SciELO - Scientific Electronic Library Online

vol.11 número1Intergenerational and collaborative use of tablets: «in-medium» and «in-room» communication, learning and interactionLimitaciones en el uso y apropiación de tecnologías de información y comunicación para la negociación internacional en empresas colombianas índice de autoresíndice de assuntospesquisa de artigos
Home Pagelista alfabética de periódicos  

Serviços Personalizados




Links relacionados

  • Não possue artigos similaresSimilares em SciELO


Observatorio (OBS*)

versão On-line ISSN 1646-5954

OBS* vol.11 no.1 Lisboa mar. 2017


A Rhetorical Analysis of English and Persian Online Comments on the News Articles Related to Iran’s Nuclear Issue


Fatemeh Irajzad*, Mohsen Kafi**, Hesamoddin Shahriari***

* MA in TEFL, Department of English Language and Literature, Ferdowsi University of Mashhad, Iran Researcher at Ferdowsi University of Mashhad, Iran (

** MA in Translation Studies, Department of English Language and Literature, Ferdowsi University of Mashhad, Iran (

*** Assistant Professor of Applied Linguistics, Department of English Language and Literature, Ferdowsi University of Mashhad, Iran (



The comments section of news articles is intended for the public to share their views on posted news stories. Accordingly, this section has been subject to numerous analyses from a wide range of perspectives. The current study analyzes the tone and purpose of comments posted on news articles dealing with Iran’s nuclear issue. The data consisted of 150 online comments in English and 150 in Persian. The comments were taken from three news websites in English and three in Persian. These websites were chosen based on their different political leanings in order to take account of the various approaches to Iran’s disputed nuclear talks. The results did not indicate a significant difference between the tone of English and Persian comments, but the purposes intended by English and Persian comments varied extensively. It was also found that the political leaning of news agencies is a major cause of the observed differences.

Keywords: Comment analysis, Iran’s nuclear dispute, news articles, purpose, tone.



The rapid growth of online technologies has provided online users with the chance to be involved in a two-way interaction as both receivers and producers of information. One of the ways through which readers can express their ideas about news stories is by publishing their comments online. A comment is defined as “a short value judgment about a specific topic which can be a piece of news, image or any common document” (Granados, Garzón, Urueña, & Guarín, 2012, p. 113). In 1998, the Rocky Mountain news website was a pioneer in providing its readers with same-page comment section on movie and restaurant reviews (Santana, 2011). The idea of publishing comments does not merely serve commercial purposes, as online newspapers and political news outlets also publish their readers’ comments in order to help demonstrate public opinion. The users of online websites have different motivations for writing and reading comments. According to McQuail's classification (1994), providing or gaining information, expressing personal identity, establishing social interactions and being entertained are the most common reasons for media use. In fact, the comments may sometimes prove to be more appealing than the news article itself. Hence, a wide array of scholars have shown an increased interest in different aspects of online comments, such as the properties of social interaction in online comments (e.g., Mcmillen, 2013), user agency of readers' comments in online news media (e.g., Milioni, Vadratsikas, & Papa, 2012), and racism in online comments (e.g., Hughey & Daniels, 2013).

Another potential approach to studying online comments is rhetorical analysis. According to Selzer (2004), rhetorical analysis involves understanding how certain ideas and attitudes are conveyed through a spoken or written message, with a special focus on the methods of persuasion. More specifically, online comments may be investigated through the prism of comparative rhetoric, the main objective of which is to discover universal and distinctive features of two or more cultures. In fact, news articles can be a fruitful genre for comparative rhetorical studies. For instance, a study conducted by Chana and Tangkiengsirisin (2012) looked into the discursive differences between news articles appearing on English-language newspapers published in Thailand and those published in the United States. Another study conducted by Domingo et al. (2008) sought to outline the structural properties of audience participation in 16 online newspapers from eight European countries as well as the United States. Trakulkasemsuk and Pingkarawat (2010) also worked on magazine articles to find out the particular features of Thai English compared to British English concerning the use of noun modifiers. In the same vein, the current study seeks to investigate how the tone and purpose of Persian and English comments differ from one another.

 The Iran nuclear dispute and the negotiations taking place between the country and world powers have received wide coverage by online news media. The high stakes involved in this issue have encouraged users to share their views on the topic, which is why the present study focuses on news stories on this particular topic. Our aim is to apply a content analysis on users’ comments posted on three Persian news websites (PNWs) and three English news websites (ENWs) with different political leanings, in order to find out how the tone and purpose(s) of user comments vary across each of the mentioned groups. To the authors’ best of knowledge, no previous study has analyzed the comments produced by two groups of commenters through focusing on the multiple functions a comment may potentially serve. Therefore, this study was driven by three basic questions:

  1. What are the possible differences between English and Persian comments with regards to their tone and purpose(s)?
  2. What purpose(s) are mostly intended by emotional/rational comments?
  3. Is there a relationship between the political leanings of news websites and the tone/purpose(s) of the posted comments?


Review of the Related Literature

Political rhetoric has long been a topic of interest among researchers. “The topic of political rhetoric concerns the strategies used to construct persuasive arguments in formal public debates and in everyday political disputes” (Condor, Tileaga, & Billig, 2013, p. 262). Contemporary authors have extended the scope of rhetorical scholarship to include texts (e.g., Spurr, 1993) and informal talk (e.g., Billig, 1991). The rapid growth of the World Wide Web has greatly affected the means through which political rhetoric is produced. Most online news outlets are now allowing readers to express their ideas on different topics and news stories. Bruns (2007) describes "a new hybrid form of simultaneous production and usage, or produsage …" (p. 100), in which readers gradually move away from being mere recipients and take on the role of authors. Therefore, the previous monologue which dominated most news outlets has slowly given way to a dialogue that foregrounds the reader as a ‘produser’ of information.

The new status quo which has resulted in a large database of online comments on many websites has fueled a substantial body of research by Media and Mass Communication researchers. Also, scholars from a variety of disciplines have looked into this new situation from different perspectives. Several scholars have explored the potential of the internet to form public spheres (Al-Saggaf, 2006; Poor, 2005) and the identities of internet users (Leung, 2009; Nagar, 2011), while some have analyzed the motivations behind user participation (LaRose & Eastin, 2004; Milioni, Vadratsikas, & Papa, 2011). Promoting the quality of comments has been another popular line of research. In this context, Quality refers to “a degree of excellence in communicating knowledge or intelligence and normatively includes notions of accuracy, reliability, validity, currency, relevancy, comprehensiveness, and clarity” (Diakopoulos & Naaman, 2011b, p. 133).

Opinion and sentiment analysis is another domain which has received considerable attention. According to Sun, Kong, and Zhao (2011), the main aims of opinion analysis of online comments are to identify the purposes intended by commenters and determine the polarities of comments. They proposed a new method—Distance-Weighted Count (DWC) of polarized words— to classify the polarity of comments. Regarding the positivity and negativity of comments, Diakopoulous and Naaman (2011a) found that news stories with a negative topic attract more commenters, and users who comment more regularly are still more negative in their comments. Sentiment analysis “encompasses a broad area of information mining and natural language processing to appraise a judgment as positive or negative, and determine the emotional state of the writer and subjectivity of the opinion” (Granados et al., 2012, p. 113).

In another study, Glenn, Champion, and Spence (2012) applied a qualitative content analysis on online news media coverage of weight loss surgery and the related reader comments. They found that when considering the tone, articles were divided into supportive/unsupportive, while comments were categorized as supportive/unsupportive, aggressive/passive, and negative/positive/neutral. Also, McCluskey and Hmielowski (2012) compared online reader comments with letters to the editor. They argue that due to issues such as anonymity, absence of moderation systems, and participation of younger users, online reader comments are different from letters to the editor. Despite some criticisms of anonymous comments, many netizens consider comment anonymity as an important factor in the free expression of opinions (McMillen, 2013).

As mentioned above, previous studies have analyzed online comments from different perspectives and with diverse scopes and objectives. In the present study, a detailed analysis of English and Persian online comments regarding Iran’s nuclear talks will be provided. It has to be mentioned that in this study, the terms English and Persian merely refer to the language of the comment and do not have any reference to the nationality of the commenter. However, based on the information found for each commenter, nearly all the Persian comments were written by Iranian users and a very high percentage of English comments were posted by American and European users.



Selecting the News Websites

It is commonly known that the political leaning of a news outlet influences the opinion of its users (e.g., Kuypers, 2002; Petrova, 2008). Accordingly, many researchers have focused on identifying the political orientation of different forms of media (e.g., Schneider & Lewis, 1985; D’Allesio & Allen, 2000). More specifically, Groseclose and Milyo (2005) attempted to identify the number of citations to the members of a particular political group in order to determine the political orientation of some newspapers. Also, Weatherly, Petros, Christopherson, and Haugen (2007) analyzed the headlines of CNN and Fox News to determine their possible political bias. Similarly, in this study, the websites were chosen based on their different political leanings, so as to capture a general picture of the various standpoints to Iran’s disputed nuclear talks and investigate the impact of the news websites’ political leanings on the political bias of the users.  In fact, this line of research has previously been recommended by some researchers (e.g., Zhitomirsky-Geffet, David, Koppel, & Uzan, 2016). The Left-Right dichotomy (also known as Liberal-Conservative or Democrat-Republican) was used to categorize the news websites.

  • Left orientation: Optimistic towards foreign relations and any international agreement which would prevent conflicts/war
  • Right orientation: Pessimistic view of foreign relations and often talking about the possibility of war

The political leanings of the selected news outlets were determined based on the findings of a number of previous studies (e.g., Weatherly et al. 2007), the information found on popular online forums (e.g., Yahoo Answers, Quora), and informal talks with a number of Iranian/American political activists. In fact, there was a high consensus among the consulted sources regarding the political leaning of the left-oriented and right-oriented news websites. Table 1 presents the list of the selected English and Persian news websites and their political leaning.



Data Collection

An initial step for gathering the data was to search the phrase ‘Iran’s nuclear talks’ in the selected ENWs and ‘مذاکرات هسته ای ایران’ in the selected PNWs. The resulting news stories with more than five comments were saved in separate folders. The top five qualifying comments on each news article were documented for later analysis. Duplicate comments, material directly-quoted from other sources, and comments irrelevant to the content of the news story did not qualify for inclusion in the sample. Overall, 300 comments were collected over a six-week time period. More specifically, 150 comments were taken from English news websites (50 comments from each ENW) and 150 were extracted from Persian news websites (50 comments from each PNW).


Analysis of the Comments

Once the data were gathered, two independent raters who had previously agreed to analyze the comments were called upon. The raters were trained linguists experienced in rhetorical analysis. After a calibration session in which the aims of the analysis were explained in detail by the lead author and potential ambiguities were clarified, a piloting phase was conducted in which a number of selected comments were rated and discussed.

As for their tone, the comments were judged based on the Aristotelian trichotomy of Pathos, Logos, and Ethos. In this trichotomy, pathos accounts for the persuasive arguments which are mainly derived from the values held by a community, Logos refers to solid reasons emerging from intellectual thinking, and Ethos accounts for the features of a text which are related to the credibility of the speaker. Since many online comments are written by anonymous users and the personality type of the commenters cannot be determined based on a few lines of writing, Ethos was excluded from our categorization.

For the purpose of this study, comments which had an appeal to pathos were considered as emotional and those which had an appeal to logos were rated as rational. More specifically, the comments which contained linguistic devices describing emotional states, linguistic strategies (e.g., interjections) which directly portrayed emotional attitudes and grammatical/rhetorical means (e.g., an exclamative sentence or a marked structure) which highlighted the emotive meaning (Maynard, 2002), were considered to be emotional. On the other hand, the comments which included factual information and step-by-step explanation were rated as rational. In fact, tone is an elusive aspect of political discourse, especially since rational and emotional tones are deeply integrated. Thus, the authors attempted to overcome this limitation by asking the raters to take into account the dominant tone of each comment. The raters also considered the alteration/consistency of the tone throughout the comment. If the tone shifted in a comment, that comment was categorized as having an altering tone.

In the next step, a rubric was designed in order to provide a framework for the raters to identify the purpose(s) intended by each comment. In fact, the rubric drew upon a taxonomy mentioned by S-TEAM (Science-Teacher Education Advanced Methods) (2010) with slight adaptations for this study. The taxonomy covers a range of purposes which could potentially differentiate among the target comments. More specifically, it includes the following purposes: explain, inform, anger, persuade, amuse, motivate, sadden, ridicule (replaced by offend), defend, and attack (warn and threaten were later added). Some categories of this taxonomy (e.g., attack, amuse, defend, and persuade) have also been mentioned by Aristotle in his well-known work Rhetoric. An early draft of the rubric was piloted with 50 comments given to the two raters. The raters were told to identify the comments which did not fall into any of the above-mentioned categories. They were later asked to add any other categories that would allow for the inclusion of all the comments.

Through the process of purpose analysis, the raters realized that aggressive criticism, threat, and warning are also commonly-observed purposes intended by the comments. Thus, warn and threaten were added to the former list. Also, the category ‘ridicule’ was replaced by ‘offend’ which led to the inclusion of more comments. Raters were not limited in the number of purposes that they identified within a single comment. They judged each comment independently and neither of them were aware of the source of the comment or the other rater’s decision. In cases where the two raters agreed on the tone and purpose(s) of a comment, the decision was finalized. However, when the raters did not agree on the tone and/or purpose(s) of a comment, a third rater was called upon to reach a final decision.



The results of this study are presented in three parts. In the first section, the comments from Persian and English news websites are compared in terms of their tone and purpose(s) in order to address the first question of the research. In the second part, comments from various PNWs will be compared with each other and the third section will draw a similar comparison among the comments posted on ENWs. The findings presented in these two sections would be used to address the second and third research questions.


Comparison of English and Persian Comments

Tone analysis

In order to address the first research question, the English and Persian comments were compared in terms of their tone and intended purpose(s). The results indicate a considerable degree of similarity between PNWs (Figure 1) and ENWs (Figure 2) regarding the tone of the comments that appear on each.




In the case of PNWs (Figure 1), 58% of the comments (87/150) were judged to be emotional, 26% (39/150) were rational, while only 16% (24/150) were seen to have an altering tone. As for ENWs (Figure 2), the majority of comments (62%; 93/150) were rated as emotional, 24% (36/150) were rational, while only 14% (21/150) were seen to have an altering tone. This indicates that although the difference is not significant, Iranian commenters are still more rational in their tone.

Most statisticians agree that Cohen’s Kappa is a suitable measure for interrater reliability (Cohen, 1960; Spitzer, Cohen, & Fleiss, 1968). Therefore, in order to determine if there was an acceptable agreement between the two raters’ judgements regarding the tone of the comments, Cohen’s Kappa was run on SPSS version 23. The results of this test are presented in Table 2 and Table 3.




As shown in Table 3, there was an acceptable agreement between the two raters’ judgements (κ = .709, p < .0005). According to Streiner, Norman, and Cairney (2014), kappa values between 0.6-0.74 are considered to be ‘good’.

For illustrative purposes, three original comments are reproduced and analyzed in this section to familiarize the readers with the different types of tone (i.e., emotional, rational, and altering). The researchers have provided the readers only with English comments, since translating the Persian comments into English may have resulted in an unintended change in tone and intended purpose(s).

The following comment has been taken from Fox News. The use of offensive words such as fool, criminal, and incompetent (a technique known as name calling) as well as the frequent use of metaphorical language makes this comment exemplar of the emotional tone. A large number of comments appearing on Fox News strongly opposed the nuclear talks between Iran and 5+1 countries1, mainly on the basis that Iran’s nuclear program is a threat to all countries, especially Israel. Having mentioned the stance of Fox News towards Iran’s nuclear talks, it is not surprising that the users posting on this website also oppose these negotiations.

When the mushroom cloud expands over Israel and we are heading for shelters, Lurch will be on his yacht waiting for Barry to send him to swift boat some other poor nation. I have never seen an administration with 100% incompetents, fools and criminals. Strong memo to follow.

Below is a comment posted on Reuters. This comment was published under a news which discussed the idea of Hillary Clinton to keep all options on the table in case the nuclear negotiations failed. The commenter partly disagrees with the stance of Ms. Clinton, but expresses his idea through a dominant rational tone:

That may be true, and we should keep all options open in any case, but I don’t think it’s helpful for Clinton to be saying this. Iran should be given the chance to abandon its nuclear ambitions and rejoin the international community as a normal country. Even if this is a long shot, it must be given the chance anyway. If worse comes to worst, then yes we do have “all options on the table”, but part of being the world policeman means offering other people a fair deal. That is why Barack Obama’s approach is the correct one, and why the Republicans’ approach is the wrong one.

Following is an altering comment posted on Reuters. It is interesting that the commenter starts and ends his comment with an emotional tone, while using a rational tone in between to solidify the argument. In fact, this is a widely-used rhetorical technique in political discourse. Also, in the concluding sentences, the commenter associates the West with Israel, a strategy which is known as guilt by association.

Israel is the only illegal nuclear weapons state in the Middle East, the sanctions should be on Israel. Iran has, to date, done absolutely nothing illegal [emotional]. Their rights to the entire fuel cycle, including enrichment, are guaranteed under the NPT, the burden of proof that there is a military aspect to Iran’s program lies with the West, not Iran, as there simply is no way to prove the non-existence of a thing. It’s like saying “give me proof that there’s no such thing as the tooth fairy.” The proof should be that nobody’s ever seen the tooth fairy, but if someone asks for concrete evidence she doesn’t exist, how is that possible? It isn’t [rational]. The West are just stepping to the rhythm of Israel’s fiddle. As Israel is the primary antagonist in the middle east with its violent settlers committing daily war crimes of land theft in the occupied territories, with its illegal nuclear weapons program and threats to preemptively destroy Iran on the basis of little more than emotion, with its modern apartheid state, the West should be ashamed to be associated with Israel at all. Sanction and boycott Israel [emotional].


Purpose analysis

Despite the high similarity between the tone of English and Persian comments, variation was commonly observed among Persian and English news outlets with regard to the purpose(s) intended by the commenters. The comparative analysis of the intended purpose(s) of English and Persian comments is seen in Figure 3.



As can be seen in Figure 3, most categories of purpose were intended more frequently by English commenters. More specifically, English comments intended to attack, offend and anger far more often than Persian comments. Paradoxically, English comments also intended to explain, inform, persuade and amuse more than Persian comments. On the other hand, the categories of motivate, sadden, defend, and warn were more commonly seen among the comments posted by Iranian users.

In order to better demonstrate the different purposes intended by the users, three comments are provided below. The purposes corresponding to each section of the comment have been indicated in square brackets.

The following is a comment taken from Reuters which intends to warn, attack, explain, and persuade. The first two purposes are associated with an emotional tone, while the next two are fulfilled through a rational tone.

A successful result is achievable ONLY IF the P5+1 can keep Netanyahu out of the room. Netanyahu will pull out all the stops to torpedo an agreement. He will marshall his army of lobbyists putting pressure on the US and EU members of the P5+1 [warn], [attack].

If an agreement is near, he will try to have the P5+1 move the goal posts (eg ridiculously low enrichment, destruction of Arak, destruction of conventional weapons/missiles, etc) …. anything known to be totally unacceptable by the Iranians. He will also try to ensure that Israel’s secret nuclear weapons program never gets close to the table. In the interests of peace in the Middle East, for God’s sake keep Netanyahu out of the room. We must avoid a (P5+1)xN [explain], [persuade].

The following comment was posted on Aljazeera. The commenter intended to inform, anger, sadden, offend, attack, defend, and warn, most of which are accompanied by an emotional tone.

Who threatened Israel 50 years ago with nuclear weapons?

One would assume that you meet threat with like threat. That a country has these weapons of 'mass destruction' and is outside the NTP is unnerving [inform] and that it is an aggressive and paranoid state that ignores international law in it's illegal occupations of Palestinian lands is even more unnerving! [attack], [warn]

If Iran gets attacked...they would not have the nerve, bombing the Gaza ghetto defenseless population with chemical weapons is more in their line. Cowardly acts. [sadden], [offend], [defend] As for 'full international support' I assume you mean USA and a few Pacific atolls as usual.

Don't have the source but remember a survey in Europe a few years ago that rated Israel the greater threat to world peace. It was leaped on by the Israel lobby as showing anti-Semitism! [inform] Hypocrisy on stilts!

Below is a comment posted on Fox News which intends to anger, sadden, and offend.

Kerry is rates up there with Jane Fonda.  I can’t believe he is broking a deal for us. But I can’t believe the List of Actors in this Democratic Presidency. Starting with Obama [anger, offend]. It’s worse than any nightmare that anyone can imagine. I just wish it was just a bad dream. God Bless & Help us survive the next three years [sadden].

As for the second question of the study, the authors found a meaningful relationship between the tone of the comments and the purposes intended by the commenters, each influencing the other. In fact, in most cases, it was possible to make decisions about the tone of the comment based on the intended purposes. In other words, certain purposes were fulfilled through an emotional/rational tone. For instance, the categories of explain, inform, persuade, and motivate are linked to the rational tone, whereas anger, amuse, offend, attack, sadden, defend, warn, and threaten are associated with an emotional tone. This finding harkens to mind Aristotle’s Rhetorical Triangle in which there is a relationship between different rhetorical elements, two of which are tone and purpose.


Analysis of comments posted on Persian news websites

In this section, the differences among Persian news agencies with respect to the tone and purpose of their user comments are highlighted. The results of our data analysis regarding the tone and purposes of the comments posted on Iranian news agencies are plotted in Figure 4 and Figure 5.




The observed differences could be explained with reference to the political orientation and content features of the selected media outlets. For instance, by reviewing the news articles on Iranian diplomacy as a specialized foreign policy website with moderate political orientations, readers will soon realize that the agency strives for more objectivity compared to the other websites which lean towards a certain political group/ideology. Accordingly, the degree of rationality found in the comments posted on Iranian Diplomacy (40%, 20/50) was higher than the other two news agencies. Moreover, compared to Tabnak and FarsNews, explanation (26%, 13/50), informing (48%, 24/50), and persuasion (22%, 11/50) were more commonly found in the comments posted on this website. The comments posted on Fars News and Tabnak demonstrated almost the same degree of emotionality (66%, 33/50; 70%, 35/50, respectively). However, whereas commenters posting on Fars News had a strong tendency to attack the opposing side (48%, 24/50), Tabnak users had an inclination to defend their stance or favored group (30%, 15/50).


Analysis of Comments Posted on English News Websites

In this section, the tone and purpose of comments posted on each of the selected ENWs are compared in order to investigate the impact of the political leanings of ENWs on the tone and purpose of user-produced content (i.e., comments). The results are presented in Figures 6 and 7.




A large number of comments posted on Fox News and Rueters had an emotional tone (74%, 37/50; 64%, 32/50, respectively), but the comments posted on Aljazeera were less emotional (44%, 22/50). Although some of the comments posted on Fox News and Rueters had a rational tone (14%, 7/50; 26%, 13/50, respectively), the level of rationaliy was higher among the comments appearing on Aljazeera (30%, 15/50).

Regarding the purpose of comments, it was found that the users of Aljazeera had a greater tendency to explain (18%, 9/50), persuade (24%, 13/50), motivate (22%, 11/50), and defend (36%, 18/50). As for Reuters, the categories of inform (46%, 23/50), amuse (18%, 9/50), warn (30%, 15/50), and threaten (6%, 3/50) were more frequntly discovered in user comments. As shown above, commenters posting on Fox News were seen to offend (58%, 29/50) and attack (54%, 27/50) more frequently compared to the other functions considered in this study. Also, the comments posted on Aljazeera and Fox News intended to anger (54%, 27/50) and sadden (28%, 14/50) equally.


Discussion and Conclusions

This study has been the first attempt to draw a comparison between and among Iranian and English news outlets regarding the tone and purpose of user comments. For this purpose, a comparitive rhetorical analysis of user comments was conducted. The findings of this study regarding the first research question indicated that while there is not a meaningful difference in the tone of comments posted on ENWs and PNWs, the purposes intended by Persian and English comments are significantly different. Also, with regard to the second question of the study, the researchers found a meaningful relationship between the tone of the comments and the purposes intended by the commenters, each influencing the other. Finally, the findings with regard to the third research question indicated that the political leaning of news websites has a meaningful influence on user comments. Regardless of language background, Persian and English news websites form a continuum.

On the one hand, there are news websites with a right (conservative) political leaning (i.e., Fox News and Farsnews). Fox News is considered to have strong right-wing political leanings with support for the political right and the republican party, mostly attracting readers who do not support the policies of the Democratic party in the United States. On the other hand, Farsnews is known to have an anti-Western stance and biased viewpoint against the US in particular. Despite the obvious differences between the political stance of Fox News and Farsnews, their radical views towards political issues result in a relative similarity between the tone and purpose of comments posted on each. As seen in the results section, comments posted on both of these news agencies show a high level of emotionality. Moreover, a high percentage of comments posted on Farsnews and Fox News intended to attack the opposing sides. These findings are in accordance with Kuypers (2002) and Petrova (2008) who argue that news outlets have the potential to influence public opinion by the way they frame political issues.

 On the other side of the continuum, there are news websites with left (liberal) political leanings (i.e., Aljazeera  and Iranian Diplomacy).  These websites are more optimistic towards Iran’s nuclear talks. Accordingly, the comments posted on both of these websties tended to be more rational. While there is a high percentage of explanation, persuation, and motivation in the comments posted on Aljazeera, users of Iranian Diplomacy intended to explain, inform, persuade, motivate, anger, amuse, and offend more often.

In between, there are news agencies with moderate (almost neutral) political leanings which often attempt to present news stories neutrally, without taking any sides (i.e., Rueters and Tabnak). These news websites often adapt an analytical view in their news articles. Although the comments posted on both of the moderate news agencies have a high degree of emotionality, the purposes intended by each are different. For instance, Rueters’ users intended to warn, inform, amuse, and threaten more often than users posting on the other two ENWs, while Tabnak users intended to defend more than the users of the other PNWs.

The authors acknowledge certain limitations associated with this exploratory study. The news stories and comments selected for our study were chosen based on a narrowed-down topic (i.e., Iran’s nuclear issue), which limits the generalizability of the findings. Future research on  the analysis of comments should broaden the scope to include other topics to better address the issue of generalizability. Despite the use of idependent raters and focus on the dominant tone of the comments, the authors acknowledge that tone analysis, especially in the political discourse, can be notoriously difficult and quite subjective. There is much future work to be done in this area, such as analyzing the figures of speech applied by commenters and delving into the different strategies used by commenters to achieve their intended purposes. Also, in order to yield more generalizable results, future studies may focus on larger sample of comments taken from more sources.



Al‐Saggaf, Y. (2006). The online public sphere in the Arab world: The war in Iraq on the Al Arabiya website. Journal of ComputerMediated Communication, 12(1), 311-334.         [ Links ]

Billig, M. (1991). Ideology and opinions: Studies in rhetorical psychology. Thousand Oaks, CA: Sage.         [ Links ]

Bruns, A. (2007). Produsage. Proceedings of the 6th ACM SIGCHI Conference on Creativity & Cognition, USA, 99-106. doi: 10.1145/1254960.1254975        [ Links ]

Chana, C., & Tangkiengsirisin, S. (2012). Comparative rhetoric of the discourse structure of news stories between news articles published in Thailand and in the USA. LEARN Journal: Language Education and Acquisition Research Network, 5(1), 97-109.         [ Links ]

Cohen, J. A. (1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20(1), 37-46.         [ Links ]

Condor, S., Tileaga, C., & Billig, M. (2013). Political rhetoric. In L. Huddy, D. O. Sears & J. S. Levy (Eds.), Oxford handbook of political psychology (pp. 262-300). New York, NY: Oxford University Press.

D'Alessio, D., & Allen, M. (2000). Media bias in presidential elections: A meta‐analysis. Journal of Communication, 50(4), 133-156.         [ Links ]

Diakopoulos, N., & Naaman, M. (2011a). Topicality, time, and sentiment in online news comments. Proceedings of the CHI'11 Extended Abstracts on Human Factors in Computing Systems, USA, 1405-1410. doi: 10.1145/1979742.1979782        [ Links ]

Diakopoulos, N., & Naaman, M. (2011b). Towards quality discourse in online news comments. Proceedings of the ACM 2011 conference on Computer supported cooperative work, USA, 133-142. doi: 10.1145/1958824.1958844        [ Links ]

Domingo, D., Quandt, T., Heinonen, A., Paulussen, S., Singer, J. B., & Vujnovic, M. (2008). Participatory journalism practices in the media and beyond: An international comparative study of initiatives in online newspapers. Journalism Practice, 2(3), 326-342.         [ Links ]

Glenn, N. M., Champion, C. C., & Spence, J. C. (2012). Qualitative content analysis of online news media coverage of weight loss surgery and related reader comments. Clinical Obesity, 2(5-6), 125-131.         [ Links ]

Groseclose, T., & Milyo, J. (2005). A measure of media bias. The Quarterly Journal of Economics, 120(4), 1191-1237.         [ Links ]

Hughey, M. W., & Daniels, J. (2013). Racist comments at online news sites: A methodological dilemma for discourse analysis. Media, Culture & Society, 35(3), 332-347.         [ Links ]

LaRose, R., & Eastin, M. S. (2004). A social cognitive theory of Internet uses and gratifications: Toward a new model of media attendance. Journal of Broadcasting & Electronic Media, 48(3), 358-377.         [ Links ]

Leung, L. (2009). User-generated content on the internet: An examination of gratifications, civic engagement and psychological empowerment. New Media & Society, 11(8), 1327-1347.         [ Links ]

McCluskey, M., & Hmielowski, J. (2012). Opinion expression during social conflict: Comparing online reader comments and letters to the editor. Journalism, 13(3), 303-319.         [ Links ]

McMillen, S. R. (2013). Threads of deliberation: A textual analysis of online news comments. (Doctoral dissertation). Retrieved from

McQuail, D. (1994). Mass communication theory: An introduction. London, UK: Sage.         [ Links ]

Kuypers, J. A. (2002). Press bias and politics: How the media frame controversial issues. Westport, CT: Greenwood.         [ Links ]

Maynard, S. K. (2002). Linguistic emotivity: Centrality of place, the topic-comment dynamic, and an ideology of pathos in Japanese discourse. Amsterdam, Philadelphia: John Benjamins.         [ Links ]

Milioni, D., Vadratsikas, K., & Papa, V. (2012). Their two cents worth: Exploring user agency in readers' comments in online news media. Observatorio, 6(3), 21-47.         [ Links ]

Miller, K. (2005). Communication theories: Perspectives, processes, and contexts. Boston, MA: McGraw-Hill.         [ Links ]

Nagar, N. (2011). The loud public: The case of user comments in online news media (Doctoral dissertation). Retrieved from ProQuest database. (UMI No. 3460834).

Pérez-Granados, D., Lozano-Garzón, C., López-Urueña, A., & Jiménez-Guarín, C. (2012). Sentiment analysis in Colombian online newspaper comments. In F. Gaol (Ed.), Recent progress in data engineering and internet technology (pp. 113–119). Heidelberg, Germany: Springer.

Petrova, M. (2008). Political economy of media capture. In R. Islam (Ed.), Information and public choice (pp. 121-139). Washington, DC: The World Bank.

Poor, N. (2005). Mechanisms of an online public sphere: The website slashdot. Journal of Computer-Mediated Communication, 10(2). doi:10.1111/j.1083-6101.2005.tb00241.x        [ Links ]

Santana, A. D. (2011). Online readers' comments represent new opinion pipeline. Newspaper Research Journal, 32(3), 66-81.         [ Links ]

Schneider, W., & Lewis, I. A. (1985). Views on the news. Public Opinion, 8(4), 6-11.         [ Links ]

Selzer, J. (2004). Rhetorical analysis: Understanding how texts persuade readers. In C. Bazerman & P. Prior (Eds.), What writing does and how it does it: An introduction to analyzing text and textual practices (pp. 279- 307). London, UK: Routledge.

Spitzer, R. L., Cohen, J. A., & Fleiss, J. L. (1968). Quantification of agreement in psychiatric

diagnosis: A new approach. Archives of General Psychiatry, 17(1), 83-87.

Spurr, D. (1993). The rhetoric of empire. Durham, NC: Duke University Press.         [ Links ]

S-TEAM (Science-Teacher Education Advanced Methods). (2010). Argumentative competence: How to develop it?. Trondheim: S-TEAM/NTNU.         [ Links ]

Streiner, D. L., Norman, G. R., & Cairney, J. (2014). Health measurement scales: A practical guide to their development and use. Oxford, UK: Oxford University Press.         [ Links ]

Sun, S., Kong, G., & Zhao, C. (2011). Polarity words distance-weight count for opinion analysis of online news comments. Procedia Engineering, 15, 1916-1920.         [ Links ]

Trakulkasemsuk, W., & Pingkarawat, N. (2010). A comparative analysis of English feature articles in magazine published in Thailand and Britain: Linguistics aspects. In R. Facchinetti, D. Crystal, & B. Seidlhofer (Eds.), Linguistic insights: Studies in language and communication (pp. 79-91). Bern, Switzerland: Peter Lang.         [ Links ]

Weatherly, J.N., Petros, T.V., Christopherson, K., & Haugen, E. (2007). Perceptions of political bias in the headlines of two major news organizations. The International Journal of Press/Politics, 12(2), 91-104.         [ Links ]

Zhitomirsky-Geffet, M., David, E., Koppel, M., & Uzan, H. (2016). Utilizing overtly political texts for fully automatic evaluation of political leaning of online news websites. Online Information Review, 40(3), 362-379.         [ Links ]



1 5+1 countries are the permanent members of the UN Security Council including China, France, Russia, The United States, The United Kingdom, and Germany.

Creative Commons License Todo o conteúdo deste periódico, exceto onde está identificado, está licenciado sob uma Licença Creative Commons