative
discourse, SwissUniversities Doctoral Programme
Language & Cognition
University of Fribourg, Fribourg, Switzerland
2019-09-02
2. Today’s Lecture
• My book w/computational linguist Manfred
Stede: Argumentation Mining
• What is argumentation?
• Argumentation mining: a first look
• Argumentative language
• Challenges for argumentation mining
• Argumentation structures
• Corpus annotation
• Why study argumentation mining?
3. In Swissbib, search for “Synthesis
Lectures on Human Language
Technologies”:
https://www.swissbib.ch/Search/Res
ults?lng=en&type=ISN&lookfor=194
74040
Main Chapters
1. Introduction
2. Argumentative Language
3. Modeling Arguments
4. Corpus Annotation
5. Finding Claims
6. Finding Supporting and
Objecting Statements
7. Deriving the Structure of
Argumentation
8. Assessing Argumentation
9. Generating Argumentative Text
10. Summary and Perspectives
https://doi.org/10.2200/S00883ED1V01Y201811HLT040
5. Example argument
We really need to tear down that building.
Granted, it will be expensive, but the degree of
asbestos contamination is not tolerable
anymore. Also, it is one of the most ugly
buildings in town!
Stede & Schneider, Argumentation Mining, 1.1, page 6
7. An argument has a central claim
Tweet by @robeastaway
https://twitter.com/robeastaway/status/135838892694839296
Claim: Jaffa Cakes are cakes
Justification: When stale, cakes go hard.
Was official EU ruling.
8. An argument supports the claim
Tweet by @robeastaway
https://twitter.com/robeastaway/status/135838892694839296
Claim: Jaffa Cakes are cakes
Justification: When stale, cakes go hard.
Was official EU ruling.
9. “Argumentation is a verbal, social, and rational
activity aimed at convincing a reasonable critic
of the acceptability of a standpoint by putting
forward a constellation of propositions justifying
or refuting the proposition expressed in the
standpoint.”
van Eemeren and Grootendorst [2004, p. 1], as quoted by
Stede & Schneider, Argumentation Mining, 1.1, page 1
12. Possible applications of argumentation
mining
• Sensemaking
• Practical reasoning
• Argument Retrieval
• Web-Scale Discourse and Debate
• Sentiment Analysis
• Writing Support
• Essay Scoring
• Dialogue Systems
Stede & Schneider, Argumentation Mining, 1.3, pages 8-9 &
10.3, pages 140-143
13. Full-fledged argumentation mining
1. Identify argumentative text
2. Segment the text into argumentative discourse
units (ADUs)
3. Identify the central claim.
4. Identify the role/function of ADUs.
5. Identify relations between ADUs.
6. Build the overall structural representation.
7. Identify the type and the quality of the
argumentation.
Stede & Schneider, Argumentation Mining, 1.2, pages 6-7
14. Example argument
We really need to tear down that building.
Granted, it will be expensive, but the degree of
asbestos contamination is not tolerable
anymore. Also, it is one of the most ugly
buildings in town!
Stede & Schneider, Argumentation Mining, 1.1, page 6
15. Segment into
argumentative discourse units (ADUs)
• [We really need to tear down that building.]0
[Granted, it will be expensive,]1 [but the
degree of asbestos contamination is not
tolerable anymore.]2 [Also, it is one of the
most ugly buildings in town!]3
Stede & Schneider, Argumentation Mining, 1.1, page 6
16. Identify the central claim
We really need to tear down that building.
Granted, it will be expensive, but the degree of
asbestos contamination is not tolerable
anymore. Also, it is one of the most ugly
buildings in town!
17. Identify the role of the other ADUs
• [We really need to tear down that building.]0
[Granted, it will be expensive,]1 [but the
degree of asbestos contamination is not
tolerable anymore.]2 [Also, it is one of the
most ugly buildings in town!]3
Stede & Schneider, Argumentation Mining, 1.1, page 6
18. Role: Anticipating a counterargument
We really need to tear down that building.
Granted, it will be expensive, but the degree of
asbestos contamination is not tolerable
anymore. Also, it is one of the most ugly
buildings in town!
19. Role: Rebutting a counterargument
We really need to tear down that building.
Granted, it will be expensive, but the degree of
asbestos contamination is not tolerable
anymore. Also, it is one of the most ugly
buildings in town!
20. Role: Supporting the central claim
We really need to tear down that building.
Granted, it will be expensive, but the degree of
asbestos contamination is not tolerable
anymore. Also, it is one of the most ugly
buildings in town!
21. Identify the role of the other ADUs
• 0: main claim [We really need to turn down
that building.]
• 1: counterargument [Granted, it will be
expensive,]
• 2: rebuttal [but the degree of asbestos
contamination is not tolerable anymore.]
• 3: support [Also, it is one of the most ugly
buildings in town!]
22. Identify relations between ADUs
• 0: main claim [We really need to turn down that
building.]
• 1: counterargument [Granted, it will be
expensive,] ATTACKS 0
• 2: rebuttal [but the degree of asbestos
contamination is not tolerable anymore.]
UNDERCUTS THE COUNTERARGUMENT ATTACK
• 3: support [Also, it is one of the most ugly
buildings in town!] SUPPORTS 0
24. “Argumentation is a verbal, social, and rational
activity aimed at convincing a reasonable critic
of the acceptability of a standpoint by putting
forward a constellation of propositions justifying
or refuting the proposition expressed in the
standpoint.”
van Eemeren and Grootendorst [2004, p. 1], as quoted by
Stede & Schneider, Argumentation Mining, 1.1, page 1
25. Full-fledged argumentation mining
1. Identify argumentative text
2. Segment the text into argumentative discourse
units (ADUs)
3. Identify the central claim.
4. Identify the role/function of ADUs.
5. Identify relations between ADUs
6. Build the overall structural representation.
7. Identify the type and the quality of the
argumentation.
Stede & Schneider, Argumentation Mining, 1.2, pages 6-7
27. Subjectivity
• ‘Objective’ statements are a matter of
intersubjective agreement or disagreement:
– There is a cat on the mat.
– Winston Churchill came to office in 1940.
• Private states are not
Stede & Schneider, Argumentation Mining, 2.1, page 11-12,
following [Wiebe et al., 2005], [Quirk et al., 1985]
28. Various types of private states
• Reveal an emotion: Hooray!
• Give an opinion: That’s a really bad wine.
• Make a judgment: You don’t deserve the
prize.
• Make a prognosis: There will be snow
tomorrow.
• Give an estimate or speculation: I guess that’s
a llama over there.
Stede & Schneider, Argumentation Mining, 2.1, page 12
29. Sentiment & Opinion
• Sentiment: Various subjective utterances
(opinion, judgement, emotion)
• Opinion is narrower for us: Subjective
evaluations of some entity.
Stede & Schneider, Argumentation Mining, 2.1, page 12
30. Arguments & Opinions
• Not all opinions are arguments. But if a reason
for an opinion is provided, we can analyze it as
an argument:
“I never enjoyed Wagner’s operas, as they are
so enormously overloaded.”
31. Verifiability
Verifiable: objective assertions (not about personal
feelings or interpretations)
– Verifiable-private: statements that concern the
speaker’s personal state or experience.
– Verifiable-public: public information is sufficient
to verify the statement: no personal state or
experience is involved.
Unverifiable: statements that cannot be proven
with objective evidence.
Park and Cardie [2014], cited by Stede & Schneider, Argumentation
Mining, 2.2, page 13-14
32. Verifiable-private examples
(Park and Cardie [2014])
My son has hypoglycemia.
They flew me to NY in February.
The flight attendant yelled at the passengers.
Park and Cardie [2014], cited by Stede & Schneider, Argumentation
Mining, 2.2, page 13-14
33. Supported differently in
argumentation
• Verifiable-public: can provide evidence:
“I tell you Winston Churchill came to office in 1940. I
saw it on Wikipedia!”
• Verifiable-private: may provide evidence, but this is
optional.
“I have a headache”.
Hearers are not in a position to question this, but the
speaker may go on to explain: “Maybe I had too much
wine last night”.
• Unverifiable: can give a reason as support:
“I don’t like this wine, because it has so much tannin”.
Park and Cardie [2014], cited by Stede & Schneider, Argumentation
Mining, 2.2, page 13-14
34. Supported differently in
argumentation
• Verifiable-public: can provide evidence:
“I tell you Winston Churchill came to office in 1940. I
saw it on Wikipedia!”
• Verifiable-private: may provide evidence, but this is
optional.
“I have a headache”.
Hearers are not in a position to question this, but the
speaker may go on to explain: “Maybe I had too much
wine last night”.
• Unverifiable: can give a reason as support:
“I don’t like this wine, because it has so much tannin”.
Park and Cardie [2014], cited by Stede & Schneider, Argumentation
Mining, 2.2, page 13-14
35. Supported differently in
argumentation
• Verifiable-public: can provide evidence:
“I tell you Winston Churchill came to office in 1940. I
saw it on Wikipedia!”
• Verifiable-private: may provide evidence, but this is
optional.
“I have a headache”.
Hearers are not in a position to question this, but the
speaker may go on to explain: “Maybe I had too much
wine last night”.
• Unverifiable: can give a reason as support:
“I don’t like this wine, because it has so much tannin”.
Park and Cardie [2014], cited by Stede & Schneider, Argumentation
Mining, 2.2, page 13-14
36. Implicit Opinions
• [camera review] The viewfinder is somewhat
dark.
• [hotel review] The rooms turned out to be
small.
Rajendran et al. [2016] Stede & Schneider, Argumentation
Mining, 2.2, page 15
37. Need domain knowledge
(“polar facts” imply an evaluation)
• The viewfinder is somewhat dark (+ A dark
viewfinder is considered bad). I am not in
favor of the camera.
• The rooms turned out to be small (+ A small
room is considered bad). I am not in favor
of the hotel.
Stede & Schneider, Argumentation Mining, 2.2, page 15
38. Hedging
• “There is no clear symptom of diabetes.”
Stede & Schneider, Argumentation Mining, 2.1, page 12 & 2.2 page
14. See also Farkas et al. 2010.
39. Stance classification
Given a topic under debate, determine whether
a contribution is pro or con.
For example:
• “vaccination of children (should/should not)
be mandatory”
• “behavior X (is/is not) acceptable”
• “legislation X (should/should not) be dropped”
Stede & Schneider, Argumentation Mining, 2.2, page 16
41. Classifying speech acts
• Representatives: Speaker commits to the truth of
an assertion.
• Directives: Speaker tries to make addressee
perform some action.
• Expressives: Speaker expresses an emotional
state.
• Declaratives: Speaker changes the state of the
world by means of performing the utterance.
• Commissives: Speaker commits to doing some
action in the future.
Stede & Schneider, Argumentation Mining, 2.2, page 17. See also
Searle [1976]
42. Performative verbs
• You shouldn’t read that book.
• I advise you not to read that book.
• What is your name?
• I ask you to tell me your name.
• Yesterday I met a philosopher.
• I assert that I met a philosopher yesterday.
• Max is the smartest kid in the world.
• I claim that Max is the smartest kid in the world.
• It’s about 15 degrees out there.
• I estimate that it is about 15 degrees out there.
Stede & Schneider, Argumentation Mining, 2.2, pages 16-17. See
also Austin [1975]
43. Components of a speech act
• Locutionary act: producing the linguistic
utterance by speaking, writing, gesturing.
• Illocutionary act: the intention or goal the
speaker has in mind when performing the act.
• Perlocutionary act: the effect that the
performance of the act has on the addressee.
Stede & Schneider, Argumentation Mining, 2.2, page 18. See also
Austin [1975] & Searle [1976]
45. Politeness & indirect speech acts
• “Can you pass the salt?”
• Apologies, with varying levels of indirectness
– I apologize for being late.
– Sorry I’m late.
– I’m a bit late, unfortunately.
– I’m afraid I didn’t quite make it on time.
Stede & Schneider, Argumentation Mining, 2.2, pages 18-19. See
e.g. Ogiermann [2009]
46. Argument, Explanation, and
Justification
• Using airplanes is really a bad idea because they
are among the worst air polluters we have ever
created. (Argumentation/Persuasion)
• An airplane is able to take off because the shape
of the wings produces an upward force when the
air flows across them. (Explanation)
• I need to use airplanes a lot because my job
requires me to be in different parts of the country
every week. (Justification)
Stede & Schneider, Argumentation Mining, 2.2, page 20.
47. Discourse Mode [Smith 2003]
• Narrative
• Description
• Report
• Information
• Argument
Stede & Schneider, Argumentation Mining, 2.2, page 20-22, See also
Smith [2003], Werlich [1975].
48. Discourse Mode [Smith 2003]
• Narrative: The passengers landed in New York in the middle
of the night and then moved on to Hoboken immediately.
• Description: Hundreds of people occupied the square. In
front of them, the speaker was standing on a small podium.
• Report: My sister visited the new exhibition yesterday.
• Information: Krypton is one of the noble gases. It is one of
the rarest elements on earth.
• Argument: The award was given to Paul, but he did not
deserve it. His work is very shallow.
Stede & Schneider, Argumentation Mining, 2.2, pages 20-22, See
also Smith [2003], Werlich [1975].
49. Situation Entity types help distinguish
argumentative vs. non-argumentative text
[Becker 2016]
• State: Armin has brown eyes.
• Event: Bonnie ate three tacos.
• Report: The agency said applications had increased.
• Generic sentence: Scientific papers make arguments.
• Generalizing sentence: Fei travels to India every year.
• Fact: Georg knows that Reza won the competition.
• Proposition: Georg thinks that Reza won the competition.
• Resemblance: Reza looks like he won the competition.
• Question: Why do you torment me so?
• Imperative: Listen to this.
Becker et al. [2016], Stede & Schneider, Argumentation Mining,
2.2, page 22. See also Song et al. [2017]
51. Rhetoric
• Logos: Speakers employ rules of sound
reasoning.
• Ethos: Speakers signal their authority or
credibility (or that of their source)
• Pathos: Speakers seek to communicate
their standpoint in a manner that seeks
to evoke an emotional response
Stede & Schneider, Argumentation Mining, 1.1, page 5, based on
Aristotle.
52. Rhetoric
• Logos: (Logic)
That building needs to be demolished, because it is full of
asbestos, which is known to be hazardous, and there is no way
to stop its diffusion from the different parts of the building.
• Ethos: (Authority/Credibility)
That building needs to be demolished, because it is full of
asbestos, as the report by the university engineers has shown.
• Pathos: (Emotion)
That building needs to be demolished, because it is an
irresponsible source of danger to the health and indeed the
life of our children who spend so many hours in those
poisonous rooms every day!
Stede & Schneider, Argumentation Mining, 1.1, page 5, based on
Aristotle.
53. Rhetoric studies choices
• Close the window! It’s cold.
• It is rather cold in here and I’m already not
feeling so well. Would you be so kind as to
close the window for me?
Stede & Schneider, Argumentation Mining, 2.6, page 23. See
[Fahnestock, 2011] and others for more on these sorts of choices
54. Rhetoric studies choices
• Small changes in word choice, sentence
construction, and passage construction
produce different rhetorical effects.
Stede & Schneider, Argumentation Mining, 2.6, page 23. See
[Fahnestock, 2011] and others for more on these choices
55. Rhetorical figures
• Hyperbole: Phew, the distance from Newark
to Manhattan is a hundred miles!
• Tautology: That nice restaurant was really
great.
Stede & Schneider, Argumentation Mining, 2.6, page 24.
56. Rhetorical figure mining
Rhetorical figures and mining has been studied
extensively by Harris, DiMarco, and colleagues.
See also “Harnessing rhetorical figures for
argument mining”, Lawrence, Visser, and Reed
[2017].
Stede & Schneider, Argumentation Mining, 2.6, page 24. See Harris
& DiMarco [2017], Harris et al. [2018] among others.
57. Rhetorical moves – Swales’ Create a
Research Space
Move 1 Establishing a territory
Step 1 Claiming centrality and/or
Step 2 Making topic generalization(s) and/or
Step 3 Reviewing items of previous research
Move 2 Establishing a niche
Step 1A Counter-claiming or
Step 1B Indicating a gap or
Step 1C Question-raising or
Step 1D Continuing a tradition
Move 3 Occupying the niche
Step 1A Outlining purposes or
Step 1B Announcing present research
Step 2 Announcing principal findings
Step 3 Indicating research article structure
Swales’ Create a Research Space – see also Stede & Schneider,
Argumentation Mining, 2.6, pages 24-25.
58. Rhetorical moves – Teufel’s
Argumentative Zoning
Teufel, Simone. "Scientific Argumentation Detection as Limited-domain Intention
Recognition." ArgNLP. 2014.– see also Stede & Schneider, Argumentation Mining, 2.6,
pages 24-25..
Teufel and Moens [2002] is widely used. A recent hierarchy:
59. Hierarchical Text Structure
A text can be viewed as a hierarchy of text spans
that are recursively connected via coherence
relations:
• Elaboration: [The new Smart Watch was
introduced today.] [It costs $50 more than the old
model.]
• Motivation: [ [The new Smart Watch was
introduced today.] [It costs $50 more than the old
model.] ] [You should really buy it.]
Stede & Schneider, Argumentation Mining, 2.6, pages 25-26
60. Rhetorical Structure Theory [Mann
and Thompson, 1988]
Stede & Schneider, Argumentation Mining, 2.6, pages 25-26. Figure 2.1
from [Stede, 2011, p.115]. See also [Mann and Thompson, 1988]
61. Hierarchical Text Structure resources
• Penn Discourse Tree Bank – relations are
annotated independently of one another. Now
in its third version, see
https://www.seas.upenn.edu/~pdtb/
• RST Discourse Treebank [Carlston et al., 2003]
• Discourse parsers [see Stede Discourse
Parsing, 2011, Section 4.4]
Stede & Schneider, Argumentation Mining, 2.6, pages 25-26.
63. Full-fledged argumentation mining
1. Identify argumentative text
2. Segment the text into argumentative discourse
units (ADUs)
3. Identify the central claim.
4. Identify the role/function of ADUs.
5. Identify relations between ADUs
6. Build the overall structural representation.
7. Identify the type and the quality of the
argumentation.
Stede & Schneider, Argumentation Mining, 1.2, pages 6-7
64. Challenges for argumentation mining
1. Identify argumentative text
– Is there a standpoint on which people disagree?
• Context or domain knowledge may be needed to
determine whether or not a text is argumentative (e.g.
“The rooms turned out to be small.”)
– Prosody can impact the meaning of a text.
• “Really?” vs. ”Really.” vs. “Really!”
65. Challenges for argumentation mining
2. Segment the text into argumentative
discourse units (ADUs)
– Argumentative units occur at multiple levels,
including clauses, sentences, and multi-sentence
groupings.
– Additionally, there are choices: arguments can be
analyzed at different granularities.
66. Challenges for argumentation mining
3. Identify the central claim.
– Sometimes there are multiple plausible
interpretations.
– Argument structure is recursive, so there may be
multiple different arguments.
67. Challenges for argumentation mining
4. Identify the role/function of ADUs
– Support/attack may be implicit.
– Domain knowledge or context may be needed.
5. Identify relations between ADUs.
– Support/attack may be implicit.
– Domain knowledge or context may be needed.
68. Challenges for argumentation mining
7. Build the overall structural representation.
– Possibility of missing information because
arguments are rarely fully specified.
69. Challenges for argumentation mining
8. Identify the type and the quality of the
argumentation.
– Current approaches look at relevance,
acceptability, sufficiency.
See also, Stede & Schneider, Argumentation Mining, 8.3.2,
pages 116-120
72. Kinds of argument
• epistemic: some proposition is true or false;
• ethical or esthetical: something is good or bad
(or: beautiful or ugly);
• deontic: some action should be done or not
done.
Eggs [2000], as quoted in Stede & Schneider,
Argumentation Mining, 1.1, page 3
73. Esthetical argument
• Of all Greenaway’s works (...), this is probably the British
filmmaker’s least effective. As with all of his films, the
choreography of people and objects before the camera (...)
is elaborate and splendid. The film also marks some of
Greenaway’s favorite thematic obsessions, including (in no
particular order) spiritual and corporeal rotting, Sir Isaac
Newton and arcane mullings on things historic, classical
and numerical.
• But Greenaway’s narrative and his direction of actors—two
elements which only recently has he concerned himself
with—are without foundation. After the effects of the
visual presentation have worn off, the film becomes rather
tiresome to follow.
Desson Howe, The Washington Post. Review of ‘The Belly of an Architect’, June 29, 1990.
http://www.washingtonpost.com/wpsrv/style/longterm/movies/videos/thebellyofanarchitectnrhowe_a0b289.htm ,
as cited in Stede & Schneider, Argumentation Mining, 1.3, page 3
75. Argument structures are varied
Rahwan [2008] as cited in Stede & Schneider, Argumentation Mining, Figure 3.1, page 28
76. Argument structures can be analyzed
at different levels
Stede & Schneider, Argumentation Mining, Table 3.1, page 33, based on
Bentahar, Moulin, and Bélanger’s taxonomy of argumentation models, modified from
Bentahar et al. [2010, p. 215]
77. Arguments can be studied at different
granularities.
• Arguments are recursive, zippered structures.
• The conclusion of one argument may be the
premise of the next argument.
Stede & Schneider, Argumentation Mining, 3.1, pages 28-29
79. Argument, Case, Debate
• Argument - “a one-step reason for a claim”
cannot be subdivided into any other parts that
are still arguments
• Case - “a chain of reasoning leading toward a
claim” has only supporting evidence for a
claim
• Debate - “reasons for and against a claim” has
both supporting and conflicting evidence
[Wyner et al., 2015, p. 51], as cited in Stede & Schneider, Argumentation Mining, 3.1, pages 28-29
80. Case example: only supporting
evidence for a claim
Tweet by @robeastaway
https://twitter.com/robeastaway/status/135838892694839296
81. Debate example: has both supporting
and conflicting evidence
We really need to tear down that building.
Granted, it will be expensive, but the degree of
asbestos contamination is not tolerable
anymore. Also, it is one of the most ugly
buildings in town!
Stede & Schneider, Argumentation Mining, 1.1, page 6
83. Argumentation is defeasible.
Argumentation considers tentative conclusions,
which can be revised when new information
comes to light.
This is different from classical inference. In
classical logic, the conclusion is guaranteed to
hold whenever the premises hold.
Stede & Schneider, Argumentation Mining, 3.2 page 29
84. Classical inference rule:
syllogism
Socrates is a man; (Minor premise)
Every man is mortal; (Major premise)
Therefore, Socrates is mortal. (Conclusion)
Stede & Schneider, Argumentation Mining, 3.2 page 29
85. Classical inference rule:
modus ponens
Given if p, then q.
Given p.
Therefore, q.
Stede & Schneider, Argumentation Mining, 3.2 page 29
86. Classical inference rule:
modus tollens
Given if p, then q.
Given not q.
Therefore, not p.
Stede & Schneider, Argumentation Mining, 3.2 page 29
88. Argumentation schemes
An argumentation scheme expresses a defeasible
inference rule for showing the acceptability of a
standpoint.
There are multiple approaches, including:
• Walton/Reed/Macagno: 60 schemes
• Pragma-dialectic school: 3 schemes (sign,
comparison, cause) with variations and subtypes
• Argumentum model of topics (AMT): maxims
(from topoi & loci) activate rules.
Stede & Schneider, Argumentation Mining, 3.3, page 30
89. Argumentation Schemes – Position to
Know
• Major Premise: Source a is in a position to
know about things in a certain subject domain
S containing proposition A.
• Minor Premise: a asserts that A (in Domain S)
is true.
• Conclusion: A is true.
Walton, Reed, Macagno [2008] as cited in Stede & Schneider, Argumentation Mining, 3.3, page 30
90. Argumentation Schemes – Position to
Know
Critical Questions:
1. Is a in a position to know whether A is true?
2. Is a an honest (trustworthy, reliable) source?
3. Did a assert that A is true?
Walton, Reed, Macagno [2008] as cited in Stede & Schneider, Argumentation Mining, 3.3, page 30
91. AMT
Rigotti and Greco Morasso [2010, p. 499], as cited in Stede & Schneider, Argumentation Mining, Fig 3.2, page 32
97. Microtext scheme explained
•Grey boxes: text
•Round: proponent
•Square: opponent
•Arrow: support
•Circle/square: attack
[Peldszus and Stede, 2013] as cited by Stede & Schneider, Argumentation Mining, 4.1, page 45-47
98. Annotation scheme example:
Argumentation schemes on Wikipedia
Schneider, Samp, Passant, Decker, Arguments about Deletion: How Experience Improves the Acceptability of Arguments in Ad-hoc Online Task Groups. CSCW 2013.
99. Annotation scheme example:
Factors on Wikipedia
Schneider, Passant, Decker. Deletion Discussions in Wikipedia: Decision Factors and Outcomes. WikiSym 2012
100. Annotation scheme example:
Kirschner’s science scheme
Kirschner et al. [2015, p. 1]
as cited by Stede & Schneider, Argumentation Mining, 4.1, page 47-48
101. Annotation scheme example: Modified
Toulmin scheme
Habernal and Gurevych [2017, p. 144]
as cited by Stede & Schneider, Argumentation Mining, 4.1, page 48-49
102. Annotation scheme example: Cornell
eRulemaking
Niculae et al. [2017, p. 985]
as cited by Stede & Schneider, Argumentation Mining, 4.1, page 49-50
103. Annotation scheme example:
Inference Anchoring Theory
Budzyńska and Reed [2011]
as cited by Stede & Schneider, Argumentation Mining, 4.1, page 50-51
IF=illocutionary force
RA=inference between
propositions
TA=dialogue rule
relating utterances
106. Example Corpora: Online Interactions
• Internet Argument Corpus
• Agreement by Create Debaters
• Agreement in Wikipedia Talk Pages
• ComArg
• Technorati technical blogs
• Web Discourse
Stede & Schneider, Argumentation Mining, 4.2, pages 51-55
107. Agreement by Create Debaters corpus
example
[Rosenthal and McKeown, 2015, p. 169]
as quoted by Stede & Schneider, Argumentation Mining, Figure 4.6, page 52
108. ComArg comment/argument
[Boltužić and Šnajder, 2014, p. 54]
as quoted by Stede & Schneider, Argumentation Mining, Figure 4.7, page 53
s=implicit support; S=Explicit support
109. Other examples of corpora
• Araucaria
• Argumentative Microtext Corpus
• Cornell eRulemaking corpus
• Webis Editorial corpus
• Persuasive Essay Corpus
Stede & Schneider, Argumentation Mining, 4.2, pages 51-55
114. Quality and consistency assessment
• Assess:
– Is the argument sound?
– Is the argument convincing?
• When information conflicts, what is the
maximally consistent subset?
– Argumentation solvers (e.g. ASPARARTIX,
Carneades, GrappaViz, etc.) are designed to do
this, given a manual analysis or controlled natural
language
Stede & Schneider, Argumentation Mining, 1.3, pages 8-9 &
10.3, pages 140-143
115. Get argument structure from text
(1) Households should pay tax for their
garbage.
115
Arrow: premise
(4) (1)
Paying tax for garbage increases
recycling, so households should pay.
(3) (1)
Recycling more is good, so people
should pay tax for their garbage.
Adam Wyner and Tom van Enger. Towards Web-based Mass Argumentation in Natural
Language. (long version of EKAW 2010 poster).
http://wyner.info/research/Papers/WynerVanEngersEKAW2010.pdf
116. Recycling Debate manually extracted
from BBC “Have your Say”
116
Arrow: premise
Dashed arrow: attacks
Adam Wyner and Tom van Enger. Towards Web-based Mass Argumentation in Natural
Language. (long version of EKAW 2010 poster).
http://wyner.info/research/Papers/WynerVanEngersEKAW2010.pdf
117. Maximal consistent sets
117
Adam Wyner and Tom van Enger. Towards Web-based Mass Argumentation in Natural
Language. (long version of EKAW 2010 poster).
http://wyner.info/research/Papers/WynerVanEngersEKAW2010.pdf
119. Visualization
• Help an individual structure their thoughts
• Show the flow of topics in a debate
• Visualize the points of agreement and
disagreement
127. In Swissbib, search for “Synthesis
Lectures on Human Language
Technologies”:
https://www.swissbib.ch/Search/Res
ults?lng=en&type=ISN&lookfor=194
74040
Main Chapters
1. Introduction
2. Argumentative Language
3. Modeling Arguments
4. Corpus Annotation
5. Finding Claims
6. Finding Supporting and
Objecting Statements
7. Deriving the Structure of
Argumentation
8. Assessing Argumentation
9. Generating Argumentative Text
10. Summary and Perspectives
https://doi.org/10.2200/S00883ED1V01Y201811HLT040
Editor's Notes
Stede & Schneider pages 1-2:
A verbal activity. People can and do gesture and frown at each other, and occasionally this might be a way of resolving some disagreement. In our conception here, however, argumentation is an inherently linguistic activity—either in spoken or written mode. This is to be distinguished from, for instance, a fistfight, which can be a different means of sorting out a conflict.
A social activity. Social emphasizes that argumentation is a matter of interaction among a number of people, with a minimum of two. Granted, many of us munch on a difficult decision for some time by mentally walking through the consequences of the alternatives, but genuine arguing requires a person to argue with.
A rational activity. Obviously, we can perform verbal and social activities in very many ways. Among these, argumentation targets specifically the dimension of reason. When one person reminds the other “Be reasonable!”, the point is to call for a style of dialogue that is not driven by emotional outbreak, power struggle, personal offense, etc., but by the sober exchange of— reasonable arguments.
A standpoint. In argumentation, the heart of the matter is some issue on which people may have divergent views. Thus, the argument does not target an undisputed ‘fact’ but a ‘standpoint’ (or a ‘stance’). Where there is no potential disagreement, there is no need to argue.
Convincing of acceptability. At the end of an argument, the parties involved may not nec- essarily be completely ‘on the same page’ about the standpoint. An argument (usually) does not prove beyond any doubt that one particular viewpoint is the single correct one. Instead, the party who initially was skeptical may now be somewhat more inclined to accept the view of the other party (if the argument was successful).
A constellation of propositions. The shape of a justification might be quite simple, for in- stance a single convincing sentence. But quite often it will be more complex and involve a web of interrelated points that the speaker carefully assembles in a non-arbitrary way; an example of such a constellation will be shown at the end of this section.
Justifying the proposition of the standpoint. If the reader is more inclined to accept the par- ticular standpoint after the argument is finished, that result is not just due to the speaker having pointed out that he is the boss, or some such; it is due to the speaker having successfully justified their view.
A reasonable critic. This phrase in fact combines two distinct aspects. Reasonable reinforces the point made earlier by the ‘rational activity’: the argument is subject to scrutiny on the level of rational thought. But in addition, the presence of the external critic points to a social context in which the argumentation takes place. This can be common political discourse for which most people have the background, or it can be a much more specialized realm such as the scientific, medical, or legal. Whatever the context, it includes particular rules of conduct that have become convention, which the implicit judge of the argument watches over.
Stede & Schneider pages 1-2:
A verbal activity. People can and do gesture and frown at each other, and occasionally this might be a way of resolving some disagreement. In our conception here, however, argumentation is an inherently linguistic activity—either in spoken or written mode. This is to be distinguished from, for instance, a fistfight, which can be a different means of sorting out a conflict.
A social activity. Social emphasizes that argumentation is a matter of interaction among a number of people, with a minimum of two. Granted, many of us munch on a difficult decision for some time by mentally walking through the consequences of the alternatives, but genuine arguing requires a person to argue with.
A rational activity. Obviously, we can perform verbal and social activities in very many ways. Among these, argumentation targets specifically the dimension of reason. When one person reminds the other “Be reasonable!”, the point is to call for a style of dialogue that is not driven by emotional outbreak, power struggle, personal offense, etc., but by the sober exchange of— reasonable arguments.
A standpoint. In argumentation, the heart of the matter is some issue on which people may have divergent views. Thus, the argument does not target an undisputed ‘fact’ but a ‘standpoint’ (or a ‘stance’). Where there is no potential disagreement, there is no need to argue.
Convincing of acceptability. At the end of an argument, the parties involved may not nec- essarily be completely ‘on the same page’ about the standpoint. An argument (usually) does not prove beyond any doubt that one particular viewpoint is the single correct one. Instead, the party who initially was skeptical may now be somewhat more inclined to accept the view of the other party (if the argument was successful).
A constellation of propositions. The shape of a justification might be quite simple, for in- stance a single convincing sentence. But quite often it will be more complex and involve a web of interrelated points that the speaker carefully assembles in a non-arbitrary way; an example of such a constellation will be shown at the end of this section.
Justifying the proposition of the standpoint. If the reader is more inclined to accept the par- ticular standpoint after the argument is finished, that result is not just due to the speaker having pointed out that he is the boss, or some such; it is due to the speaker having successfully justified their view.
A reasonable critic. This phrase in fact combines two distinct aspects. Reasonable reinforces the point made earlier by the ‘rational activity’: the argument is subject to scrutiny on the level of rational thought. But in addition, the presence of the external critic points to a social context in which the argumentation takes place. This can be common political discourse for which most people have the background, or it can be a much more specialized realm such as the scientific, medical, or legal. Whatever the context, it includes particular rules of conduct that have become convention, which the implicit judge of the argument watches over.
From Stede & Schneider page 11:
when the speaker decides to reveal a private state, there is no point in replying Not true! We may very well like or dislike that particular revelation, but objecting to its truth is usually not an option.
From Stede & Schneider page 12:
a listener may legitimately react to these statements with some objection, but then the objection applies to the expressed content of the private state, not to its revelation. For example, a listener responding with Wrong! to 2.7 objects to the notion of the creature over there being a llama—and not to the fact that the speaker thinks otherwise (more technically: wrong does not scope over I guess). Very often, these cases are ambiguous (and can sometimes produce mis- understanding) when those verbs of cognition are not explicitly stated but left implicit by the speaker, as is the case in Examples 2.4–2.6.
Stede & Schneider page 17:
The fact that people in daily life typically do not constantly use those extended versions con- tributes to ambiguity and sometimes to misunderstanding (but, arguably, makes communica- tion much more interesting); recall our discussion of Examples 2.3–2.7 at the beginning of the chapter. This ambiguity in the roles that utterances can play also makes the task of argumenta- tion mining difficult, because in some way or another, the status of individual statements that take part in argumentation needs to be determined (and that would be much easier if those performative verbs were present).
Stede & Schneider p 18
One reason to distinguish illocution from perlocution is that an utterance may well have—from the speaker’s perspective—unintended or even undesired effects. A speaker could comment on the addressee’s new hairstyle in a neutral way, and the addressee might derive the literal meaning (as intended by the speaker) but in addition surmise a secondary meaning and feel offended (as not intended by the speaker).
For argumentation, the perlocution is of obvious relevance, as it characterizes precisely that ‘change of mind’ on the addressee’s side, which is the purpose of the speaker devising their argument. Research on persuasion seeks to detect or even measure the strength of such effects. For speech act theory, however, the perlocution is a private state that is of limited interest; the heart of the matter is the perspective of the speaker and thus the illocutionary force of her utterance.
Stede and Schneider p23
In both cases, we have a directive speech act and a justification, but stated very differently. Which one is more appropriate, and correspondingly, more effective, of course depends very much on context: Who is talking to whom under what circumstances. This may seem trivial, but the point is that a similar array of linguistic options for articulating a sentence is essentially always at the writer’s disposal.
Stede and Schneider p23
In both cases, we have a directive speech act and a justification, but stated very differently. Which one is more appropriate, and correspondingly, more effective, of course depends very much on context: Who is talking to whom under what circumstances. This may seem trivial, but the point is that a similar array of linguistic options for articulating a sentence is essentially always at the writer’s disposal.
From Stede and Schneider p25:
Rhetorical Structure Theory or RST [Mann and Thompson, 1988] postulates that a well-formed tree structure can be built, which spans the text completely.7 The authors postulate that some 25 different coherence relations are sufficient for analyzing most texts. A sample RST tree (for an excerpt of a text) is shown in Figure 2.1. The best-known English RST corpus is the RST Discourse Treebank or RST-DT [Carlson et al., 2003], and it has led to numerous proposals of auto- matic parsers that try to build the tree structures automatically. Stede [2011, Sct. 4.4] gives an overview.
From Stede and Schneider p25:
Evidently, the analysis of argumentation is not entirely different from the analyses in the PDTB and in the RST-DT. For RST, in particular, parallels between the tree structures and similar analyses of argumentation structure (recall the example shown in Figure 1.1) are easy to notice. However, there is no consensus yet as to how the relationship is to be construed ex- actly. A summary of the discussion was provided by Peldszus and Stede [2013], and later on,
Stede & Schneider page 33
One of the most influential models of argumentation, the Toulmin model, was first proposed by British philosopher Stephen Toulmin in 1958. This model takes a mono- logical view of argumentation, focusing on the internal structure of an argument, which is viewed as a tentative proof. The Toulmin model has a detailed internal structure, as shown in Figure 3.3. In the model, a claim is defeasibly supported by its grounds (also called the data), according to some warrant. If necessary, the warrant may be further supported by a backing. Qualifiers or rebuttals can also be diagrammed, but are not required.
In Figure 3.3, the claim is “Anne now has red hair”. Its grounds is “Anne is one of Jack’s sisters”. Its warrant is “any sister of Jack’s may be taken to have red hair”, which has the backing, “All his sisters have previously been observed to have red hair.” Its qualifier is “presumably”. Its rebuttal is “unless Anne has dyed/ gone white/ lost her hair...”.
Argument mapping and education have been among the areas influenced by the Toul- min model. Furthermore, it has been taken up in Computational Linguistics and—with modifications—used for annotating texts by Habernal and Gurevych [2017]; we will introduce that work in the next chapters.
They don’t say how they extracted these – but they say
Someone makes statement (1)
Someone else gives (4) as a reason/premise for (1)
Someone else gives (3) as an additional reason for (1)
(2) Is a counterproposal with a range of supporting reasons
===
Icons:
http://findicons.com/icon/27954/girl_5?id=27964#
http://findicons.com/icon/27930/boy_8?id=27939#
http://findicons.com/icon/27955/girl_4?id=27965#