Welcome to the Interactive Data Visualisation

On this page you can examine and compare the characteristics of antisemitic comments from different discourse events. To do this, select two discourse events at the upper left of this page that you want to compare.
The discourse-specific information and visualisations are made comparable in two main columns. A short description of the content and some general information about the data as well as the number of (antisemitic) comments are displayed in the top row. The different visualisations can be selected and are displayed in the lower row.
For each of these visualisations, there are settings that can be used to adjust the display of the results. To help you understand each graph type and the results, there is an interpretation example for the default discourse events. The visualisations are fully interactive. Move the mouse over graphical elements, zoom in or move elements to get a better insight.
The different visualisations can be selected via the tabs in this part of the window. You can choose to display:

  1. the frequency distributions of the different labels
  2. the co-occurrence of these labels
  3. the frequency distribution of antisemitic specific keywords
  4. the network of these keywords

Author and Replication Materials

Stefan Munnes (WZB; Decoding Antisemitism)
Replication Material available on GitHub.
Many thanks to the Berlin Social Science Center (WZB) for hosting this application.

Plot options

Of the classic antisemitic tropes, ‘Evil/The Devil’ is the most common in social media comments on the Arab-Israeli conflict. This attribution is found, depending on the country, in 30% to 40% of the antisemitic comments in 2021. It is most common in German data, least common in French data. Compared to the 2023 discourse event, for which data is only available for the UK, the percentage has decreased to just over 10%.

Classic Antisemitic Tropes

Label Concept name
Evil_Devil Evil/The Devil
Child_Murder Blood Libel/Child Murder
Repul_Dehum Repulsiveness and Dehumanisation
Immorality Immorality
Lie_Deceit Lie and Deceit
Vengefulness Vengefulness
Disloyalty Disloyalty/Jewish Loyalty
Blame_For_AS Blame for Antisemitism

Tropes of Political or Financial Power

Label Concept name
Greed Greed, Exploitation and Identification with Capitalism
Power Power
Conspiracy Conspiracy Theories
Disintegration Disintegration
Self_Victim Self-victimisation

Secondary Antisemitism

Label Concept name
Reject_Guilt Rejection of Guilt
Clean_Break Clean Break (Schlussstrich)
Dist_Den_Hol Distortion and Denial of the Holocaust
Rel_Den_AS Relativisation and Denial of Antisemitism
Instr_AS_Hol Instrumentalisation of Antisemitism and the Holocaust
NZ_Jew_Collab Nazi-Jewish Collaboration
Admonisher Admonishers
Taboo_Crit Taboo of Criticism
Not_Learn_Past Jews Have Not Learned from the Past

Further Post-Holocaust Concepts

Label Concept name
Jews_Resp_IL Holding Jews Collectively Responsible for Israel’s Actions
Privilege Jewish Privilege and the ‘Free Pass’
Pro_NZ_Hit_Hol Affirmation of Nazis, Hitler, and the Holocaust

Attacks on Israel’s Legitimacy

Label Concept name
NZ_Anlgy Nazi Analogy
Fasc_Anlgy Fascism Analogies
Apartheid_Anlgy Apartheid Analogy
Racist_State Racist State
Colonial_Anlgy Colonialism Analogies
Terror_State Terrorist State
Genocide Genocide
Dble_Standards Double Standards
Den_Right_Exist Denial of Israel’s Right to Exist
BDS_Boycott BDS/Calls for Boycott
IL_Sole_Guilt Israel’s Sole Guilt in the Conflict

Speech Acts

Label Concept name
Insults Insults
Threats Threats
Curses Curses
Death_Wishes Death Wishes
Pro_Violence Affirming, Desiring, Calling for Violence

Plot options

At both points in time in the discourse event of the Arab-Israeli conflict, the label ‘Israel’s Sole Guilt in the Conflict’ appears most frequently with other labels in the same comment, mainly with the classic antisemitic trope of 'Evil/The Devil'. Compared to 2021, in the 2023 comments Israel-related antisemitic labels are much more frequently combined with the speech act of 'Affirming, Desiring, Calling for Violence’.

Classic Antisemitic Tropes

Label Concept name
Evil_Devil Evil/The Devil
Child_Murder Blood Libel/Child Murder
Repul_Dehum Repulsiveness and Dehumanisation
Immorality Immorality
Lie_Deceit Lie and Deceit
Vengefulness Vengefulness
Disloyalty Disloyalty/Jewish Loyalty
Blame_For_AS Blame for Antisemitism

Tropes of Political or Financial Power

Label Concept name
Greed Greed, Exploitation and Identification with Capitalism
Power Power
Conspiracy Conspiracy Theories
Disintegration Disintegration
Self_Victim Self-victimisation

Secondary Antisemitism

Label Concept name
Reject_Guilt Rejection of Guilt
Clean_Break Clean Break (Schlussstrich)
Dist_Den_Hol Distortion and Denial of the Holocaust
Rel_Den_AS Relativisation and Denial of Antisemitism
Instr_AS_Hol Instrumentalisation of Antisemitism and the Holocaust
NZ_Jew_Collab Nazi-Jewish Collaboration
Admonisher Admonishers
Taboo_Crit Taboo of Criticism
Not_Learn_Past Jews Have Not Learned from the Past

Further Post-Holocaust Concepts

Label Concept name
Jews_Resp_IL Holding Jews Collectively Responsible for Israel’s Actions
Privilege Jewish Privilege and the ‘Free Pass’
Pro_NZ_Hit_Hol Affirmation of Nazis, Hitler, and the Holocaust

Attacks on Israel’s Legitimacy

Label Concept name
NZ_Anlgy Nazi Analogy
Fasc_Anlgy Fascism Analogies
Apartheid_Anlgy Apartheid Analogy
Racist_State Racist State
Colonial_Anlgy Colonialism Analogies
Terror_State Terrorist State
Genocide Genocide
Dble_Standards Double Standards
Den_Right_Exist Denial of Israel’s Right to Exist
BDS_Boycott BDS/Calls for Boycott
IL_Sole_Guilt Israel’s Sole Guilt in the Conflict

Speech Acts

Label Concept name
Insults Insults
Threats Threats
Curses Curses
Death_Wishes Death Wishes
Pro_Violence Affirming, Desiring, Calling for Violence

Shown are the 30 most clearly antisemitic keywords from the two discourse events on the Arab-Israeli conflict in 2021 and 2023 that appear in at least 5 comments. These words are more likely to appear in comments labelled as antisemitic and were calculated using the chi-squared independence test. They may also appear in non-antisemitic comments but with a lower probability, taking into account the uneven distribution.

For the two discourse events, some keywords remain relatively constant and are formative for the entire discourse, e.g. 'apartheid' or 'zionist'. However, other keywords are more specific to the respective antisemitic discourse, e.g. 'settlers' or 'unarmed' in 2021 and 'occupation' or 'genocide' in 2023.

Overall, there is a strong reference to political subjects ('palestinians', 'israeli') as well as strong negative descriptions of activities ('oppression', 'kill', 'cleansing').

These network plots show the relationships between keywords. Node size indicates how many comments the keyword appears in, and edge thickness indicates how often keywords appear together in a comment. For a better overview, only the 30% most important connections are shown.

For the discourse event Arab-Israeli conflict in 2021, the central importance of the two conflict parties 'palestinians' and 'israel' is shown, which are also strongly linked. It is therefore not surprising that most of the keywords are relevant for both terms. There are some small differences, such as the one-sided reference to 'oppression' and 'palestinians', as well as the stronger connection between 'israel' and 'killing' and 'terrorist'.

A similar picture emerges for the year 2023. Particularly striking is the slogan 'free' and 'palestine' plus 'emoji', which indicates a stronger political mission on the one hand, and an attribution of guilt on the other with 'regime', 'israel', 'apartheid', 'illegal' and 'occupation'.