=Paper=
{{Paper
|id=Vol-3834/paper94
|storemode=property
|title=Revolution + Love: Measuring the Entanglements of State Violence and Emotions in Early PRC
|pdfUrl=https://ceur-ws.org/Vol-3834/paper94.pdf
|volume=Vol-3834
|authors=Maciej Kurzynski,Aaron Gilkison
|dblpUrl=https://dblp.org/rec/conf/chr/KurzynskiG24
}}
==Revolution + Love: Measuring the Entanglements of State Violence and Emotions in Early PRC==
https://ceur-ws.org/Vol-3834/paper94.pdf
Revolution + Love: Measuring the Entanglements of
State Violence and Emotions in Early PRC
Maciej Kurzynski1,∗ , Aaron Gilkison2
1
Advanced Institute for Global Chinese Studies, Lingnan University, Hong Kong
2
Department of East Asian Languages and Cultures, Stanford University, USA
Abstract
This paper examines the relationship between violent discourse and emotional intensity in the early
revolutionary rhetoric of the People’s Republic of China (PRC). Using two fine-tuned bert-base-chinese
models—one for detecting violent content in texts and another for assessing their affective charge—
we analyze over 185,000 articles published between 1956 and 1989 in the People’s Liberation Army Daily
(Jiefangjun Bao), the ofÏcial journal of China’s armed forces. We find a statistically significant correla-
tion between violent discourse and emotional expression throughout the analyzed period. This strong
alignment between violence and affect in ofÏcial texts provides a valuable context for appreciating how
other forms of writing, such as novels and poetry, can disentangle personal emotions from state power.
Keywords
violent discourse, sentiment analysis, People’s Liberation Army Daily, revolutionary rhetoric
1. Introduction
The concept of “Revolution Plus Love” (geming jia lian’ai 革命加恋爱) became prominent
during the New Culture Movement in China (ca. 1915-1919) and continued to shape Chinese
literary practice throughout the long twentieth century. It has also provided a lens through
which sinologists have examined socio-political changes in the Republic (1912–1949) and the
People’s Republic (1949-) of China. Jianmei Liu [25] shows how Chinese writers personalized
revolution and revolutionized their romantic adventures, often finding themselves confronted
with dilemmas between personal fulfilment and national ideals. Haiyan Lee [21] emphasizes
how sentimental discourse replaced the kin-based sociality that defined the pre-modern world
with a modern one that transformed strangers into compatriots. Eugenia Lean [20] investigates
a startling case of Shi Jianqiao (1905-1979), a woman who murdered the warlord Sun Chuan-
fang (1885-1935) and then managed to galvanize what Lean calls “public sympathy” to regain
freedom. Elizabeth Perry [29] focuses on the “emotion work” launched by the Communist
Party as a deliberate strategy of psychological engineering.
The theoretical premise of this paper is that both emotional engagement and violent dis-
course leave formal traces in texts which can be identified with the help of statistical methods
CHR 2024: Computational Humanities Research Conference, December 4–6, 2024, Aarhus, Denmark
∗
Corresponding author.
£ maciej.kurzynski@ln.edu.hk (M. Kurzynski); amgilki@stanford.edu (A. Gilkison)
ç https://www.qhchina.org (M. Kurzynski)
ȉ 0009-0006-5466-2423 (M. Kurzynski)
© 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
1012
CEUR
ceur-ws.org
Workshop ISSN 1613-0073
Proceedings
of literary inquiry. We build upon existing scholarship on the political signification of sen-
timents to suggest a computational perspective on the entanglements between violence and
affect in Chinese revolutionary discourse. In particular, we focus on the texts published be-
tween 1956 and 1989 in the People’s Liberation Army Daily (PLA Daily, or Jiefangjun Bao 解放
军报), one of the major PRC journals and the ofÏcial publication of China’s armed forces, to
analyze how such entanglements manifested in ofÏcially sanctioned documents.
2. Related Works
2.1. Violent Discourse and Hate Speech
“Violent discourse” refers to the use of language to inflict harm, perpetuate power structures,
and normalize physical violence [27, 1]. While it is closely related to “hate speech,” the two
categories are not identical. Violent discourse does not need to contain targeted abuse or foul
language and is often produced by public institutions rather than private individuals. Con-
versely, hate speech might include mockery and racial stereotypes without any direct link to
violent behavior, let alone military confrontations [36]. The distinction between “hate speech”
and “violent discourse” is productive in the analysis of ofÏcial publications in the early PRC,
which often decried racial discrimination in the United States [16, 6]. On the surface, the de-
tailed accounts of US racism contrasted the Chinese revolution with the malfeasance of the
capitalist world. In fact, such accounts served to promote state violence against the “enemies
of the People” identified within the country. In other words, the anti-hate rhetoric fueled vio-
lent behavior.
Automatic hate speech detection includes research related to sexism, racism, cyberbullying,
and toxicity in the public realm. The literature focused on these topics is extensive and we
refer the reader to multiple surveys for comprehensive overviews [33, 34, 30, 10]. The rise of
large language models (LLMs) in hate speech detection research has been a significant devel-
opment [18], but, as discussed by Elsafoury [7] and Cooper et al. [3], these models continue to
struggle with nuanced interpretations, which can perpetuate stereotypes and reinforce harm-
ful narratives. Studies such as those by Röttger et al. [32] and Lee et al. [22] emphasize that
hate speech detection models must account for cultural biases to be effective across different
linguistic and social contexts. The problem is further compounded by the scarcity of related
research in Chinese. There are still relatively few Chinese hate speech datasets available, al-
though the situation seems to be improving [37, 4]. Finally, whereas automatic hate speech
detection has been at the forefront of NLP research during the last decade, violent discourse as
a theoretical category has been relatively underrepresented in computational literary studies,
many projects focusing on extra-literary content such as social media posts or movie dialogues
[26, 2, 19, 17].
2.2. Sentiment Analysis in Political Contexts
Similar to hate speech detection, automatic sentiment analysis has seen significant contribu-
tions and surveys during the last two decades. Notable studies include those by Liu [24],
Wankhade et al. [35], and Zhang et al. [38], which provide comprehensive overviews of the
1013
methods and applications of sentiment analysis in various domains. In the context of literary
texts, Jockers’ work with the syuzhet package [15] exemplifies the application of sentiment
analysis to understand emotional arcs in literature.
Related to this article are the numerous studies focused on political contexts, revealing the
nuanced ways public opinion is shaped and expressed and highlighting the role of social media
in political discourse [5, 23, 31]. Advanced techniques such as emotion mining and aspect-
based sentiment analysis (ABSA) have been employed to capture the sentiment in political
texts [9, 14]. These approaches facilitate the extraction of sentiment from complex political
narratives, providing insights into voter behavior and sentiment polarization.
3. Methodology
3.1. Data Collection
This project used two training datasets:
⇒ Violent/Non-Violent Texts: The first dataset was constructed from texts sourced from
the PLA Daily.1 Texts were classified as “violent” if they included language depicting
physical and military violence, as described in our related work on the distribution of
violent discourse in the journal [8], which adopts a dictionary-based approach to detect
violent texts for model training. “Non-violent” texts were characterized by the absence
of such vocabulary. The dataset includes a total of 5,728 examples, split evenly between
the two classes, with no more than 100 examples taken from each year for either class.
⇒ Strong/Weak Emotion: The second dataset was derived from the Douban Dushu
Dataset [39], containing more than 3.7 million Chinese book reviews. As there is no
large dataset containing labeled emotional intensity specific to military-related Chinese
texts from the mid-20th century, which otherwise would be an ideal training corpus for
this project, we searched for a dataset that would capture a broad range of emotional
expressions independent of specific subject matter. The Douban Dushu Dataset meets
this requirement, including reviews of a wide variety of books and thus preventing the
trained model from focusing on any particular topic. We labeled 1-star and 5-star com-
ments as representing “strong” emotions due to their clear expression of either negative
or positive sentiment, while 3-star comments were considered “weak” emotions. Only
the comments that were at least 200 characters long were included, and we selected
62,000 examples from each class (“strong” and “weak” emotion) for training.
3.2. Training & Validation
For this study, we used PyTorch to fine-tune two open-source models bert-base-chinese on the
datasets described above: one for classifying violent versus non-violent texts, and the other
for categorizing the strength of emotions within texts. Notice that the sentiment analysis task
considered in this project differs from the usual NLP applications which distinguish positive
1
The PLA Daily corpus has been acquired from the digitized version of the journal available through the East View
library (https://dlib.eastview.com).
1014
from negative sentiments or categorize them into different classes (anger, surprise, happiness,
etc). Here, we focus on the intensity of the expressed sentiments rather than their quality.
Bert-base-chinese is a lightweight model (with 102 million parameters) pre-trained on a large
corpus of Chinese text, which makes it suitable for various natural language processing tasks. It
requires relatively modest computational resources and enables fast training. For both tasks—
violence detection and emotional intensity assessment—we fine-tuned bert-base-chinese using
sequence classification with the respective dataset, a batch size of 16, a learning rate of 2e-5,
and the Adam optimizer. We used validation loss to find the optimal number of training epochs.
Texts were tokenized into input sequences using the bert-base-chinese tokenizer, which splits
by Chinese character (there are no spaces in Chinese). We have achieved F1 score of 0.981 for
violence analysis (600 test samples) and 0.926 for sentiment analysis (12,000 test samples).
3.3. Quantitative Analysis
After training, the 185,472 articles from the PLA Daily published between 1956 and 1989 were
segmented into non-overlapping chunks of 500 characters, yielding 629,734 texts in total. The
period in question begins with the establishment of the journal in 1956 and ends in 1989, a year
marked by nation-wide pro-democratic protests. Each text was evaluated by the two models for
the probability of being classified as “violent” or “strong” (emotionally intense), respectively.
We then computed the average monthly probability of the “violent” class and the “strong” class.
4. Results
Figure 1: Yearly average and monthly probabilities of violence and emotional intensity in PLA Daily
texts from 1956 to 1989.
As illustrated in Figure 1, there is a clear alignment between violent discourse and emotional
expression in the journal. Affectively-charged texts are often about violence, and violence is
described in affective terms. Both lines demonstrate an increasing trend from the late 1950s,
1015
peaking around 1968, followed by a decline through the late 1970s and 1980s. This trend in-
dicates heightened periods of violence-related, emotionally-charged content published in the
journal during the Cultural Revolution (1966-1976) and a subsequent decrease as China moved
towards more stabilized periods in the post-Mao era. A very strong Pearson correlation be-
tween the monthly averages (408 months; r: 0.8468, p-value: 2.4296e-113) and yearly averages
(34 years; r: 0.9092, p-value: 1.0183e-13) of violent discourse and emotional intensity can be
observed, demonstrating how the People’s Liberation Army Daily “emotionalized revolution
and revolutionized emotions” in the early PRC.2 This relationship can be further illustrated
by plotting the percentage of articles displaying both high-violence and high-emotion scores
throughout the analyzed period (Figure 2). Between 1966 and 1968, the number of such articles
rises to nearly 50% of the total published content.
Figure 2: Percentage of articles with both violence and emotion scores exceeding 0.9, published in the
PLA Daily between 1956 and 1989. Prior to plotting, the values in each category have been normalized
to map onto a [0, 1] scale based on the actual observed ranges.
Examples from the extrema of the distribution have been provided in Table 1 in the Appendix.
In high-violence, high-emotion texts, the sentiments are channelled towards the Communist
Party and the leader Mao Zedong in the yiku-sitian 忆苦思甜 (“remember the bitter past and
think of the sweet present”) mode. In the high-violence, low-emotion texts, the focus is placed
on military matters analyzed from a professional perspective. The low-violence, high-emotion
passages convey gratitude to the Communist Party and its members, with little to no mention
of military history. The low-violence, low-emotion texts focus on civilian matters.
2
It is important to notice that the relative values (trends) within models are more informative than absolute compar-
isons between the models, as they have been trained on different amounts and types of data. For example, emotional
intensity of 0.7 and violence score of 0.5 does not entail that a given text is “more emotional than violent.”
1016
5. Discussion
The above findings offer additional evidence that emotional mobilization was one of the cru-
cial aspects of revolutionary violence in modern China, fostering a collective identity among
the populace confronted with state-designated enemies [29, 28]. Although well-documented in
non-DH sinology, the discovered alignment between emotion and violence is surprising inso-
far as the sentiment-analysis model has been trained on texts (book reviews) that have little in
common with the military-related content published in the PLA Daily. The results demonstrate
the applicability of out-of-distribution datasets in quantitative explorations of literary phenom-
ena, including even such intangible features as emotional valence of political texts. Moreover,
the focus on continuous intensity rather than discrete flavors of emotions mitigates some of
the shortcomings of computational sentiment analysis. By identifying highly-emotional mo-
ments in texts rather than labeling them as either “positive” or “negative,” we give some of the
interpretive power back not only to the researcher but also the individuals who actually read
those texts.
This last point is particularly important given that the intended reader was supposed to not
only sympathize with the suffering of proletarian heroes (Patrick Hogan’s “complementary
emotions” [11]) but also empathize with them by partaking in the revolutionary fervor (“paral-
lel emotions”). These reactions could thus simultaneously feature sentiments at the extrema of
the positive-negative spectrum. Consider the following excerpt from the article “A Communist
Party Member Must Fight,” published on October 12th, 1969:
Fire means a command, and the scene of the fire is a battlefield! Lu Bingyi and his
comrades from the propaganda team were the first to arrive at the site. A raging
fire was engulfing a local alleyway’s plastic processing factory. Through the thick,
acrid smoke, they could hear the desperate cries of women trapped inside. [...] The
fire, fanned by plastic products, raged ever higher. Thick black smoke, carrying a
pungent odor, stung Lu’s nose, causing it to bleed. With the combined heat of the
flames and the suffocating smoke, Lu felt dizzy and gasped for breath. Over and
over, he silently recited,“Be resolute, fear no sacrifice, overcome all difÏculties to
win victory.”Chairman Mao’s teachings, heavy with meaning, strengthened Lu as
he charged into the flames and fought bravely. Foam from the fire extinguishers
sprayed into Lu’s left eye, causing sharp pain, yet he persisted, helping Master
Zhou rescue five class sisters in quick succession.
火光就是命令，火场就是战场！小陆和宣传队的同志首先赶到现场。烈火，在里弄塑料加工
厂里熊熊地燃烧。从浓浓的臭烟里发出了姐妹们焦急的呼救声... 熊熊的烈火，卷着塑料制品，
越烧越旺，浓浓的乌烟，发出一股特殊的臭味，呛得小陆的鼻子直流血。又是火烤，又是烟
熏，小陆感到头昏脑胀，窒息得喘不过气来。他一遍又一遍地背诵着“下定决心，不怕牺牲，
排除万难，去争取胜利”
。毛主席的教导，字字重千斤，鼓励着小陆出入火海，英勇战斗。外
边射进火海的泡沫酸碱喷进了小陆的左眼，痛得厉害，他仍坚持和周师傅一起接连救出了五
个阶级姐妹。
In this and similar passages, the vicarious details are meant to invoke both positive and nega-
tive responses in the reader, embedding ideological instruction at an affective level. A binary
understanding of emotions risks oversimplifying such emotional dynamics and missing the
nuanced ways in which political power can be intertwined with affect.
1017
Our paper thus suggests a special role that can be played by narrative arts. If literature has
the potential to “personalize revolution and revolutionize romantic adventures,” as Liu puts
it [25], it can also disentangle private passions from violent discourse and redirect feelings
towards other facets of life [12, 13]. Depictions of simple everyday interactions, deliberately
paired with non-violent sentiment, may generate affective-discursive spaces that resist political
manipulation. The computational approach proves useful not only in conceptualizing such
spaces in quantitative terms but also identifying them within large textual corpora. We will
further explore this line of thought in the sequels to this paper.
Limitations
Several limitations of this project must be acknowledged. Our primary dataset consists of
articles from the PLA Daily, a single source that does not represent the full spectrum of revolu-
tionary discourse in the PRC. Furthermore, the binary classification of texts as either violent or
non-violent and as conveying strong or weak emotions simplifies the complex nature of human
language. More refined classification systems could be developed to capture such subtleties.
References
[1] R. G. Apressyan. “Violent speech”. In: Peace Review 10.4 (1998), pp. 587–591. doi: 10.108
0/10402659808426209.
[2] C. M. Castorena, I. M. Abundez, R. Alejo, E. E. Granda-Gutiérrez, E. Rendón, and O. Ville-
gas. “Deep Neural Network for Gender-Based Violence Detection on Twitter Messages”.
In: Mathematics 9.8 (2021). doi: 10.3390/math9080807.
[3] P. Cooper, M. Surdeanu, and E. Blanco. “Hiding in Plain Sight: Tweets with Hate Speech
Masked by Homoglyphs”. In: Findings of the Association for Computational Linguistics:
EMNLP 2023. Ed. by H. Bouamor, J. Pino, and K. Bali. Singapore: Association for Compu-
tational Linguistics, 2023, pp. 2922–2929. doi: 10.18653/v1/2023.findings-emnlp.192.
[4] J. Deng, J. Zhou, H. Sun, C. Zheng, F. Mi, H. Meng, and M. Huang. “COLD: A Benchmark
for Chinese Offensive Language Detection”. In: Proceedings of the 2022 Conference on
Empirical Methods in Natural Language Processing. Ed. by Y. Goldberg, Z. Kozareva, and
Y. Zhang. Abu Dhabi, United Arab Emirates: Association for Computational Linguistics,
2022, pp. 11580–11599. doi: 10.18653/v1/2022.emnlp-main.796.
[5] S. Dorle and N. Pise. “Political Sentiment Analysis through Social Media”. In: 2018 Sec-
ond International Conference on Computing Methodologies and Communication (ICCMC).
2018, pp. 869–873. doi: 10.1109/iccmc.2018.8487879.
[6] R. Duan. “Solidarity in Three Acts: Narrating U.S. Black Freedom Movements in China,
1961–1966”. In: Modern Asian Studies 53.5 (2019), pp. 1351–1380. doi: 10.1017/s0026749x
17000489.
1018
[7] F. Elsafoury. “Thesis Distillation: Investigating The Impact of Bias in NLP Models on
Hate Speech Detection”. In: Proceedings of the Big Picture Workshop. Ed. by Y. Elazar,
A. Ettinger, N. Kassner, S. Ruder, and N. A. Smith. Singapore: Association for Computa-
tional Linguistics, 2023, pp. 53–65. doi: 10.18653/v1/2023.bigpicture-1.5.
[8] A. Gilkison and M. Kurzynski. “Vectors of Violence: Legitimation and Distribution of
State Power in the People’s Liberation Army Daily (Jiefangjun Bao), 1956-1989”. In: Jour-
nal of Cultural Analytics 9.1 (2024), pp. 1–25. doi: 10.22148/001c.115481.
[9] Z. Gou, Q. Guo, and Y. Yang. MvP: Multi-view Prompting Improves Aspect Sentiment Tuple
Prediction. 2023. arXiv: 2305.12627 [cs.CL].
[10] M. S. Hee, S. Sharma, R. Cao, P. Nandi, T. Chakraborty, and R. K.-W. Lee. Recent Advances
in Hate Speech Moderation: Multimodality and the Role of Large Models. 2024. arXiv: 240
1.16727 [cs.CL].
[11] P. C. Hogan. “Simulation and the Structure of Emotional Memory: Learning from Arthur
Miller’s After the Fall”. In: Cognitive Literary Science: Dialogues between Literature and
Cognition. Ed. by M. Burke and E. T. Troscianko. New York: Oxford University Press,
2017, pp. 113–134. doi: 10.1093/acprof:oso/9780190496869.003.0007.
[12] P. C. Hogan. Understanding Nationalism: On Narrative, Cognitive Science, and Identity.
Ohio State University Press, 2009.
[13] P. C. Hogan. What Literature Teaches Us About Emotion. Cambridge University Press,
2011.
[14] Y. C. Hua, P. Denny, K. Taskova, and J. Wicker. A Systematic Review of Aspect-based Sen-
timent Analysis (ABSA): Domains, Methods, and Trends. 2024. arXiv: 2311.10777 [cs.CL].
[15] M. Jockers. Syuzhet: An R Package for the Extraction of Sentiment and Sentiment-Derived
Plot Arcs from Text. 2015. url: https://github.com/mjockers/syuzhet.
[16] R. Kelley and B. Esch. “Black Like Mao: Red China and Black Revolution”. In: Souls: A
Critical Journal of Black Politics, Society, and Culture 1.1 (1999), pp. 6–39. doi: 10.1080/1
0999949909362150.
[17] M. Kim, H. Yeen, and M.-W. Koo. “Towards Context-Based Violence Detection: A Korean
Crime Dialogue Dataset”. In: Findings of the Association for Computational Linguistics:
EACL 2024. Ed. by Y. Graham and M. Purver. St. Julian’s, Malta: Association for Compu-
tational Linguistics, 2024, pp. 603–623.
[18] T. Kumarage, A. Bhattacharjee, and J. Garland. Harnessing Artificial Intelligence to Com-
bat Online Hate: Exploring the Challenges and Opportunities of Large Language Models in
Hate Speech Detection. 2024. arXiv: 2403.08035 [cs.CL].
[19] V. Kumari, K. Memon, B. Aslam, and B. S. Chowdhry. “An Effective Approach for Vi-
olence Detection using Deep Learning and Natural Language Processing”. In: 2023 7th
International Multi-Topic ICT Conference (IMTIC). 2023, pp. 1–8. doi: 10.1109/imtic58887
.2023.10178618.
[20] E. Lean. Public Passions: The Trial of Shi Jianqiao and the Rise of Popular Sympathy in
Republican China. University of California Press, 2007.
1019
[21] H. Lee. Revolution of the Heart: A Genealogy of Love in China, 1900-1950. Stanford Uni-
versity Press, 2007.
[22] N. Lee, C. Jung, and A. Oh. “Hate Speech Classifiers are Culturally Insensitive”. In: Pro-
ceedings of the First Workshop on Cross-Cultural Considerations in NLP (C3NLP). Ed. by
S. Dev, V. Prabhakaran, D. Adelani, D. Hovy, and L. Benotti. Dubrovnik, Croatia: Asso-
ciation for Computational Linguistics, 2023, pp. 35–46. doi: 10.18653/v1/2023.c3nlp-1.5.
[23] B. Li, H. Fei, F. Li, Y. Wu, J. Zhang, S. Wu, J. Li, Y. Liu, L. Liao, T.-S. Chua, and D. Ji. DiaASQ
: A Benchmark of Conversational Aspect-based Sentiment Quadruple Analysis. 2023. arXiv:
2211.05705 [cs.CL].
[24] B. Liu. Sentiment Analysis and Opinion Mining. Morgan & Claypool Publishers, 2012.
[25] J. Liu. Revolution Plus Love: Literary History, Women’s Bodies, and Thematic Repetition in
Twentieth-Century Chinese Fiction. University of Hawaii Press, 2003.
[26] E. Mensa, D. Colla, M. Dalmasso, M. Giustini, C. Mamo, A. Pitidis, and D. P. Radicioni.
“Violence detection explanation via semantic roles embeddings”. In: BMC Medical Infor-
matics and Decision Making 20.1 (2020), p. 263. doi: 10.1186/s12911-020-01237-4.
[27] P. E. O’Connor. “Introduction: Discourse of Violence”. In: Discourse and Society 6.3 (1995),
pp. 309–318.
[28] L. Pang. The Art of Cloning: Creative Production During China’s Cultural Revolution. Verso
Books, 2017.
[29] E. Perry. “Moving The Masses: Emotion Work in The Chinese Revolution”. In: Mobiliza-
tion: An International Quarterly 7.2 (2002), pp. 111–128. doi: 10.17813/maiq.7.2.70rg70l2
02524uw6.
[30] F. Poletto, V. Basile, M. Sanguinetti, C. Bosco, and V. Patti. “Resources and benchmark
corpora for hate speech detection: a systematic review”. In: Lang. Resour. Eval. 55.2
(2021), pp. 477–523. doi: 10.1007/s10579-020-09502-8.
[31] P. Rita, N. António, and A. P. Afonso. “Social media discourse and voting decisions influ-
ence: sentiment analysis in tweets during an electoral period”. In: Social Network Analysis
and Mining 13.1 (2023), p. 46. doi: 10.1007/s13278-023-01048-1.
[32] P. Röttger, H. Seelawi, D. Nozza, Z. Talat, and B. Vidgen. Multilingual HateCheck: Func-
tional Tests for Multilingual Hate Speech Detection Models. 2022. arXiv: 2206 . 09917
[cs.CL].
[33] A. Schmidt and M. Wiegand. “A Survey on Hate Speech Detection using Natural Lan-
guage Processing”. In: Proceedings of the Fifth International Workshop on Natural Lan-
guage Processing for Social Media. Ed. by L.-W. Ku and C.-T. Li. Valencia, Spain: Associa-
tion for Computational Linguistics, 2017, pp. 1–10. doi: 10.18653/v1/W17-1101.
[34] F. D. Vigna, A. Cimino, F. Dell’Orletta, M. Petrocchi, and M. Tesconi. “Hate Me, Hate Me
Not: Hate Speech Detection on Facebook”. In: Italian Conference on Cybersecurity. 2017,
pp. 86–95.
1020
[35] M. Wankhade, A. C. S. Rao, and C. Kulkarni. “A survey on sentiment analysis methods,
applications, and challenges”. In: Artif. Intell. Rev. 55.7 (2022), pp. 5731–5780. doi: 10.10
07/s10462-022-10144-1.
[36] A. Weber. Manual on Hate Speech. Council of Europe Publishing, 2009.
[37] Y. Xiao, H. Bouamor, and W. Zaghouani. Chinese Offensive Language Detection:Current
Status and Future Directions. 2024. arXiv: 2403.18314 [cs.CL].
[38] W. Zhang, Y. Deng, B. Liu, S. J. Pan, and L. Bing. Sentiment Analysis in the Era of Large
Language Models: A Reality Check. 2023. arXiv: 2305.15005 [cs.CL].
[39] J. Zhao and Z. Ji. LSICC: A Large Scale Informal Chinese Corpus. 2018. arXiv: 2403.18314
[cs.CL].
1021
Appendix
Excerpt Source V E
The heavens and earth are not as great as the Party’s kindness; Chairman
Mao is truly the most dear person to us poor and lower-middle peasants. In
the wicked old society, eleven of my relatives were killed. When I was
fourteen, my father was beaten to death by a heartless landlord while
working for him. After my father’s death, my mother led us siblings to beg for
food. [...] A few years after joining my husband’s family, four members of his “Don’t Forget Class Bitterness:
family died from hunger and exhaustion. I gave birth to eight sons, but two of Always Loyal To Chairman Mao”
0.9955 0.9941
them starved to death. I begged for food from Shandong to Guandong, from a 不忘阶级苦: 永远忠于毛主席 Jun
young child until I was over fifty years old. A landlord’s vicious dog bit off my 17, 1968
right ear, leaving me covered in wounds and nearly dead.
我想，天大地大不如党的恩情大，毛主席真是咱贫下中农最亲的人。万恶的旧社会，活活坑死了我
十一个亲人。我十四岁那年，爹爹在给地主扛活时，被狼心狗肺的地主活活打死了。爹爹死后，妈
妈领着我们姊妹几个到处要饭... 过门后不几年，婆家又饿死和累死了四口人。我生了八个儿子，
也活活饿死两个。我要饭从山东要到关东，从不懂事的孩子要到五十多岁。地主的恶狗咬掉了我
的右耳朵，咬得我满身是伤，险些送了命。
In recent years, the Soviet military has paid great attention to the
synchronization of air defense weapons and combat units, studying issues
such as the deployment, firing, and logistical support of air defense units
during movement to ensure the success of their large-scale mobile operations.
However, Western analysts believe that the mobility of the Soviet field air
defense is far from meeting the requirements of rapid army offensives. When “Three Characteristics of Soviet
Army Field Air Defense” 苏陆军野 0.9995 0.2271
the troops begin to move, the effectiveness of the air defense drops sharply. 战防空三个特点 Aug 24, 1984
Combined with a low level of electronic warfare capabilities, significant
technical and tactical improvements are still needed.
近年来，苏军很注意防空武器与作战部队的同步运动，研究运动中防空部队的展开、发射和补充
保障等问题，以保证其大规模机动作战的胜利。但西方认为，苏军野战防空的机动性远没有达到
陆军快速进攻的要求。部队一运动起来，防空的效能骤减，加上电子战水平低，在技术、战术上还
需大大改进。
I was so emotional that I couldn’t speak. I lost my mother when I was very
young, and my father was constantly running around to make ends meet,
leaving no one to take care of me as I was tormented by illness. But today, in
the revolutionary forces, my superiors care for and look after me with such
meticulous attention, like my own parents. As I thought about this, tears
welled up in my eyes. With trembling hands, I accepted the fruits and snacks
brought by Section Chief Yang. These were not just fruits and snacks, but a “When Section Chief Yang and I
symbol of the heartfelt care from revolutionary comrades to their fellow Were Together” 杨股长和我在一 0.0079 0.9948
soldiers, continually warming my heart. Dear Section Chief Yang, you worked 起的时候 Jun 6, 1959
tirelessly for the revolutionary cause, exhausting yourself to the point of
illness, and now you have left us forever!
我激动得说不出话来。我在很小的时候失去了母亲，父亲整天为我们的生活奔跑，疾病把我折磨
得死去活来，也无人照顾。而今天在革命部队里，上级对自己是这样无微不至地体贴和照顾，象亲
生父母一样。我想着想着，热泪禁不住夺眶而出。我用颤抖的手接过股长送来的水果和点心。这不
是水果，也不是点心，而是革命同志对战友的一颗火热的心，它不断地温暖着我。亲爱的杨股长，
为了革命事业废寝忘食，劳累成疾，终于和我们永别了！
Systematically and in an organized manner, military officers are being
dispatched to civilian enterprises and local universities to gain life experience
and learn from the strengths of these institutions. This initiative aims to
enhance and improve the educational work of military academies. The main
considerations for this new attempt by the Air Self-Defense Force Officers
“Japan Air Self-Defense Force’s
School are: first, to broaden the horizons of young military officers, helping New Attempt to Train Command
0.0007 0.2716
them understand society, learn from the strengths of civilian enterprises and Officers” 日本航空自卫队培养指
local institutions, and address their own shortcomings; second, to deepen the 挥干部的新尝试 Oct 10, 1988
understanding of the military within local and civilian communities...
有计划、有组织地把军队干部派往民间企业、地方大学体验生活，吸取民间企业、地方大学的长
处，以加强和改善军队院校的育人工作。航空自卫队干部学校进行新尝试的主要考虑是：一、使年
轻军队干部开阔视野，了解社会，取地方、民间之长，补自己之短；二、加深地方、民间对军队的
了解...
Table 1
Sample texts with corresponding violence (V) and emotion (E) scores.
1022