Content note: comments are quoted as written. Some contain insults, racist or sectarian language, or calls to violence. Discursa reports them because they are part of the conversation it measures; it does not endorse them. Whether to remove such content is a moderation decision for the platform and the page.
The most common position is a quarrel between commenters rather than an argument about the topic, held by 16% of items. Next is the view that religious practice is a personal choice, at 12% of items. At 7% of items each come the view that public life should follow the dominant faith's rules and the view that governments in the region oppress or fail their people. The personal-choice and dominant-faith positions oppose each other directly, and both centre on the Ramadan public-eating debate. Positions about the channel or its host are small: 2% of items hold a position supporting the channel and none opposing it, while 4% hold the reverse, based on the positions people hold. A further 45% of items hold no listed position, and 489 of those hold a clear view that is not on the list.
% of items holding each position, of the 3,596 read for positions. A comment can hold more than one, so the bars do not sum to 100%.
Derived from the positions people hold, using the analyst's tags of which positions support and which oppose the DW Jafar Talk channel; the model is never asked. <1% hold positions on both sides, 50% only positions that are neither, and 45% no listed position.
Among the positions held by 30 items or more, those in which more than 15% of items imply the position rather than state it in the comment itself are: "Religious practice is a personal choice that no state or institution should impose" in 30% of its items; "Governments in the region oppress or fail their people" in 19% of its items; "Public life should follow the religious majority's rules, such as not eating in public during Ramadan" in 16% of its items; "Women are restricted or mistreated and should have the same freedom as men" in 21% of its items; "Arabs and Muslims are judged by a double standard in the West" in 37% of its items; "The channel is biased or promotes a specific agenda" in 20% of its items. 666 items (19%) hold a view not on the list (489 of them hold no listed position), among them, in no order of size: "People should stop killing and torturing animals and adopt a plant-based diet"; "States have the right to restrict behaviours and expressions to preserve their cultural and social identity and prevent foreign influence"; "Abla Kamel is a humble and beloved actress".
| Period | Items | A quarrel with another commenter, not about the topic | Religious practice is a personal choice that no state or institution should impose | Governments in the region oppress or fail their people | Public life should follow the religious majority's rules, such as not eating in public during Ramadan | Women are restricted or mistreated and should have the same freedom as men |
|---|---|---|---|---|---|---|
| 16 February 2026 | 104 | 8% | 4% | 13% | 11% | 15% |
| 17 February 2026 | 392 | 21% | 33% | 8% | 14% | 3% |
| 18 February 2026 | 369 | 16% | 23% | 4% | 9% | 2% |
| 19 February 2026 | 295 | 14% | 7% | 6% | 3% | 9% |
| 20 February 2026 | 298 | 12% | 4% | 8% | 2% | 14% |
| 21 February 2026 | 178 | 16% | 1% | 4% | 1% | 3% |
| 22 February 2026 | 124 | 14% | 2% | 9% | 0% | 3% |
| 23 February 2026 | 237 | 19% | 9% | 8% | 3% | 5% |
| 24 February 2026 | 511 | 15% | 16% | 7% | 10% | 5% |
| 25 February 2026 | 423 | 16% | 11% | 9% | 7% | 2% |
| 26 February 2026 | 317 | 18% | 6% | 7% | 7% | 1% |
| 27 February 2026 | 177 | 23% | 5% | 5% | 3% | 1% |
| 28 February 2026 | 119 | 15% | 8% | 3% | 9% | 2% |
| 1 March 2026 | 52 | 19% | 8% | 4% | 6% | 2% |
By day, audience items whose positions were read and that carry a date; periods with fewer than 30 items are left out.
Comments are quoted as written; see the content note in the Summary.
One comment for each of the leading positions other than quarrels between commenters, translated in full. Every quoted comment is in the appendix.
% of the 3,596 items counted · tone of the item
Negative items (45% of items) are led by quarrels between commenters that are not about the topic. The next largest group argues that religious practice is a personal choice no state should impose, followed by complaints that governments in the region oppress or fail their people, and then the view that public life should follow the majority religion's rules. Others say Arabs and Muslims face a double standard in the West, in words of resentment or despair, and some say women are restricted and deserve the same freedom as men. Religious freedom, social media trends and public morality laws are among the subjects these items turn on, and some use contemptuous words about other people's looks, behaviour or morals. Neutral items (36% of items): about half of them (656 of 1,283) hold no listed position, and those that do lead with the argument that religious practice is a personal choice, then that public life should follow the religious rules of the people who live there, such as not eating in public during Ramadan. They argue in a calm, matter-of-fact way, with people correcting each other or offering a practical fix, such as bringing your own meal. Others in this group cover government failure, women's freedom, and disputes over which country a dish belongs to. Positive items (19% of items) mostly hold no listed position; among those that do, arguments over which country a dish or tradition belongs to and the view that religious practice is a personal choice are about level, ahead of calling the host charismatic and likable. These items use warm words, blessings, heart emojis and thanks, and they touch on cultural heritage, social media trends and Ramadan traditions.
% of items · theme (one item may carry multiple themes)
Not charted, at 0.5% of items or more: Racial Discrimination (4%), Economic Hardship (2%).
Commenters in this theme split over whether religion should be imposed in public life. 38% of items in this theme argue that religious practice is a personal choice no state or institution should impose, with some saying fasting cannot be forced on others and that respecting those who fast does not mean fasting with them. 14% of items in this theme argue that public life should follow the religious rules of the dominant faith in a country, such as not eating in public during Ramadan, and one says that people who claim to value democracy should respect other people's customs. Some commenters point out that Muslims make up most of Syria's population, so they say Muslims should set its laws. Another 21% of items in this theme are quarrels between commenters that drift away from the topic, such as disputes over Islamic doctrine, the reliability of hadith collections, and the direction of prayer. A further 8% of items in this theme argue that Arabs and Muslims face a double standard in the West.
Commenters in this theme quarrel with each other more than they discuss the channel: 30% of items in this theme are quarrels between commenters. They mock one another's writing ("try to form a useful sentence"), insult each other's country or upbringing, and tell each other not to speak about things they do not know. Others drift to the entertainment shows being discussed, with some calling a prank-show host trivial or immoral, one saying they have never watched an episode because it is unacceptable in form and content, and another saying it is only a comedy programme so no one should be upset. A smaller group says the channel has an agenda: 4% of items in this theme hold that the channel is biased or promotes a specific agenda. One commenter urges others to boycott the channel and DW, claiming they stir up conflict between Arab societies and incite against Islam. Another 4% of items in this theme argue that religious practice is a personal choice no state or institution should impose, and 3% of items in this theme call the content trivial or offensive.
Commenters split over whether public life should follow the rules of the religion that the larger part of the country practises. 27% of items in this theme say it should: they argue that Syria and Iraq are Muslim countries, that minorities must respect the larger religious group, and that eating in public during Ramadan is flaunting. Some say a person who does not fast can eat at home, and one asks whether a student has no shame eating in front of her classmates. Against them, 19% of items in this theme say religion is a personal choice that no state should impose. They point to the sizeable non-Muslim population in Iraq and ask what basis a law has to bind everyone; one notes that it would stop even children from eating. Another 14% of items in this theme are quarrels between commenters rather than arguments about the topic, and 10% say governments in the region oppress or fail their people.
In this theme, 27% of items argue that a dish or tradition belongs to one country and that neighbours copied it. Commenters from North Africa trade claims over couscous and tagine. One says Moroccan cookbooks spread into western Algeria before the border closed, and that Morocco exports its traditions there. A Tunisian commenter says that couscous is shared Amazigh heritage across 4 countries (Libya, Tunisia, Algeria and Morocco), and one says harira reached Algeria from Morocco through traders or migration. Another 8% of items in this theme are quarrels between commenters that drop the topic and become insults about history, poverty and colonial rule. A smaller group argues that religious practice is a personal choice no state should impose (7% of items in this theme). A further 5% of items in this theme say public life should follow the rules of the dominant religion, such as not eating in public during Ramadan. One commenter working in Europe says he misses family gatherings.
In this theme, 32% of items say governments in the region oppress or fail their people. One commenter says Iraq has no rule of law, only religious parties playing on sectarian feeling for their own gain, and another that Islamic parties pass laws that strip Iraq of democracy, and one jokes that even America treats migrants as if it were a Middle Eastern state that crushes freedoms. Another 24% of items in this theme argue that religious practice is a personal choice that no state or institution should impose. One commenter says people in the Arab world want others to know God "by force, not by reason and not by faith", others object to beliefs being imposed by force, and accuse some Muslim societies of denying other faiths freedoms they demand for themselves. A further 12% of items in this theme say Arabs and Muslims are judged by a double standard in the West. One points to bans on the headscarf in European countries, the demolition of mosques in India, and children taken from Muslim families. In 19% of items in this theme, commenters quarrel with each other instead of discussing the topic, trading insults and accusations such as ignorance or sexual abuse by clerics.
% of items · narrative frame used (one item may use multiple)
Not charted, at 0.5% of items or more: Appeal to Expertise (1%), Economic Reasoning (1%).
| Position | Most used frames among its holders |
|---|---|
| A quarrel with another commenter, not about the topic | Sarcasm and Ridicule 38%; Insult and Abuse 37%; Moral Condemnation 26% |
| Religious practice is a personal choice that no state or institution should impose | Moral Condemnation 41%; Sarcasm and Ridicule 18%; Comparison with Elsewhere 14% |
| Governments in the region oppress or fail their people | Moral Condemnation 47%; Sarcasm and Ridicule 22%; Comparison with Elsewhere 20% |
| Public life should follow the religious majority's rules, such as not eating in public during Ramadan | Moral Condemnation 60%; Comparison with Elsewhere 20%; Sarcasm and Ridicule 14% |
| Women are restricted or mistreated and should have the same freedom as men | Moral Condemnation 51%; Sarcasm and Ridicule 18%; Comparison with Elsewhere 12% |
| Arabs and Muslims are judged by a double standard in the West | Comparison with Elsewhere 46%; Moral Condemnation 28%; Sarcasm and Ridicule 25% |
| A dish or tradition belongs to one country and others copied it | Personal Testimony 32%; Comparison with Elsewhere 27%; Sarcasm and Ridicule 24% |
| The channel is biased or promotes a specific agenda | Moral Condemnation 51%; Sarcasm and Ridicule 26%; Comparison with Elsewhere 14% |
| Women's images or conduct in public should be restricted for decency | Moral Condemnation 66%; Sarcasm and Ridicule 15%; Question 12% |
| The content is trivial or offensive | Sarcasm and Ridicule 46%; Moral Condemnation 21%; Comparison with Elsewhere 14%; Personal Testimony 14% |
| Jafar is a charismatic and likable host | Blessing or Greeting 62%; Personal Testimony 23% |
| Jafar is arrogant or distasteful | Insult and Abuse 40%; Moral Condemnation 24%; Sarcasm and Ridicule 24% |
| Speaks well of the DW Jafar Talk channel, without giving a reason | Blessing or Greeting 77% |
| The content is engaging and high-quality | Sarcasm and Ridicule 40%; Blessing or Greeting 30% |
| The channel sheds light on important human issues | Personal Testimony 30% |
| Speaks badly of the DW Jafar Talk channel, without giving a reason | Insult and Abuse 57% |
| The channel is irrelevant or out of touch | Question 75% |
Of the items holding each position, the share using each frame. An item can use more than one frame. Each row shows its three most used frames, more where tied; frames held by fewer than 3 of a position's items are left out. 8 of these groups hold fewer than 30 items (4 to 28), so each of their shares stands for a few comments, not a steady rate; any table that breaks a figure down by source, period or dialect shows no shares within such groups. "The channel is courageous and principled" is left out: none of its frames is used by 3 of its items.
Frame descriptions
| Moral Condemnation | This frame is the largest, at 24% of items, and appears as judgments that someone lacks shame, should bear the consequences of defying a conservative society, or holds a hostile attitude toward Islam. |
| Sarcasm and Ridicule | This frame covers 19% of items and appears as laughing emojis, mocking remarks about dishes or religion, and jabs exchanged between commenters in quarrels. |
| Personal Testimony | This frame appears in 10% of items, with commenters speaking from their own situation, such as a student with a medical condition, or commenters who avoid eating in front of those fasting out of respect. |
| Comparison with Elsewhere | This frame appears in 10% of items and sets one place against another, such as Germany, Europe, the United States, China or Syria, to argue about respect for the religion of those around you or about standards of living. |
| Question | This frame appears in 9% of items, as hypothetical scenarios, challenges to other commenters, and pointed queries about why a rule or practice exists. |
| Blessing or Greeting | This frame appears in 8% of items and takes the form of prayers, Ramadan wishes, supplications for Palestine and Sudan, and warm greetings to other commenters or to Algerians. |
| Insult and Abuse | This frame appears in 7% of items. Its definition: slurs, curses or name-calling aimed at a person or group. |
| Religious Citation | This frame appears in 3% of items. Its definition: quoting scripture, hadith or a religious authority to make the point. |
| Plea or Demand | This frame appears in 3% of items. Its definition: asking or demanding that someone act. |
| Day | Volume | Tone |
|---|---|---|
| 16 February 2026 | 123 | 16%↑ 41%↓ |
| 17 February 2026 | 433 | 10%↑ 48%↓ |
| 18 February 2026 | 399 | 19%↑ 49%↓ |
| 19 February 2026 | 346 | 27%↑ 45%↓ |
| 20 February 2026 | 326 | 21%↑ 49%↓ |
| 21 February 2026 | 183 | 24%↑ 46%↓ |
| 22 February 2026 | 126 | 23%↑ 52%↓ |
| 23 February 2026 | 286 | 12%↑ 48%↓ |
| 24 February 2026 | 558 | 15%↑ 45%↓ |
| 25 February 2026 | 490 | 20%↑ 40%↓ |
| 26 February 2026 | 329 | 24%↑ 43%↓ |
| 27 February 2026 | 192 | 21%↑ 41%↓ |
| 28 February 2026 | 122 | 24%↑ 43%↓ |
| 1 March 2026 | 55 | 23%↑ 33%↓ |
Volume counts every dated item, those the audience did not write included. Tone shares divide by the audience's items on that day.
The items start at 16:26 UTC on 16 February and end at 19:23 UTC on 1 March, so the first and last bars cover part of a day. Read the ends as partial, not as a ramp or a collapse.
Daily activity peaked on 24 February at 558 items, after 433 items on 17 February, the first full day collected. The conversation then sagged to 126 items on 22 February before climbing back to the peak, so it moved in waves rather than a single steady build. After the peak, volume fell every day: 490 items on 25 February, 192 on 27 February and 55 on 1 March. The threads under the 24 February posts faded (257 audience items that day, 41 on 27 February), and the nine posts of the last three days drew few comments (72 shown on them when collected).
% of items · language, or Arabic dialect where the text is Arabic
The bars above cover 3,405 of the 3,968 items (86%). The other 563: 370 published by the pages themselves, which carry no dialect; 2 holding only punctuation; 191 the report could not place in one of its six varieties: 178 unclear, 2 in other languages or varieties, 8 mixing two languages or varieties and 3 that are only a tagged name or a platform translation, not counted as English.
| Dialect | Items | A quarrel with another commenter, not about the topic | Religious practice is a personal choice that no state or institution should impose | Governments in the region oppress or fail their people | Public life should follow the religious majority's rules, such as not eating in public during Ramadan | Women are restricted or mistreated and should have the same freedom as men | Arabs and Muslims are judged by a double standard in the West | A dish or tradition belongs to one country and others copied it |
|---|---|---|---|---|---|---|---|---|
| MSA | 1,583 | 13% | 15% | 8% | 7% | 5% | 3% | 3% |
| Levantine | 781 | 22% | 15% | 9% | 11% | 4% | 7% | 1% |
| Gulf | 442 | 21% | 12% | 5% | 9% | 5% | 2% | <1% |
| Egyptian | 358 | 19% | 9% | 4% | 4% | 7% | 1% | 0% |
| Maghrebi | 174 | 17% | 5% | 1% | 2% | 2% | 0% | 34% |
| English | 67 | 9% | 0% | 15% | 0% | 13% | 1% | 1% |
Share of each dialect group's items; groups under 30 items are left out.
Modern Standard Arabic items are the largest group at 40% of items, followed by Levantine-speaking commenters at 20% of items and Gulf-speaking commenters at 11% of items. Tone differs by variety: 57% of Levantine items and 55% of Gulf items are negative in tone, against 40% of Modern Standard Arabic items and 38% of Maghrebi items. English items are the least negative at 21% of English items, though that group is small. On what people argue, the largest groups share the same top concerns: quarrels with other commenters (22% of Levantine items and 21% of Gulf items) and the view that religious practice is a personal choice no state should impose (15% of Levantine items and 15% of Modern Standard Arabic items). Maghrebi-speaking commenters stand apart: 34% of their items argue that a dish or tradition belongs to one country and others copied it.
The five most-engaged accounts together hold 13% of the engagement the audience's items drew, so engagement is not concentrated in a few hands. Voice_1 leads with an engagement of 280 across 45 items, and its tone is mixed with no single position leading. Voice_5 is also mixed with no single position leading. Voice_2, Voice_3 and Voice_4 each hold the same main position: religious practice is a personal choice that no state or institution should impose. Voice_2 holds it in 21 of its 48 items with a neutral tone; it is also one of the accounts that posted the same message repeatedly, so its item count includes copies. Voice_3 holds it in 11 of 23 items and Voice_4 in 6 of 15, and both write in a negative tone. The pattern is a scattered audience. Voice_1, the most-engaged account, often mocks Islam and the Prophet Muhammad (about a dozen of its 45 items, some likening him to Epstein), and Voice_4 argues for personal religious freedom in openly hostile terms, calling Islam the enemy of all humanity. The pages' own posts and replies drew 82% of all likes, far more than these accounts drew. Voice_11 wrote the most, with 66 items, and it is not among these five.
| Account | Platform | Items | Engagement | Share of Voice ⓘ | Main position |
|---|---|---|---|---|---|
| Voice_1 | YouTube | 45 | 280 | 3.6% | No main position |
| Voice_2 | YouTube | 48 | 208 | 2.7% | Religious practice is a personal choice that no state or institution should impose |
| Voice_3 | YouTube | 23 | 187 | 2.4% | Religious practice is a personal choice that no state or institution should impose |
| Voice_4 | YouTube | 15 | 166 | 2.1% | Religious practice is a personal choice that no state or institution should impose |
| Voice_5 | YouTube | 4 | 149 | 1.9% | No main position |
| Voice_6 | YouTube | 10 | 129 | 1.6% | Religious practice is a personal choice that no state or institution should impose |
| Voice_7 | YouTube | 26 | 113 | 1.4% | No main position |
| Voice_8 | YouTube | 19 | 110 | 1.4% | No main position |
| Voice_9 | YouTube | 17 | 110 | 1.4% | No main position |
| Voice_10 | YouTube | 34 | 109 | 1.4% | No main position |
| Voice_11 | YouTube | 66 | 53 | 0.7% | Religious practice is a personal choice that no state or institution should impose |
| Voice_12 | YouTube | 60 | 46 | 0.6% | A quarrel with another commenter, not about the topic |
Share of Voice: each account's engagement as a percentage of the engagement recorded across every item the audience wrote in this dataset. Engagement = likes and replies. Shares are not counted. It is a share of the reactions these items drew, not of how much of the conversation an account wrote: the items column is that. The pages' own posts and replies drew 82% of all likes and are not in this table. The table lists the ten accounts with the most engagement, then the two that wrote the most among the rest (Voice_11 to Voice_12).
| Items | Favourable | Unfavourable | |
|---|---|---|---|
| MSA | 1,583 | 2% | 4% |
| Levantine | 781 | 2% | 4% |
| Gulf | 442 | 2% | 4% |
| Egyptian | 358 | 2% | 4% |
| Maghrebi | 174 | 1% | 4% |
| English | 67 | 1% | 1% |
Favourable and unfavourable come from the analyst's tags on each position; the model is never asked. An item holding positions on both sides counts in both. Groups under 30 items are left out.
| Side | Items | Most used frames |
|---|---|---|
| Favourable | 61 | Blessing or Greeting 51%; Personal Testimony 15%; Moral Condemnation 10%; Sarcasm and Ridicule 10%; Plea or Demand 7% |
| Unfavourable | 133 | Moral Condemnation 38%; Sarcasm and Ridicule 29%; Insult and Abuse 14%; Comparison with Elsewhere 11%; Question 8% |
Of the items on each side, the share using each frame. An item can use more than one frame. An item holding positions on both sides counts in both.
Favourable and unfavourable positions are both small everywhere: on YouTube, the only platform, 2% of items are favourable and 4% unfavourable. Egyptian-dialect items (2% favourable, 4% unfavourable), Gulf-dialect items (2% and 4%), Levantine items (2% and 4%) and Modern Standard Arabic items (2% and 4%) sit close together. Maghrebi-dialect items are 1% favourable and 4% unfavourable, and English items hold 1 favourable and 1 unfavourable item of 67, too few to read. Favourable and unfavourable items argue differently. Among favourable items, 51% are blessings or greetings and 15% are personal testimony. Among unfavourable items, 38% argue through moral condemnation and 29% use sarcasm and ridicule.
2 signals detected across the 3,598 items the audience wrote (the 3,596 counted in the other figures and 2 items holding only punctuation).
Not checked: uniform hashtags: fewer than 10 items carry hashtags.
Context: the 3,598 items the audience wrote average 11.4 items/hour across the 315 hours the items span.
The scanner fired 2 of the 5 indicators it could check, which gives a LOW rating. The first is repeated text from single accounts: some accounts posted the same message more than once, and one account posted near-identical items again and again. In all, 31 items were repeated this way, a very small share of items. The second is burst posting: 3 bursts in which a large number of distinct accounts posted in the same half-hour, and together the busy half-hours hold 6% of items. Every burst opened soon after a post was published on the channel, and every busy half-hour fell within a day of one, which is what an audience arriving looks like. The scanner cannot tell this from coordination. The repeated messages come from 4 accounts, each repeating its own text, and the scan found no text copied across accounts. Seven of the 104 accounts in the bursts post in more than one; each is a regular commenter here, with 11 to 66 items and differing positions, which is what a returning audience looks like. The uniform hashtag check was not run, because too few items carry hashtags.
This analysis covers comments from the DW Jafar Talk channel's 100 posts, and it counts what the people who commented here argued, not what they want more of. Moderate first the threads under the fasting posts, where the argument over religion in public life sits: 12% of the 3,596 audience items hold that religious practice is a personal choice that no state or institution should impose, and 67% of those items sit under its two largest posts, the posts on the Iraqi interior ministry's circular against eating in public during Ramadan and a Syrian university's ban on the same, of the 37 posts the position's items appear under. Across the set, 7% of items hold that public life should follow the religious rules of those around them, such as not eating in public during Ramadan, so both sides of that argument are present, and the choice between them is a value question that belongs to the editorial team rather than to this report. What the team can do is set a moderation approach for those threads, since quarrels between commenters make up 21% of items under those two posts (158 of 769), against 16% across the set, and the channel's own repeated reminder about civil discussion appeared under 37 posts.
It is the most held unfavourable claim with a reason behind it (see 03). Where the claim is a matter of fact, it can be answered with the facts; where it is a judgement, say what the choice was and why.
Comments are quoted as written; see the content note in the Summary.
Each comment is shown whole, as collected, with a translation of the whole comment. Grouped by the first position each holds.
The channel itself, DW Jafar Talk, is named most, in 72 items, followed by Syria in 68 items and Iraq in 60 items. The leading sentiment for the channel is neutral and for Syria it is mixed, while Egypt, named in 40 items, is also mixed. The leading sentiment is positive for Algeria, named in 54 items, and for Morocco, named in 34 items. It is negative for Iraq, Europe (48 items), the USA (49 items), Jeffrey Epstein (33 items) and Germany (33 items).
| Entity | Type | Mentions | Leading sentiment |
|---|---|---|---|
| DW Jafar Talk (the channel) | Media | 72 | Neutral |
| Syria (سوريا) | Location | 68 | Mixed |
| Iraq (العراق) | Location | 60 | Negative |
| Algeria (الجزائر) | Location | 54 | Positive |
| USA (أمريكا) | Location | 49 | Negative |
| Europe (أوروبا) | Location | 48 | Negative |
| Egypt (مصر) | Location | 40 | Mixed |
| Morocco (المغرب) | Location | 34 | Positive |
| Jeffrey Epstein (ابستين) | Person | 33 | Negative |
| Germany (المانيا) | Location | 33 | Negative |
Mentions: items in which the classifier listed the entity; a word search can find more. Leading sentiment: the sentiment the classifier recorded toward the entity where it recorded one, otherwise the tone of the item that names it. The two can differ: a comment angry at those who insulted someone is negative in tone and positive toward him. Mixed means the leading tone led the runner-up by a fifth or less.
| Label | Question | Example from this report | How it is set |
|---|---|---|---|
| Position | What does the comment claim? | “Religious practice is a personal choice that no state or institution should impose” | From a list proposed from the data and approved by the analyst before any comment is classified. A comment can hold several positions, or none. |
| Theme | What is it about? | Religious Freedom | Neutral topics drawn from this conversation. A theme names a subject, never a verdict on it; verdicts are positions. |
| Frame | How does it argue? | Moral Condemnation | One fixed set of eleven ways of arguing (ridicule, insult, moral condemnation, religious citation, personal testimony, economic reasoning, expertise, comparison, plea, question, greeting), defined the same way in every report so frames can be compared across reports. |
| Tone | How does it sound? | Positive, negative or neutral | The emotional register of the comment itself, whatever it is about: a blessing is positive, an insult negative. Tone is not a view of the subject; that is what positions record. |
For and against are not asked of the model. The analyst tags each position as supporting, opposing or neither; a comment counts as for or against from the positions it holds.
Classification, writing and translation use a number of the latest frontier language models.
According to our internal evaluation, the classification of dialect and tone is between 80 and 90 percent accurate. Read a difference of a few points between two figures as no difference.
Some comments contain offensive language: insults, slurs and attacks on groups or individuals. The platforms and their moderation left these comments up, so they are part of the conversation this report describes: they are counted with every other comment, and some appear among the quoted examples, reproduced as collected, so that a position is shown as its holders put it. The language and content of the comments do not reflect the views of the Discursa team, and the team does not endorse them.