Right-Wing Rock or Just Rock? A Computational Linguistic Analysis of Frei.Wild

Carlotta Schneeberger; Kevin Tang

cs.CLSep 30, 2026

Right-Wing Rock or Just Rock? A Computational Linguistic Analysis of Frei.Wild

Authors: Carlotta Schneeberger, Kevin Tang

Organizations: Heinrich Heine University Düsseldorf · Faculty of Arts and Humanities, Heinrich Heine University Düsseldorf 1Institute of Linguistics · University of Florida · Department of English Language and Linguistics ·

Abstract

Rechtsrock is a subgenre of rock music that spreads right-wing ideology, often instrumentalized to recruit adolescents into the radical scene. Monitoring institutions counteract this by manually examining and, in some cases, banning extremist content; however, there are border cases that evade regulation. We present a study aimed at determining whether such a case, the band Frei $.$ Wild, should be classified as politically right-leaning or as part of the general German rock genre. We sampled a German rock dataset and created a corpus for right-wing rock to use as reference in this analysis and found that we can confirm the intuitions from previous investigations that Frei $.$ Wild successfully maintains an ambiguity with regard to their political affiliation. However, the tendency is towards the right-wing spectrum. Lexical analyses reveal nationalistic narratives and two high-performing classifiers (up to 97% ROC-AUC score) label more than half of their songs as right-wing extremist. Our analysis provides insight into how computational methods can improve the process of identifying right-wing extremist tendencies in music, especially in borderline cases like Frei $.$ Wild. The code and data are made available for future research.

Figures & tables

Corpus	Songs	Bands	1990s	2000s	2010s	avg. song len
Rechtsrock	99	9	22	57	20	224 words
German Rock	99	9	22	55	22	237 words
Frei.Wild	47	1	0	25	22	264 words
Frei.Wild (full)	303	1	0	68	235	243 words

Table 1: Summary statistics of the three corpora.

Figure 1: A heatmap showing the cosine similarities between bands from the Rechtsrock corpus ( red ), the Deutschrock corpus (black) and the band Frei.Wild ( blue ). Acronyms: ‘DLV’ = Die Lunikoff-Verschwörung ; ‘AMK’ = AnnenMayKantereit ; ‘GG’ = Großstadtgeflüster ; ‘MMW’ = Marius Müller-Westernhagen ; ‘SpS’ = Sportfreunde Stiller

Figure 2: A density plot with means of the test data predictions from the TF-IDF classifier for Rechtsrock ( red ), Deutschrock (black) and the unlabeled data from Frei.Wild ( blue ).

Figure 3: A density plot with means of the test data predictions from the sentence embedding classifier for Rechtsrock ( red ), Deutschrock (black) and the unlabeled data from Frei.Wild ( blue ).

Figure 4: Two density plots for the classification scores of the Frei.Wild songs per year, separated by model.

Appendix figures & tables13 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 5: A density plot with means of the training data predictions from the TF-IDF classifier for Rechtsrock ( red ) and Deutschrock (black).

Figure 6: A density plot with means of the training data predictions from the sentence embedding classifier for Rechtsrock ( red ) and Deutschrock (black).

Figure 7: A density plot with means of the test data predictions from the TF-IDF classifier for Rechtsrock (red), Deutschrock (black) and the unlabeled data from Böhse Onkelz (blue).

Figure 8: A density plot with means of the test data predictions from the sentence embedding classifier for Rechtsrock (red), Deutschrock (black) and the unlabeled data from Böhse Onkelz (blue).

Figure 9: Two density plots for the classification scores of the Böhse Onkelz songs per year, separated by model.

German Title	English Translation
Sturmführer	storm-leader (lowest officer rank in Nazi Germany)
Bis hier her	this far (as in “so far and no further”)
Wahre Werte	true values
Südtirol	South Tyrol
Antiwillkommen	anti-welcome
Gutmenschen und Moralapostel	do-gooder and moralizer

Appendix

Table 2: German song titles and their English translations.

Frei.Wild song title	TF-IDF	sentence embedding
15 Jahre	67	44
20 Jahre Seite an Seite	47	57
AIDS	42	35
Akzeptierter Faschist	53	63
Alarm im Proberaum	56	36
Alkohol	44	44

Appendix

Table 3: Rechtsrock probability (in %) assigned to the individual Frei.Wild songs by the classifiers. If the value was >50, they were classified as Rechtsrock.

Frei.Wild song title	TF-IDF	sentence embedding
Corona Weltuntergang	49	52
Corona Weltuntergang V2	61	56
Daheim	22	47
Das gewisse Nichts	36	45
Das Land der Vollidioten	50	52
Dein zweites Leben	47	21

Appendix

Table 4: Rechtsrock probability (in %) assigned to the individual Frei.Wild songs by the classifiers. If the value was >50, they were classified as Rechtsrock. (continued)

Frei.Wild song title	TF-IDF	sentence embedding
1860	31	58
Frei.Wild	35	72
Frei.Wild’s Ländereien	57	66
Freiheit	60	48
Freiheit, Freundschaft, Brüderlichkeit	74	63
Freundschaft	45	38

Appendix

Table 5: Rechtsrock probability (in %) assigned to the individual Frei.Wild songs by the classifiers. If the value was >50, they were classified as Rechtsrock. (continued)

Frei.Wild song title	TF-IDF	sentence embedding
Ich helf dir auf	58	42
Ich kann auf die Fresse geben	58	59
Ich lache über dich	37	45
Ich und mein Scheiss	59	45
Ich weiß wer ich war	57	53
Ich werde mich jetzt nicht ändern	41	51

Appendix

Table 6: Rechtsrock probability (in %) assigned to the individual Frei.Wild songs by the classifiers. If the value was >50, they were classified as Rechtsrock. (continued)

Frei.Wild song title	TF-IDF	sentence embedding
Mehr als tausend Worte	45	37
Mein Lachen, dein Leiden	37	45
Mein Leben, meine Geschichte, meine Lehre	42	37
Meine Augen durch deine Augen	40	40
Mensch oder Gott	32	37
Mimimuttersöhnchen	40	23

Appendix

Table 7: Rechtsrock probability (in %) assigned to the individual Frei.Wild songs by the classifiers. If the value was >50, they were classified as Rechtsrock. (continued)

Frei.Wild song title	TF-IDF	sentence embedding
Sommerland	31	30
Sorgenleer	35	28
Spirit of 96	30	29
Steine deiner Mauer	40	47
Sternenstaub	55	52
Stück für Stück	54	48

Appendix

Table 8: Rechtsrock probability (in %) assigned to the individual Frei.Wild songs by the classifiers. If the value was >50, they were classified as Rechtsrock. (continued)

Frei.Wild song title	TF-IDF	sentence embedding
Weiter immer weiter	44	55
Wenn alles bricht	44	56
Wenn alles in Trümmern liegt	48	61
Wenn die Erinnerung erwacht	48	27
Wenn mein Licht erlischt	37	45
Wer nichts weiß wird alles glauben	52	33

Appendix

Table 9: Rechtsrock probability (in %) assigned to the individual Frei.Wild songs by the classifiers. If the value was >50, they were classified as Rechtsrock. (continued)

Political Discourse Analysis

Political discourse analysis studies how language expresses ideology, frames issues, persuades audiences, and shapes political debate.

See all 67 papers

Towards an LLM-based method for quantifying the sexual content in song lyrics
Aug 9, 2026Ignacio M. SticcoLarge Language Model-Based Text Classification
Ignacio M. Sticco
German parties shifted towards intuition-based rhetoric after the far right's parliamentary breakthrough
Aug 5, 2026Peer Saleth, Segun T. Aroyehun, Fabio Carrella +3Evidence-Grounded Reasoning Political Discourse Analysis
Department of Politics and Public Administration, University of Konstanz, Universitätsstraße 10, Konstanz, 78457, Baden-Württemberg, Germany. · Artificial Intelligence Lab (Recod.ai), Institute of Computing, University of Campinas, R. Saturnino de Brito, 573, 13083-852,2026 Campinas, S˜ao Paulo, Brazil. · Department of Psychology, University of Potsdam, Karl-Liebknecht-Str. 24–25, 14476, Potsdam, Brandenburg, Germany. +1
The Punjabi Geet Toxicity Dataset (PGT-FG): Fine-Grained Multimodal Annotations for Toxic Content in Punjabi Songs
Oct 23, 2026Lakhjeet Kaur, Jaspreet Kaur, Komal Rani +3Multimodal Dataset Toxicity Detection
Department of Computer Science, Guru Gobind Singh Indraprastha University, Delhi, India. · Department of Computer Science, Malwa College Bathinda, Punjab, India. · Department of Management, Akal University Talwandi Sabo, Punjab, India. +3

Aug 9, 2026J/K move · Enter open · S save

Towards an LLM-based method for quantifying the sexual content in song lyrics

Ignacio M. Sticco

Large Language Model-Based Text Classification

Open paper AskPDF arXiv

Reggaeton is one of the most widely consumed music genres in the world, and its lyrics are commonly regarded as highly sexualized. This claim rests mostly on qualitative studies and on small-scale quantitative ones. This paper has two goals. First, we present a reproducible method that uses a large language model to quantify thematic content in song lyrics along several independent dimensions. The method is not restricted to sexual content. Second, we apply it to a corpus of 1,259 songs by 12 reggaeton artists released between 2002 and 2025. The analysis covers four topics: a dataset characterization, a per-artist comparison, an analysis of how the dimensions change over time, and a comparison between our sexual-explicitness score and Spotify's own explicit flag. We release the data collection code, the scoring prompt, and the corpus, so that other researchers can replicate the approach or apply it to their own lyrics datasets.

Abstract

Political Discourse Analysis

Explore similar work

Towards an LLM-based method for quantifying the sexual content in song lyrics

German parties shifted towards intuition-based rhetoric after the far right's parliamentary breakthrough

The Punjabi Geet Toxicity Dataset (PGT-FG): Fine-Grained Multimodal Annotations for Toxic Content in Punjabi Songs