ChangeMyView Through Concessions: Do Concessions Increase Persuasion?

In discourse studies concessions are considered among those argumentative strategies that increase persuasion. We aim to empirically test this hypothesis by calculating the distribution of argumentative concessions in persuasive vs. non-persuasive comments from the ChangeMyView subreddit. This constitutes a challenging task since concessions are not always part of an argument. Drawing from a theoretically-informed typology of concessions, we conduct an annotation task to label a set of polysemous lexical markers as introducing an argumentative concession or not and we observe their distribution in threads that achieved and did not achieve persuasion. For the annotation, we used both expert and novice annotators. With the ultimate goal of conducting the study on large datasets, we present a self-training method to automatically identify argumentative concessions using linguistically motivated features. We achieve a moderate F1 of 57.4% on the development set and 46.0% on the test set via the self-training method. These results are comparable to state of the art results on similar tasks of identifying explicit discourse connective types from the Penn Discourse Treebank. Our findings from the manual labeling and the classification experiments indicate that the type of argumentative concessions we investigated is almost equally likely to be used in winning and losing arguments from the ChangeMyView dataset. While this result seems to contradict theoretical assumptions, we provide some reasons for this discrepancy related to the ChangeMyView subreddit.

[1]  Mitsuko Narita Izutsu Contrast, concessive, and corrective: Toward a comprehensive study of opposition relations , 2008 .

[2]  Rashmi Prasad,et al.  Reflections on the Penn Discourse TreeBank, Comparable Corpora, and Complementary Annotation , 2014, CL.

[3]  Moshe Azar,et al.  Concession relations as argumentation , 1997 .

[4]  Kathleen McKeown,et al.  PDTB Discourse Parsing as a Tagging Task: The Two Taggers Approach , 2015, SIGDIAL Conference.

[5]  Hwee Tou Ng,et al.  A PDTB-styled end-to-end discourse parser , 2012, Natural Language Engineering.

[6]  D. Davidson Inquiries Into Truth and Interpretation , 1984 .

[7]  S. Thompson,et al.  A linguistic practice for retracting overstatements: ‘Concessive repair’ , 2005 .

[8]  Janyce Wiebe,et al.  Recognizing Contextual Polarity in Phrase-Level Sentiment Analysis , 2005, HLT.

[9]  Swapna Somasundaran,et al.  Recognizing Stances in Online Debates , 2009, ACL.

[10]  Joel R. Tetreault,et al.  Automatically Identifying Good Conversations Online (Yes, They Do Exist!) , 2017, ICWSM.

[11]  Rashmi Prasad,et al.  The Penn Discourse Treebank , 2004, LREC.

[12]  Marie-Francine Moens,et al.  Argumentation mining: the detection, classification and structure of arguments in text , 2009, ICAIL.

[13]  Joel R. Tetreault,et al.  Finding Good Conversations Online: The Yahoo News Annotated Comments Corpus , 2017, LAW@ACL.

[14]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[15]  S. Thompson,et al.  Concessive patterns in conversation , 2000 .

[16]  Iryna Gurevych,et al.  Annotating Argument Components and Relations in Persuasive Essays , 2014, COLING.

[17]  Elena Musi How did you change my view? A corpus-based study of concessions’ argumentative role , 2018 .

[18]  Nancy Green,et al.  Representation of Argumentation in Text with Rhetorical Structure Theory , 2010 .

[19]  Cristian Danescu-Niculescu-Mizil,et al.  Winning Arguments: Interaction Dynamics and Persuasion Strategies in Good-faith Online Discussions , 2016, WWW.

[20]  James R. Curran,et al.  Bootstrapping POS-taggers using unlabelled data , 2003, CoNLL.

[21]  Kai Zheng,et al.  Hedging their Mets: The Use of Uncertainty Terms in Clinical Documents and its Potential Implications when Sharing the Documents with Patients , 2012, AMIA.

[22]  William C. Mann,et al.  Rhetorical Structure Theory: Toward a functional theory of text organization , 1988 .

[23]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[24]  Paula Castro,et al.  Managing disagreement through yes, but… constructions: An argumentative analysis , 2015 .

[25]  Bing Liu,et al.  Mining and summarizing customer reviews , 2004, KDD.

[26]  Marilyn A. Walker,et al.  A Corpus for Research on Deliberation and Debate , 2012, LREC.

[27]  Bob Carpenter,et al.  The Benefits of a Model of Annotation , 2013, Transactions of the Association for Computational Linguistics.

[28]  Brian Ecker,et al.  Argument Mining: Extracting Arguments from Online Dialogue , 2015, SIGDIAL Conference.

[29]  Iryna Gurevych,et al.  Which argument is more convincing? Analyzing and predicting convincingness of Web arguments using bidirectional LSTM , 2016, ACL.

[30]  Debanjan Ghosh,et al.  Coarse-grained Argumentation Features for Scoring Persuasive Essays , 2016, ACL.

[31]  Ani Nenkova,et al.  Automatic sense prediction for implicit discourse relations in text , 2009, ACL.

[32]  Yi Li,et al.  Is This Post Persuasive? Ranking Argumentative Comments in Online Forum , 2016, ACL.

[33]  Livio Robaldo,et al.  The Penn Discourse TreeBank 2.0. , 2008, LREC.

[34]  K. Hyland,et al.  Writing Without Conviction? Hedging in Science Research Articles , 1996 .

[35]  K. Hyland,et al.  Talking to the Academy , 1996 .

[36]  Greg McVerry,et al.  Argumentation Schema and the Myside Bias in Written Argumentation , 2019 .

[37]  Frans H. van Eemeren,et al.  Argumentative Indicators in Discourse, A Pragma-Dialectical Study , 2007, Argumentation Library.

[38]  J. Fleiss Measuring nominal scale agreement among many raters. , 1971 .

[39]  Kevin L. Blankenship,et al.  The Role of Different Markers of Linguistic Powerlessness in Persuasion , 2005 .

[40]  Manfred Stede,et al.  Ma(r)king concessions in English and German , 1995, ArXiv.

[41]  Ch. Perelman,et al.  The new rhetoric , 1957 .

[42]  Rada Mihalcea,et al.  Co-training and Self-training for Word Sense Disambiguation , 2004, CoNLL.