• Search Menu
  • Advance articles
  • Browse content in Biological, Health, and Medical Sciences
  • Administration Of Health Services, Education, and Research
  • Agricultural Sciences
  • Allied Health Professions
  • Anesthesiology
  • Anthropology
  • Anthropology (Biological, Health, and Medical Sciences)
  • Applied Biological Sciences
  • Biochemistry
  • Biophysics and Computational Biology (Biological, Health, and Medical Sciences)
  • Biostatistics
  • Cell Biology
  • Dermatology
  • Developmental Biology
  • Environmental Sciences (Biological, Health, and Medical Sciences)
  • Immunology and Inflammation
  • Internal Medicine
  • Medical Microbiology
  • Medical Sciences
  • Microbiology
  • Neuroscience
  • Obstetrics and Gynecology
  • Ophthalmology
  • Pharmacology
  • Physical Medicine
  • Plant Biology
  • Population Biology
  • Psychological and Cognitive Sciences (Biological, Health, and Medical Sciences)
  • Public Health and Epidemiology
  • Radiation Oncology
  • Rehabilitation
  • Sustainability Science (Biological, Health, and Medical Sciences)
  • Systems Biology
  • Browse content in Physical Sciences and Engineering
  • Aerospace Engineering
  • Applied Mathematics
  • Applied Physical Sciences
  • Bioengineering
  • Biophysics and Computational Biology (Physical Sciences and Engineering)
  • Chemical Engineering
  • Civil and Environmental Engineering
  • Computer Sciences
  • Computer Science and Engineering
  • Earth Resources Engineering
  • Earth, Atmospheric, and Planetary Sciences
  • Electric Power and Energy Systems Engineering
  • Electronics, Communications and Information Systems Engineering
  • Engineering
  • Environmental Sciences (Physical Sciences and Engineering)
  • Materials Engineering
  • Mathematics
  • Mechanical Engineering
  • Sustainability Science (Physical Sciences and Engineering)
  • Browse content in Social and Political Sciences
  • Anthropology (Social and Political Sciences)
  • Economic Sciences
  • Environmental Sciences (Social and Political Sciences)
  • Political Sciences
  • Psychological and Cognitive Sciences (Social and Political Sciences)
  • Social Sciences
  • Sustainability Science (Social and Political Sciences)
  • Author guidelines
  • Submission site
  • Open access policy
  • Self-archiving policy
  • Why submit to PNAS Nexus
  • The PNAS portfolio
  • For reviewers
  • About PNAS Nexus
  • About National Academy of Sciences
  • Editorial Board
  • Journals on Oxford Academic
  • Books on Oxford Academic

Issue Cover

Article Contents

Introduction, experimental design, acknowledgments, supplementary material, author contributions, data availability.

  • < Previous

How persuasive is AI-generated propaganda?

ORCID logo

Competing Interest: The authors declare no competing interest.

  • Article contents
  • Figures & tables
  • Supplementary Data

Josh A Goldstein, Jason Chao, Shelby Grossman, Alex Stamos, Michael Tomz, How persuasive is AI-generated propaganda?, PNAS Nexus , Volume 3, Issue 2, February 2024, pgae034, https://doi.org/10.1093/pnasnexus/pgae034

  • Permissions Icon Permissions

Can large language models, a form of artificial intelligence (AI), generate persuasive propaganda? We conducted a preregistered survey experiment of US respondents to investigate the persuasiveness of news articles written by foreign propagandists compared to content generated by GPT-3 davinci (a large language model). We found that GPT-3 can create highly persuasive text as measured by participants’ agreement with propaganda theses. We further investigated whether a person fluent in English could improve propaganda persuasiveness. Editing the prompt fed to GPT-3 and/or curating GPT-3’s output made GPT-3 even more persuasive, and, under certain conditions, as persuasive as the original propaganda. Our findings suggest that propagandists could use AI to create convincing content with limited effort.

Online covert propaganda campaigns are frequent and ongoing. Recently, policymakers, technologists, and researchers have voiced concern that new artificial intelligence (AI) tools could supercharge covert propaganda campaigns by allowing propagandists to mass produce text at low cost. Could foreign actors use AI to generate persuasive propaganda targeting audiences in the United States? To investigate this, we conducted a preregistered survey experiment on 8,221 US respondents comparing the persuasiveness of English-language foreign covert propaganda articles sourced from real-world campaigns to text generated by a large language model, which is a form of AI. We found that the large language model can create highly persuasive text, and that a person fluent in English could improve the persuasiveness of AI-generated propaganda with minimal effort.

Academics, journalists, online platforms, and governments have demonstrated that online covert propaganda campaigns are frequent and ongoing ( 1 , 2 ). Disclosures of Russian disinformation campaigns on social media targeting the United States in 2016 heightened awareness of these efforts ( 3 ) and caused platforms to commit more resources to finding and suspending these operations ( 4 ). Yet, covert propaganda operations continue on websites ( 5 ), social media platforms ( 6 ), encrypted messaging apps ( 7 ), and other channels. State-backed covert propaganda campaigns use short-form content and full-length articles for a range of goals, from self-promotion to undermining confidence in democratic institutions.

Recently, many have voiced concern that new artificial intelligence (AI) tools could supercharge covert propaganda campaigns by allowing propagandists to mass produce text at low cost ( 8–10 ). The machine learning community has made major breakthroughs in language models that can generate original text in response to a text input ( 11 ). These models are quickly diffusing across society.

Despite broad concern about the use of language models for propaganda and other information campaigns, only a limited number of studies have used social science methods to assess the risk. Scholars have examined whether people rate AI-generated news articles as credible ( 12 ) and recognize when AI-generated content is false ( 13 ), and whether elected officials reply to AI-generated constituent letters ( 14 ). However, to our knowledge, no studies examine the persuasiveness of AI-generated propaganda compared to an ecologically valid benchmark.

We ran an experiment with US respondents comparing the persuasiveness of foreign covert propaganda articles sourced from real-world campaigns to text created by GPT-3 davinci, a large language model developed by OpenAI. We focused on propaganda articles, rather than snippets such as tweets, since the performance of language models typically declines as text length increases. We therefore create a relatively “hard case” for the technology. Our preregistration plan is available with the Open Science Framework.

Article selection and construction

We began by identifying six articles (ranging from 151 to 308 words long) that investigative journalists or researchers uncovered as part of covert, likely state-aligned propaganda campaigns originating from either Iran or Russia (see SI Appendix 1.A for details on article selection). We then used GPT-3 to generate articles on the same six topics. For each topic, we fed GPT-3 one or two sentences from the original propaganda article that make the article’s main point, as well as three other propaganda articles on unrelated topics. The three example articles informed the style and structure of the GPT-3-generated text, while the excerpts from the original article informed the topic. We asked GPT-3 to generate three articles on each topic, rather than one, to avoid over-indexing on any one output since each AI-generated output is unique. We discarded generations that were <686 characters or >1,936 characters. These parameters were selected to keep articles within 10% of the shortest and longest articles from the original or edited propaganda set. No other criteria were used to discard GPT-3 output. (We include full information on the article generation process in SI Appendix 1.B .)

After finding the original propaganda articles and using GPT-3 to create AI-generated versions, we compared the persuasiveness of the two. To measure persuasiveness, we first summarized in direct, plain English the main point of the original propaganda. The thesis statements, shown in Table 1 , are cleaned versions of the passages we fed to GPT-3 for each topic.

Researcher-written thesis statements for the six articles.

These sentences summarize the main point we believed the propagandist was trying to convince the target audience. In some cases, this was challenging since articles made multiple points. Several of these statements are either false or debatable.

Survey deployment

In December 2021, we interviewed US adults using Lucid, a survey company that uses quota sampling to achieve geographic and demographic representativeness. Per our preregistration, respondents who failed attention checks at the beginning of the survey were not invited to continue, and respondents who completed the survey in <3 min were excluded, resulting in a final sample of 8,221.

We asked each respondent how much they agreed or disagreed with the thesis statements for four of the six propaganda topics, selected at random, without reading an article about these topics. This serves as our control data. We then presented each respondent with articles on the remaining two topics and measured agreement with the thesis statements for those topics. Some of the articles we presented were original propaganda; others were propaganda generated by GPT-3. (We presented one article about a Syria-related topic, and one article about a non-Syria-related topic. The articles appear in the second SI Appendix , and details about our experimental deployment appear in SI Appendix 1.C .)

We then estimated how our treatments affected two measures of agreement: percent agreement, defined as the percentage of respondents who agreed or strongly agreed with each thesis statement, and scaled agreement, defined as the average score on a 5-point scale from 0 (“strongly disagree”) to 100 (“strongly agree”). Specifically, we regressed each measure of agreement on a comprehensive set of indicators for each issue and article, and used the regression coefficients to compute quantities of interest. When averaging across issues, for example, we gave equal weight to each issue, and when averaging across articles produced by GPT-3, we gave equal weight to each article. For a complete presentation of the regression models and results, see SI Appendix 2 . Below, we focus our discussion on percent agreement, but overall patterns and conclusions were similar when we analyzed scaled agreement. a

This study was approved by Stanford University’s Institutional Review Board which focused on risks to survey respondents, and also vetted by a cross-professional AI-specific Ethics Review Board that considered risks to society. All participants provided informed consent. To mitigate risks that respondents might come to believe falsehoods, we informed respondents after they completed the survey that the articles came from propaganda sources and may have contained false information. Regarding risks to society, propagandists are likely already well aware of the capabilities of large language models; historically, propagandists have been quick both to adopt new technologies and incorporate local language speakers into their work. As a result, the societal benefit of assessing the potential risks outweighs the possibility that our paper would give propagandists new ideas.

Persuasiveness of GPT-3-generated propaganda

To establish a benchmark against which we can evaluate GPT-3, we first assess the effect of reading the original propaganda compared to not reading any propaganda about that topic (the control). We start by presenting estimates pooled across topics and outputs, and later break out topics and outputs individually. As shown in Fig. 1 , the original propaganda was highly persuasive. While only 24.4% of respondents who were not shown an article agreed or strongly agreed with the thesis statement, the rate of agreement jumped to 47.4% (a 23 percentage point increase) among respondents who read the original propaganda. Thus, the original propaganda nearly doubled the share of participants who concurred with the thesis statement.

Original propaganda and GPT-3-generated propaganda were highly persuasive. The top panel shows the percentage of respondents who agreed or strongly agreed with the thesis statement. The bottom panel shows the average level of agreement on a 5-point scale, coded 0 if strongly disagree, 25 if disagree, 50 if neither agree nor disagree, 75 if agree, and 100 if strongly agree. Estimates are pooled across topics and outputs. SEs clustered by respondent and 95% CIs are shown.

Original propaganda and GPT-3-generated propaganda were highly persuasive. The top panel shows the percentage of respondents who agreed or strongly agreed with the thesis statement. The bottom panel shows the average level of agreement on a 5-point scale, coded 0 if strongly disagree, 25 if disagree, 50 if neither agree nor disagree, 75 if agree, and 100 if strongly agree. Estimates are pooled across topics and outputs. SEs clustered by respondent and 95% CIs are shown.

GPT-3-generated propaganda was also highly persuasive, and 43.5% of respondents who read a GPT-3-generated article agreed or strongly agreed with the thesis statement, compared to 24.4% in the control (a 19.1 percentage point increase). This suggests that propagandists could use GPT-3 to generate persuasive articles with minimal human effort, by using existing articles on unrelated topics to guide GPT-3 about the style and length of new articles. b While GPT-3-generated propaganda was highly persuasive, it was slightly less compelling than the original propaganda (a 3.9% point difference). Figures S7 and S8 show that the persuasive effects of the original propaganda and GPT-3 propaganda were fairly consistent across social groups. We did not find substantial heterogeneity in treatment effects when we split the sample by demographic variables, partisanship/ideology, news consumption, time spent on social media, and more. This suggests that AI-generated propaganda could be compelling to a remarkably wide range of groups in society.

In Fig. 2 , we break out the results by article topic and show each of the three GPT-3-generated outputs. While baseline agreement in the control group varied by topic, almost all GPT-3 outputs were highly persuasive. For most issues, each GPT-3-generated article was about as persuasive as the original propaganda. However, this was not always the case. For example, Syria Oil output 3 and Wall outputs 2 and 3 performed significantly worse than the original propaganda on both percent agreement and scaled agreement. c The poor performance of these articles and a few others caused GPT-3 to perform slightly less well than the original propaganda on average, when, in Fig. 1 , we had computed an average that gave equal weight to all GPT-3-generated outputs from all six issues. This suggests a potential role for human propagandists, who could review the output of GPT-3 and select the high-quality articles that make the propagandist’s point.

Most GPT-3-generated output were as persuasive as the original propaganda, but a few articles performed worse. Average agreement with the thesis statement for each treatment group for each topic. SEs clustered by respondent and 95% CIs are shown.

Most GPT-3-generated output were as persuasive as the original propaganda, but a few articles performed worse. Average agreement with the thesis statement for each treatment group for each topic. SEs clustered by respondent and 95% CIs are shown.

Human–machine teaming

In practice, propagandists might not use all of the output from a model in a propaganda campaign. Instead, they could engage in human–machine teaming to increase the efficiency of human propagandists while still having a measure of human oversight and quality control ( 15 ).

After running the model, a human could serve as a curator by weeding out articles that do not make the point the propagandist seeks to get across. To simulate this scenario, a human read through each GPT-3 output carefully to see whether either the title or the body of the article make the claim of the thesis statement. (For a description of this process, see SI Appendix 1.D .) Two of the GPT-3 propaganda articles (out of 18 total) did not advance the intended claim. When we removed those articles, and focused only on outputs that make the thesis, agreement increased to 45.6%, and the difference between original propaganda and curated GPT-3 propaganda ceased to be statistically significant (see Fig. 3 ). Thus, after discarding a small number of articles that did not include the thesis statement, GPT-3 was as persuasive as the original propaganda.

Human curation made GPT-3 as persuasive as the original propaganda. Average agreement with the thesis statement for each treatment group, averaged over topics. “All outputs (no curation)” includes all GPT-3 propaganda articles. “Outputs that make thesis” excludes GPT-3 outputs that did not make the claim of the thesis in the title or body of the article. “Best performing outputs” is the average agreement with the thesis statement for the best performing GPT-3 output for each of the six topics. SEs and 95% CIs are shown.

Human curation made GPT-3 as persuasive as the original propaganda. Average agreement with the thesis statement for each treatment group, averaged over topics. “All outputs (no curation)” includes all GPT-3 propaganda articles. “Outputs that make thesis” excludes GPT-3 outputs that did not make the claim of the thesis in the title or body of the article. “Best performing outputs” is the average agreement with the thesis statement for the best performing GPT-3 output for each of the six topics. SEs and 95% CIs are shown.

Another strategy for human involvement would be to edit the prompt given to GPT-3. The original propaganda included typos and grammatical errors, perhaps indicative of an author whose native language was not English. To simulate what would happen if a fluent English speaker wrote the prompts for GPT-3, we made two changes: (i) we provided GPT-3 with the researcher-written thesis statement, rather than an excerpt from the original article and (ii) we edited the example articles on unrelated topics, with the expectation that better-written examples would lead to better output. As Fig. 3 shows, articles generated by GPT-3 with an edited prompt were as persuasive as the original propaganda: the difference between 46.4 and 47.4% was small and not statistically significant.

Doing both—editing the prompts and curating the output—would be even better. If a propagandist edited the input and selected the best of the three outputs on each topic, the GPT-3-generated propaganda would be even more persuasive than the original propaganda (52.7% compared to 47.4%). In practice, propagandists might perform curation themselves or crowdsource curation for selecting the best articles from a set of outputs.

GPT-3 performance on additional metrics

One potential critique of our study is that our article generation process and experimental design might favor GPT-3 on the persuasiveness measure. As described above, we first determined what we thought the main point of each article was. For the GPT-3 output (without curation), we fed a snippet from the original propaganda article that makes the main point to GPT-3 in the prompt. For the scenario where we edited the example articles fed to GPT-3, we fed the researcher-written thesis statement to GPT-3. If we created GPT-3-generated articles based on an incorrect reading of the main point of the article and used that same incorrect reading for our persuasiveness measure, then our process would favor GPT-3-generated articles compared to the original propaganda. In turn, this might overstate the power of GPT-3 in propaganda campaigns.

To address this concern, we compared GPT-3 with the original propaganda on two additional dimensions: perceived credibility and writing style. We measured credibility by asking respondents whether they thought the article was trustworthy, and whether they thought the article was written to report the facts (vs. to convince the reader of a viewpoint). For a proxy for writing style, we asked respondents whether they thought the article was well written and whether they thought the author’s first language was English. On all these measures, GPT-3 performed as well, if not better, than the original propaganda (see Fig. S13 ). Our findings suggest that GPT-3-generated content could blend into online information environments on par with content we sourced from existing foreign covert propaganda campaigns. While this may not be a very high bar (only 38.7% of respondents found the original propaganda to be trustworthy, and only 52.4% thought the original propaganda was well written), language models are quickly improving. If a similar study were run with more powerful models in the future, AI-generated propaganda would likely perform even better.

Our experiment showed that language models can generate text that is nearly as persuasive for US audiences as content we sourced from real-world foreign covert propaganda campaigns. Moreover, human–machine teaming strategies (editing prompts and curating outputs) produced articles that were as or more persuasive than the original propaganda. Our results go beyond earlier efforts by evaluating the persuasiveness of AI-generated text directly (rather than focusing on metrics like credibility) and using an ecologically valid benchmark.

For two reasons, our estimates may represent a lower bound on the relative persuasive potential of large language models. First, large language models are rapidly improving. Since our study was conducted, several companies have released larger models (e.g. OpenAI’s GPT-4) that outperform GPT-3 davinci in related tasks ( 16 ). We expect that these improved models, and others in the pipeline, would produce propaganda at least as persuasive as the text we administered.

Second, our experiment estimated the effect of reading a single article, but propagandists could use AI to expose citizens to many articles. d With AI, actors—including ones without fluency in the target language—could quickly and cheaply generate many articles that convey a single narrative, while also varying in style and wording. This approach would increase the volume of propaganda, while also making it harder to detect, since articles that vary in style and wording may look more like the views of real people or genuine news sources. Finally, AI can save time and money, enable propagandists to redirect resources from creating content to building infrastructure (e.g. fake accounts, “news” websites that mask state links) that look credible and evade detection.

Our research tested the effects of propaganda about several issues, including drones, Iran, the US–Mexico border wall, and conflict in Syria. Using our experimental design, future research could test the effects of AI-generated propaganda across a wider range of issues, to assess how the effects vary by the salience of the topic and the respondent’s prior knowledge. Research could also address how much respondents are persuaded by AI-generated propaganda when they receive information from multiple sources on a topic.

Another line of research could probe strategies to guard against the potential misuse of language models for propaganda campaigns ( 25 ). If generative AI tools can scale propaganda generation, research that improves the detection of infrastructure needed to deliver content to a target (such as inauthentic social media accounts) will become more important. These detection methods are agnostic as to whether the content is AI-generated or human-written. Research into which systems are susceptible to being overrun by AI-generated text ( 14 ) and how to defend against these attacks could also mitigate the impact of AI-generated propaganda campaigns on democratic processes.

Future research could also focus on behavioral interventions to reduce the likelihood that users believe misleading AI-generated content. There is work on the conditions under which people can assess whether content is AI-generated ( 26 ), and work on how people understand labels that could be applied to misleading or AI-generated content ( 27 ). Research could build on these studies by exploring the effect of labeling AI-generated content on both engagement with the content and whether people believe the content is AI-generated.

We preregistered and implemented additional treatment arms using GPT-3 fine-tuned on articles from the Washington Post’s “Politics” section. (See preregistration plan for additional details.) We did this to emulate a propagandist who wanted to generate text in the structure and style of the Washington Post. Fine-tuning GPT-3 on these articles, however, adversely affected the substance of the AI-generated articles: only 36% of the outputs from the fine-tuned model made the thesis statement, perhaps because the fine-tuning process caused the model to prioritize the content of the Washington Post articles. In Section 3 of the SI Appendix , we include a detailed analysis of the effect of fine-tuning with the Washington Post. In practice, a propagandist fine-tuning a model to create persuasive propaganda would likely read the output to see if it made the intended point. If the process failed in this regard, they would likely adjust the fine-tuning process or abandon it altogether. We believe that the failure of the one fine-tuning approach we tried (see preregistration plan for implementation details) does not speak to the broader utility of fine-tuning. Future research could explore the conditions under which fine-tuning large language models is useful.

We made a minor error in constructing the GPT-3 prompt for the “wall” topic, when we intended to select an excerpt from the original “wall” article that makes the main point of the article. SI Appendix 2.A.2 explains this error, and Figs. S9 and S10 show results if this topic is excluded.

In addition, Syria Medical output 3 performed significantly worse than the original propaganda on scaled agreement, but not percent agreement.

Research beginning with Hasher et al. has shown that people are more likely to believe information when they are exposed to it multiple times (the “illusory truth effect”) ( 17 ). Recent experiments have shown that people were more likely to believe false headlines sourced from Facebook around the 2016 US Presidential Election if exposed multiple times ( 18 ), that repeated exposure to content on a social media platform leads to a higher rate of sharing that content ( 19 , 20 ), and that belief in false information continues to grow logarithmically with additional exposures ( 21 ). Repeated exposure has also been linked to real-world harms: Bursztyn et al. found that areas in the United States with greater exposure to a media outlet downplaying the threat of COVID-19 experienced a greater number of cases and deaths ( 22 ). Similarly, research in psychology has shown that people are more likely to believe misinformation when it comes from multiple sources, rather than one (the “multiple source effect”) ( 23 ) and more likely to adopt a position when it is supported by a greater number of arguments ( 24 ).

We thank Siddharth Karamcheti, Percy Liang, Chris Manning, Sara Plana, Girish Sastry, and participants at the International Studies Association Annual Conference and seminars at Georgetown University, the Naval War College, and Stanford University for feedback. We thank OpenAI for providing access to GPT-3 via their academic access program.

Supplementary material is available at PNAS Nexus online.

This research was funded by Stanford University’s Institute for Human-Centered AI (grant #216301) and Stanford University’s Center for Research on Foundation Models.

J.A.G. designed research, performed research, analyzed data, wrote the paper. J.C. designed research, performed research. S.G. designed research, performed research, analyzed data, wrote the paper. A.S. designed research. M.T. designed research, performed research, analyzed data, wrote the paper.

An earlier version of this manuscript was posted on a preprint server: https://osf.io/preprints/socarxiv/fp87b .

Data and replication code are available on the Harvard Dataverse: https://doi.org/10.7910/DVN/LAZ7AA .

Goldstein JA , Grossman S . 2021 Jan . How disinformation evolved in 2020. Brookings Tech Stream [accessed 2022 Dec 6]. https://www.brookings.edu/techstream/how-disinformation-evolved-in-2020/ .

Martin DA , Shapiro JN , Ilhardt JG . 2022 . Introducing the online political influence efforts dataset . J Peace Res . 60 ( 5 ): 868 – 876 .

Google Scholar

Jamieson KH . 2020 . Cyberwar: how Russian hackers and trolls helped elect a president: what we don’t, can’t, and do know . New York : Oxford University Press .

Google Preview

François C , Douek E . 2021 . The accidental origins, underappreciated limits, and enduring promises of platform transparency reporting about information operations . J Online Trust Saf . 1 ( 1 ): 1 – 30 .

Stubbs J . 2020 . Iranian disinformation effort went small to stay under big tech’s radar. Reuters [accessed 2022 Dec 6]. https://www.reuters.com/article/us-usa-election-facebook-russia/duped-by-russia-freelancers-ensnared-in-disinformation-campaign-by-promise-of-easy-money-idUSKBN25T35E .

Nimmo B , Agranovich D . 2022 . Removing coordinated inauthentic behavior from China and Russia [accessed 2022 Dec 6]. https://about.fb.com/news/2022/09/removing-coordinated-inauthentic-behavior-from-china-and-russia/ .

Frenkel S . 2021 . Iranian disinformation effort went small to stay under big tech’s radar. New York Times [accessed 2022 Dec 6]. https://www.nytimes.com/2021/06/30/technology/disinformation-message-apps.html .

Weidinger L . 2022 . Taxonomy of risks posed by language models. In: Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency (FAccT '22) (Association for Computing Machinery, New York). p. 214–229 .

Helmore E . 2023 Mar . ‘We are a little bit scared’: OpenAI CEO warns of risks of artificial intelligence. The Guardian [accessed 2023 Aug 18]. https://www.theguardian.com/technology/2023/mar/17/openai-sam-altman-artificial-intelligence-warning-gpt4 .

Nature. 2023 Jun. Stop talking about tomorrow’s AI doomsday when AI poses risks today [accessed 2023 Aug 18]. https://www.nature.com/articles/d41586-023-02094-7 .

Bommasani R , et al.  2021 . On the opportunities and risks of foundation models, arXiv, arXiv:2108.07258, preprint: not peer reviewed .

Kreps S , McCain RM , Brundage M . 2020 . All the news that’s fit to fabricate: AI-generated text as a tool of media misinformation . J Exp Polit Sci . 9 : 104 – 117 .

Spitale G , Biller-Andorno N , Germani F . 2023 . AI model GPT-3 (dis)informs us better than humans . Sci Adv . 9 ( 26 ): eadh185 .

Kreps S , Kriner DL . 2023 . The Potential Impact of Emerging Technologies on Democratic Representation: Evidence from a Field Experiment . New Media & Society . 14614448231160526.

Buchanan B , Lohn A , Musser M , Sedova K . 2021 . Truth, lies, and automation: how language models could change disinformation. Center for Security and Emerging Technology [accessed 2022 Dec 6]. https://doi.org/10.51593/2021CA003 .

Liang P , et al.  2023 . Holistic evaluation of language models . Transact Mach Learn Res . 2835 – 8856 .

Hasher L , Goldstein D , Toppino T . 1977 . Frequency and the conference of referential validity . J Verbal Learning Verbal Behav . 16 ( 1 ): 107 – 112 .

Pennycook G , Cannon TD , Rand DG . 2018 . Prior exposure increases perceived accuracy of fake news . J Exp Psychol Gen . 147 ( 12 ): 1865 .

Zhou C , Zhao Q , Lu W . 2015 . Impact of repeated exposures on information spreading in social networks . PLoS One . 10 ( 10 ): e0140556 .

Vellani V , Zheng S , Ercelik D , Sharot T . 2023 . The illusory truth effect leads to the spread of misinformation . Cognition . 236 : 105421 .

Hassan A , Barber SJ . 2021 . The effects of repetition frequency on the illusory truth effect . Cogn Res Princ Implic . 6 ( 1 ): 1 – 12 .

Bursztyn L , Rao A , Roth CP , Yanagizawa-Drott DH . 2020 . Misinformation during a pandemic. Technical Report, National Bureau of Economic Research .

Harkins SG , Petty RE . 1987 . Information utility and the multiple source effect . J Pers Soc Psychol . 52 ( 2 ): 260 .

Calder BJ , Insko CA , Yandell B . 1974 . The relation of cognitive and memorial processes to persuasion in a simulated jury trial . J Appl Soc Psychol . 4 ( 1 ): 62 – 93 .

Goldstein JA , et al . 2023 . Generative language models and automated influence operations: emerging threats and potential mitigations, arXiv, arXiv:2301.04246, preprint: not peer reviewed .

Groh M , et al . 2023 . Human detection of political speech deepfakes across transcripts, audio, and video, arXiv, arXiv:2202.12883, preprint: not peer reviewed .

Epstein Z , Fang CM , Arechar AA , Rand DG . 2023 . What label should be applied to content produced by generative AI? PsyArXiv, preprint: not peer reviewed .

Author notes

Supplementary data, email alerts, citing articles via.

  • Contact PNAS Nexus
  • Advertising and Corporate Services
  • Journals Career Network


  • Online ISSN 2752-6542
  • Copyright © 2024 National Academy of Sciences
  • About Oxford Academic
  • Publish journals with us
  • University press partners
  • What we publish
  • New features  
  • Open access
  • Institutional account management
  • Rights and permissions
  • Get help with access
  • Accessibility
  • Advertising
  • Media enquiries
  • Oxford University Press
  • Oxford Languages
  • University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

  • Copyright © 2024 Oxford University Press
  • Cookie settings
  • Cookie policy
  • Privacy policy
  • Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Propaganda During World War II Essay

The Second World War was a complicated time for both the general public and the authorities since while the former worried for their safety, family, and homeland, the latter needed to maintain the national spirit and support the soldiers at the front. For such purposes, posters were implemented involving colorful images with strong words. However, while some might think that posters from the 20th century served as inspiration or plea, they were aimed to influence people psychologically.

The first propaganda poster Every minute counts! represents the influence of lost time on the battlefield failures of their soldiers. The technique used in this poster involves fear, through which the authorities strive to scare individuals working at manufacturing factories, urging them to work harder. In this sense, the poster incorporates statistics and figures, implying that every ten minutes that are lost will lead to less ammunition and weaponry, which will, in turn, postpone the victory.

Another poster, Air defense is home defense uses the technique of connecting with the audience. In their attempt to recruit as many individuals into air defense, the authorities aim to incorporate a heart-warming illustration of a family that looks in the sky and admires the national military plane. In a way, stereotypes in posters were common during wartime (Brewer 26). Here, the objective is to emphasize the pride in national defense and show the general public endorsement of the air forces.

The last poster, England expects, incorporates the technique of calling to action via bright colors, illustration of the national flag, and words. The phrase national service is written in bold red color that is contrasted by the dark blue background, which is used to catch the attention of the audience. Moreover, the number of people illustrated in the poster serves to show the national spirit, urging others to join the forces.

Hence, while some individuals might mistakenly believe that 20th-century posters acted as calls to action or acts of inspiration, their true purpose was to affect the audience psychologically. Every minute counts! is a propaganda poster that employs the technique of fear to illustrate the impact of wasted time on their soldiers’ failures on the battlefield. Another poster, Air defense is home defense , employs the audience-connection strategy. The final poster, England expects , employs the strategy of urging action via the use of bold colors, an image of the national flag, and text.

Brewer, Susan A. To Win the Peace: British Propaganda in the United States During World War II . Cornell University Press, 2019.

  • Chicago (A-D)
  • Chicago (N-B)

IvyPanda. (2023, September 21). Propaganda During World War II. https://ivypanda.com/essays/propaganda-during-world-war-ii/

"Propaganda During World War II." IvyPanda , 21 Sept. 2023, ivypanda.com/essays/propaganda-during-world-war-ii/.

IvyPanda . (2023) 'Propaganda During World War II'. 21 September.

IvyPanda . 2023. "Propaganda During World War II." September 21, 2023. https://ivypanda.com/essays/propaganda-during-world-war-ii/.

1. IvyPanda . "Propaganda During World War II." September 21, 2023. https://ivypanda.com/essays/propaganda-during-world-war-ii/.


IvyPanda . "Propaganda During World War II." September 21, 2023. https://ivypanda.com/essays/propaganda-during-world-war-ii/.

  • World War II Propaganda Posters in America
  • The Bonds or Bondage World War II Poster Analysis
  • The World War II Propaganda Techniques
  • "Incidents in the Life of a Slave Girl" by Harriet Jacobs: Urging Abolitionism Movements
  • World War II Propaganda and Its Effects
  • Anti-Japanese Propaganda During World War II
  • "And Our Flag Was Still There" by B. Kingsolver
  • American Homefront During World War II
  • Media Propaganda: Poster Advertisement
  • War on Terror: Propaganda and Freedom of the Press in the US
  • The Office of Strategic Services Operational Groups in World War II
  • Warfare: Ethical and Moral Issues
  • The Result Japan’s Fall in World War II
  • The Role of Propaganda During World War II
  • The Effectiveness of WWII Bombing Campaigns
  • Bibliography
  • More Referencing guides Blog Automated transliteration Relevant bibliographies by topics
  • Automated transliteration
  • Relevant bibliographies by topics
  • Referencing guides

Dissertations / Theses on the topic 'Propaganda'

Create a spot-on reference in apa, mla, chicago, harvard, and other styles.

Consult the top 50 dissertations / theses for your research on the topic 'Propaganda.'

Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.

You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.

Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.

Di, Jorio Irene. "Propagandare lo Stato : l'identità nazionale nella propaganda di Vichy." Paris 10, 2004. http://www.theses.fr/2004PA100024.

Gonçalves, Ana Paula Borges. "Propaganda tradicional X propaganda via web: um estudo exploratório." reponame:Repositório Institucional do FGV, 2001. http://hdl.handle.net/10438/8479.

Hidalgo, Milexa Villegas. "Propaganda global e a função criativa das agências de propaganda." reponame:Repositório Institucional do FGV, 1991. http://hdl.handle.net/10438/10475.

Klaehn, Jeffery. "The Propaganda Model." Thesis, University of Strathclyde, 2012. http://digitool.lib.strath.ac.uk:80/R/?func=dbin-jump-full&object_id=29564.

Frtúsová, Lucia. "Médiá a propaganda." Master's thesis, Vysoká škola ekonomická v Praze, 2007. http://www.nusl.cz/ntk/nusl-1932.

DANINI, FEDERICA. "La propaganda religiosa." Doctoral thesis, Università degli studi di Genova, 2020. http://hdl.handle.net/11567/1008934.

Giorio, Laura. "War on Propaganda or PRopaganda War? : A case study of fact-checking and (counter)propaganda in the EEAS project EUvsDisinfo." Thesis, Uppsala universitet, Teologiska institutionen, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-362064.

Liebel, Vinícius. "Humor, propaganda e persuasăo: uma análise do papel das charges na propaganda nazista /." oai:ufpr.br:223137, 2006.

Hermansson, Markus. "Superhjältar och propaganda : Superhjältar och deras fiender ur ett propaganda perspektiv under andarvärldskriget." Thesis, Karlstads universitet, Institutionen för samhälls- och kulturvetenskap (from 2013), 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:kau:diva-84223.

Giambusso, Anthony F. "Is propaganda pragmatic? : a study of the relationship between classical pragmatism and American propaganda /." Available to subscribers only, 2006. http://proquest.umi.com/pqdweb?did=1136088871&sid=23&Fmt=2&clientId=1509&RQT=309&VName=PQD.

Hanna, Finell. "Hur bemöts radikalnationalistisk propaganda? : En studie av den svenska statens bemötandestrategier av radikalnationalistisk propaganda." Thesis, Försvarshögskolan, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:fhs:diva-9629.

Pedroso, Roberto. "Propaganda: comunicação artistica e científica: um roteiro metodológico para desenvolvimento de campanhas de propaganda." reponame:Repositório Institucional do FGV, 1986. http://hdl.handle.net/10438/4684.

Quadros, Doacir Gonçalves de. "Partido político e propaganda política." reponame:Repositório Institucional da UFPR, 2012. http://hdl.handle.net/1884/28011.

Steele, Margaret. "Covenanting political propaganda : 1638-89." Thesis, University of Glasgow, 1995. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.312460.

Pereira, Júlia de S. B. "Propaganda institucional do Senado Federal." reponame:Repositório Institucional do UniCEUB, 2005. http://repositorio.uniceub.br/handle/235/7398.

Faro, Clara Freire Filgueiras. "Gosto popular na propaganda televisiva." Pontifícia Universidade Católica de São Paulo, 2014. http://tede2.pucsp.br/handle/handle/2491.

Gaunt, Sarah K. "English political propaganda, 1377-1485." Thesis, University of Huddersfield, 2018. http://eprints.hud.ac.uk/id/eprint/34644/.

Lambert, James K. "REEL NAZIS a propaganda history." Thesis, University of North Texas, 2005. https://digital.library.unt.edu/ark:/67531/metadc4954/.

Haroutyunian, Anna <1979&gt. "La propaganda monumentale nell’Unione Sovietica." Master's Degree Thesis, Università Ca' Foscari Venezia, 2021. http://hdl.handle.net/10579/19511.

Pu, Hong. "l’idéologie, la propagande et le cinéma chinois d’animation entre les années 20-70." Thesis, Montpellier 3, 2019. http://www.theses.fr/2019MON30007.

Brown, Jennifer Renee. ""Alfred Hitchcock presents; `Propaganda'" a rhetorical study of Alfred Hitchcock's World War II propaganda films /." Lynchburg, Va. : Liberty University, 2010. http://digitalcommons.liberty.edu.

Schwientek, Sabine. "Das Schandkleid die Erfolgsgeschichte patriarchalischer Propaganda: Entstehung, Entwicklung und sozialkulturelle Konsequenzen ; [4000 Jahre patriarchalische Propaganda] /." kostenfrei, 2009. http://d-nb.info/996638210/34.

Silva, Evilasio dos Santos. "A propaganda publicitária e sua significação no simbólico coletivo: análise discursiva de propagandas publicitárias do governo Dilma." Universidade Católica de Pernambuco, 2013. http://www.unicap.br/tede//tde_busca/arquivo.php?codArquivo=843.

Weitoft, Kristina. "Propaganda som vapen : En analys av Islamiska Statens propagandafilmer utifrån teorier om social identitet och propaganda." Thesis, Uppsala universitet, Religionssociologi, 2016. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-302581.

Welch, David. "The Third Reich politics and propaganda /." London : Routledge, 2002. http://site.ebrary.com/id/10205184.

Andiloro, Andrea. "Propagaming : Uncovering Propaganda In War Videogames." Thesis, Uppsala universitet, Medier och kommunikation, 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-324568.

Stegbauer, Andreas. "Rechtsextremistische Propaganda im Lichte des Strafrechts /." München : VVF, 2000. http://www.gbv.de/dms/spk/sbb/recht/toc/31469207X.pdf.

Petrakis, Marina. "Propaganda in Metaxas' Greece : 1936-1940." Thesis, University of Kent, 2001. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.342133.

Beleli, Iara 1955. "Marcas da diferença da propaganda brasileira." [s.n.], 2005. http://repositorio.unicamp.br/jspui/handle/REPOSIP/279871.

Nascimento, Álvaro César. "Propaganda de medicamentos. É possível regular ?" Universidade do Estado do Rio de Janeiro, 2007. http://www.bdtd.uerj.br/tde_busca/arquivo.php?codArquivo=3265.

Pietsch, Andreas. "Tönende Verführung NS-Propaganda durch Filmmusik." Berlin mbv, Mensch-und-Buch-Verl, 2009. http://d-nb.info/998718750/04.

Chagas, Raimundo Luís Fortuna. "Arquitetura no cinema, crítica e propaganda." reponame:Repositório Institucional da UFBA, 2013. http://www.repositorio.ufba.br/ri/handle/ri/11985.

Guedes, Sandra. "ORÍGENES LESSA E A PROPAGANDA BRASILEIRA." Universidade Metodista de São Paulo, 2008. http://tede.metodista.br/jspui/handle/tede/845.

Cunha, Carlos Eduardo Freitas da. "Agências de propaganda versus mercado anunciante." Florianópolis, SC, 2002. http://repositorio.ufsc.br/xmlui/handle/123456789/83635.

Jesus, Altair Reis de. "A imagem da recriação da juventude: televisão e propaganda." Programa de Pós- Graduação em Ciências Sociais da UFBA, 2008. http://www.repositorio.ufba.br/ri/handle/ri/11354.

Hynes, Greg. "Propaganda, Perspective, and the British World: New Zealand’s First World War Propaganda and British Interactions, 1914-1918." Thesis, University of Canterbury. School of Humanities and Creative Arts, 2013. http://hdl.handle.net/10092/9126.

Bleiber, Brigitte Marianne Di Nicola Antonio. "Die Oeffentlichkeitsarbeit der Schweizerischen Krebsliga von 1910 bis 1985 /." Bern, 1991. http://www.ub.unibe.ch/content/bibliotheken_sammlungen/sondersammlungen/dissen_bestellformular/index_ger.html.

Schroeder, Christy. "Red scare propaganda in the United States a visual and rhetorical analysis /." unrestricted, 2006. http://etd.gsu.edu/theses/available/etd-01042007-155247/.

Hosterman, Alec Ry'n. "Toward a more unified conceptualization of propaganda." Virtual Press, 1997. http://liblink.bsu.edu/uhtbin/catkey/1061886.

João, Aléxon Gabriel. "Garoto-propaganda Casas Bahia: configuração e estratégia." Universidade do Vale do Rio do Sinos, 2006. http://www.repositorio.jesuita.org.br/handle/UNISINOS/2608.

Schoales, Elizabeth. "Praise and propaganda : prophetic poetry in Wales." Thesis, University of Wales Trinity Saint David, 2003. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.683199.

Sheppard, Natalie R. "Invincible: Legacy and Propaganda in Superhero Comics." ScholarWorks@UNO, 2014. http://scholarworks.uno.edu/td/1943.

Moré, Amanda Daniela. "Propaganda e publicidade de alimentos no Brasil." Universidade de São Paulo, 2011. http://www.teses.usp.br/teses/disponiveis/11/11141/tde-13022012-152341/.

álvares, Thelma Lucia Guerra. "Análise discursiva da propaganda da absolut vodka." Universidade Católica de Pernambuco, 2016. http://tede2.unicap.br:8080/handle/tede/836.

Morais, Valdenir Máximo de. "Propaganda radiofônica: estudo do processo de realização." Universidade de São Paulo, 2011. http://www.teses.usp.br/teses/disponiveis/27/27153/tde-26092011-121743/.

Pagotto, Erico Luciano. "Greenwashing: os conflitos éticos da propaganda ambiental." Universidade de São Paulo, 2013. http://www.teses.usp.br/teses/disponiveis/100/100134/tde-22072013-141652/.

Lessa, Laís Quintella Malta. "Pop art e propaganda: uma relação interdisciplinar." Universidade Presbiteriana Mackenzie, 2009. http://tede.mackenzie.br/jspui/handle/tede/2726.

Lara, Margarete Maria Fernandes do Nascimento. "Propaganda bancária: uma análise textual e orientada." Pontifícia Universidade Católica de São Paulo, 2009. https://tede2.pucsp.br/handle/handle/14593.

Nunes, Eunice Ferreira. "Gênero discursivo propaganda impressa: uma experiência pedagógica." Pontifícia Universidade Católica de São Paulo, 2010. https://tede2.pucsp.br/handle/handle/14631.

Terci, Carla Francini Hidalgo. "Propaganda e educação na sociedade cosmeticamente mediada." reponame:Repositório Institucional da UnB, 2016. http://repositorio.unb.br/handle/10482/21388.

  • Share full article

propaganda thesis paper

They Used to Award Olympic Medals for Art?

The founder of the modern Games thought they should honor both body and mind. But the tradition died years ago, and the winning artworks are largely forgotten.

The Grand Palais in Paris was home to the arts portion of the 1924 Olympics. Among the pieces on display was the Discus Thrower (foreground), which won the gold medal. Credit... Costas Dimitriadis Archives, Collection Katia Iakovidou and Yiannis Anagnostou

Supported by

John Branch

By John Branch

John Branch reported from museums in Paris, Nice, Lausanne and Dublin, and near bridges and offramps on Randall’s Island in New York.

  • May 2, 2024 Updated 10:41 a.m. ET

During all of the years that the Olympics gave out medals in arts, not just athletics — and if you didn’t know about that, the rest of this article may hold more surprises — the pinnacle came in Paris, 100 years ago this summer.

The gold medal sculpture at the 1924 Paris Olympics was by a Greek artist named Costas Dimitriadis. His nude, arching, 7-foot “Discobole” (Discus Thrower) was for weeks displayed prominently in the Grand Palais.

Two years later, before “a crowd of light-frocked women and straw-hatted men,” as The New York Times reported , the prized sculpture, cast in bronze, was planted just outside the Metropolitan Museum of Art in New York’s Central Park.

“A symbol of human perfection,” a museum official declared that day.

The statue did not stand still for long. Like the Olympic arts contests themselves, it went on quite a journey, largely to oblivion.

propaganda thesis paper

‘Pentathlon of the Muses’

For decades, beginning with the 1912 Olympics in Stockholm, the Olympics included competitions in painting, sculpture, architecture, music and literature — a “pentathlon of the Muses,” as Pierre de Coubertin, the founder and leader of the modern Olympics, called them.

“From now on they will be part of each Olympiad, on a par with the athletic competitions,” Coubertin said.

Thousands of artists, some of them famous, most of them not, submitted works. More than 150 Olympic arts medals were awarded , the same medals that athletes received. At the 1932 Olympics in Los Angeles, 400,000 people visited the monthlong exhibition of entries.

As the Olympics return to Paris this summer, thousands of gold, silver and bronze medals will be awarded — all for sport, none for arts.

“The spirit of Pierre de Coubertin did not survive,” said Nikoleta Tzani, a Greek art historian who wrote her doctoral thesis on Dimitriadis.

But some artwork did. It is scattered around the globe, some of it in museums or parks, some in private collections, much of it just lost to time and indifference.

In Lausanne, Switzerland, home of the International Olympic Committee, the Olympic Museum has a secured storage area in the basement. Curators oversee thousands of pieces of sports equipment, uniforms, medals, documents, torches, trophies — and art work.

But the only gold medal-winning paintings in the hidden collection are the two colorful pieces of a triptych that earned newspaper illustrator Jean Jacoby, of Luxembourg, first place in 1924. One represents soccer, the other rugby. The whereabouts of the third oil painting, depicting the start of a foot race, is unknown.

Upstairs, museum visitors learn all about Coubertin and see mementos of Olympic sports. There is no indication that the Olympics ever held serious arts competitions.

Outside the museum, though, overlooking Lake Geneva, a hillside garden is speckled with sculptures. A keen-eyed visitor might take note of one particular piece. It is Dimitriadis’ Discus Thrower, a late 20th-century copy of the one in New York.

“The original won first prize in the sculpture section of the art contest held during the 1924 Paris Olympic Games,” a nearby placard reads, cryptic enough to raise more questions than it answers.

Coubertin’s Vision

Coubertin, born in Paris in 1863, was raised in an aristocratic family and schooled in classical art and literature. He dreamed up the modern Olympics, based on the Games of ancient Greece, and launched them in Athens in 1896.

Like the Greeks, Coubertin believed that the Olympics should be about both body and mind. But just reviving the Olympics was difficult enough. The first three editions, to his frustration, featured only sports.

“Deprived of the aura of the Arts contests,” he once declared, “Olympic Games are only World Championships.”

In 1906, Coubertin invited dozens of artists and art figures to the Comédie Française in Paris. He proposed competitions in architecture, sculpture, painting, music and literature, “designed to be awarded every four years for unpublished works directly inspired by the idea of sport.”

He argued that athletics and academics should mix.

“The coming generation will know mental workers who will at the same time be sportsmen,” Coubertin said. “Are there not some already among the fencers?”

By day’s end, it was agreed: Art would be part of the Olympics, and medals would be awarded, beginning in 1912.

Nervous that he would receive too few submissions, Coubertin entered the literature contest himself. He submitted a florid “Ode to Sport” under the pseudonymous pairing of “Georges Hohrod and Martin Eschbach,” the surnames borrowed from the names of French villages.

In what could have become the first great I.O.C. scandal had anyone noticed before Coubertin died 25 years later, Hohrod and Eschbach won the gold medal. No other work was deemed worthy of a silver or bronze, including one by the Italian poet and novelist Gabriele d’Annunzio . (Another esteemed Italian who did not win? Sculptor Rembrandt Bugatti , of the famous car family.)

The next Olympics, planned for Berlin in 1916, were canceled by the war. The 1920 Antwerp Games were a patchwork affair held in the wake of the European devastation.

Then came Paris.

“It’s the most important arts competition,” said Christopher Young, a professor at Cambridge and co-curator of an upcoming exhibition at the Fitzwilliam Museum about the 1924 Olympics . “It’s the final year of Pierre de Coubertin’s tenure as Olympic president, it’s a moment of technological change, and the arts scene in Paris is very vibrant.”

The arts contests had found their time and place.

1924 and ‘The Liffey Swim’

While Jacoby’s winning entry from the 1924 Olympics is usually buried in storage, the painting that earned the silver medal is prominently displayed in a major European museum.

“The Liffey Swim,” an oil painting by Jack Butler Yeats , hangs in Room 14 of the National Gallery of Ireland , in Dublin. That it finished second to Jacoby’s triptych is a point of bemusement among art critics, and highlights the uneven judging of the Olympic arts competitions.

Jacoby is little known, at least outside of Luxembourg, where he has a small track stadium named for him and a bit of notoriety as only person to win two Olympic gold medals in art.

Yeats, by contrast, is considered Ireland’s most famous artist.

“He is a huge figure here,” Brendan Rooney, the National Gallery’s head curator, said. The museum owns dozens of his works — and his 1924 silver medal.

In 1923, the same year that his brother, the poet William Butler Yeats, earned the Nobel Prize in Literature, and just as Ireland was becoming independent, Yeats painted a modern scene depicting a swimming race down the Liffey River, through the heart of Dublin. The race remains a rambunctious part of Dublin’s summer calendar .

“It is an optimistic painting for a cautiously optimistic nation that had been through many very difficult years,” Rooney said.

The gallery’s accompanying placard notes that it “marks Yeats’ growing interest in Expressionism.” It is widely believed that Yeats had painted himself and his wife prominently within the crowd. “His ability to capture moments in crowds with figurative details was second to none,” Rooney said.

The painting was exhibited in London, Brussels and at the 1939-40 New York World’s Fair. With a laugh, Rooney suggested that Yeats had been the victim of artistic injustice, as his painting was modern and evocative compared to Jacoby’s conventional work.

Olympic entries skewed toward the classic and idealized, even in Paris in 1924. The Discus Thrower, the gold-winning sculpture later erected in New York, is a prime example.

“If you were a card-carrying member of the avant-garde or some kind of modernist society of the early 20th century, that wasn’t going to help you much,” Miles Osgood, a Stanford lecturer who has deeply researched the Olympic arts competitions, said. “It’s really fascinating to see how many of those artists were drawn to these contests, if only to lose them.”

Among those who submitted works but won nothing: the British poet and novelist Robert Graves in 1924, the Dadaist George Grosz, in 1928, and Walter Gropius, founder of the Bauhaus school of architecture, in 1932.

The best-known gold medal winners include French sculptor Paul Landowski and Dutch painter Isaac Israëls . Laura Knight , one of Britain’s premier female artists, earned a silver medal. That was all in 1928, a strong year for entries, buoyed by Paris in 1924.

Osgood sees Yeats’ medal-winning work reflected in the upcoming Olympics, where organizers hope to hold open-water swimming and part of the triathlon in the Seine River.

“‘The Liffey Swim’ comes to life in a new city,” Osgood wrote in an essay accompanying the Fitzwilliam exhibition. “If the Olympic Art competitions — with their rule books, juries and medals — once convinced art to imitate sport, it is now time for sport to imitate art.”

“The Liffey Swim" will be lent, along with Yeats’ silver medal, to the Fitzwilliam exhibit, scheduled to run from July 19 to Nov. 3. The Olympic Museum is sending the two Jacoby pieces, too. Art fans can judge them, side by side, as juries in Paris did 100 years ago, when Coubertin highlighted the arts at the 1924 closing ceremony.

“There is need for something else besides athleticism and sport,” he said. “We want the presence of national genius, the collaboration of the Muses, the cult of beauty, all the display pertaining to the strong symbolism incarnate in the past by the Olympic Games and which must continue to be represented in our modern times.”

They continued, but not for long.

End of an Era

The literature category of Olympic art rarely distinguished itself, but one entry for 1932 was illuminating. It was by an American, Avery Brundage, president of the American Olympic committee.

“Art, music, architecture, literature, all seem to have been prostituted in this commercial age to the greed for money,” Brundage wrote in “The Significance of Amateur Sport.” His words earned honorable mention but no medal.

Riding on momentum from Paris, the arts competitions coasted through Amsterdam in 1928 and Los Angeles in 1932. But at the Berlin Olympics in 1936, the competition fell under the direction of Joseph Goebbels, the Nazi propaganda minister.

There were international calls to boycott those Olympics, though Brundage notoriously supported American participation . France and Britain were among the countries that competed in athletics but had no arts entries. Americans had just one medalist, despite dozens of submissions. Germans dominated.

Coubertin died in 1937. World War II canceled the next two Olympics. When they resumed in London in 1948, the arts were a diluted part of the program. The United States did not participate.

A year later, Brundage, now the I.O.C. vice president, co-wrote a report on the state of the Olympics. Item 23 was about the arts. “Since art competition contestants are practically all professionals, Olympic medals should not be awarded,” the report said. “This event should be in the nature of an exhibition.”

There were other concerns. Unlike most sports, arts contests were subjective and ill-suited for the dawning age of television. The Olympics had blossomed into the pinnacle of sport, but not of art, not “on par with the athletic competitions,” as Coubertin had envisioned.

The 1952 contest in Helsinki was canceled, and in 1954, with Brundage now president, the I.O.C. voted to end the arts competitions.

They were mostly forgotten. But not by everyone.

From Paris to Randall’s Island

Patricia Reymond is senior collections manager for the Olympic Museum in Lausanne. On a weekday in March, in the quiet basement storage area, she pondered where all the Olympic art went.

“Of course we would love to have different pieces by the winning artists,” she said, standing near one of the Jacoby paintings. “But it is difficult.”

Researchers have lists of artists and artwork titles submitted for each Olympics, but descriptions are absent or vague. Artists sometimes created multiple works with similar titles, making verification difficult. Little was photographed, or was captured only in black and white.

Curators scan auctions and online sales. Language is a barrier. Not all advertised pieces are declared as having been a part of the Olympics.

Would curators notice if Carlo Pellegrini’s “Winter Sports” frieze, the first winning painting in 1912, went up for sale? (Probably, because its whereabouts have long proved vexing.) Would they catch it if “Rodeo” by Lee Blair, a Disney illustrator who won gold in 1932, hit the market? (Maybe. It was donated to a California high school and has been lost for decades .)

The case of the Discus Thrower might be the most illustrative. A Greek American tobacco executive named Ery Kehaya commissioned the first bronze version for New York. (A second was cast in 1927 and still stands near the historic Panathenaic Stadium in Athens .)

But a decade after its festive unveiling outside the Met, it was uprooted and replanted on New York’s Randall’s Island, in front of a new stadium where the sprinter Jesse Owens would qualify for the Berlin Olympics.

The sporting location made some sense, but New Yorkers have long been more likely to cross the bridges over Randall’s Island in a car or a train than to stop there. About 500 acres in size and virtually uninhabited, the island is where Manhattan, Queens and the Bronx are stitched together by the Robert F. Kennedy (formerly the Triborough) Bridge . Below is parkland and sports facilities, mostly, maintained by the Randall’s Island Park Alliance .

The organization’s logo is a silhouette of the Discus Thrower. Depicting an old-style, two-handed throwing form, it looks a bit like a muscular nudist leaning back to hand a loaf of bread to someone behind him in a crowd.

“Most people don’t know what it represents,” Deborah Maher, the alliance’s president, admitted.

The statue, like the stadium it proclaimed, did not age well. Worn, weathered and vandalized, missing an arm and the discus itself, it was quietly removed in 1970 and put in storage, and was not missed by many.

But the Discus Thrower was rediscovered, refurbished and rededicated in 1999 . With the old stadium headed for demolition, it was placed on a grassy traffic island at the foot of a Manhattan exit ramp — a lonely greeter, to a mostly empty island. He didn’t even look drivers in the eye.

It stood there until last fall, when Maher and others got nervous about encroaching road construction. They hauled it away and, on April 16, the Discus Thrower was unveiled, again, outside the 5,000-seat Icahn Stadium, which had opened in 2005 on the site of the previous, larger stadium.

The statue is the centerpiece of a new $1.6 million plaza. It stands tall within a round, raised flower bed and is illuminated on four sides.

“We wanted to highlight him,” Maher said.

That this is 2024, the 100th anniversary of the Discus Thrower’s gold medal, the centennial of one Paris Olympics and the year of another, is a happy coincidence.

“The whole thing was serendipitous,” Maher said.

To be replanted in a raised flower bed on Randall’s Island may not represent the grandest of Olympic comebacks. But it’s a fate far better than that of the old arts competitions themselves, buried by history long ago.

John Branch is a sports reporter. He won the 2013 Pulitzer Prize for feature writing for “Snow Fall,” a story about a deadly avalanche in Washington State, and is the author of three books, including “Sidecountry,” a collection of New York Times stories, in 2021. More about John Branch

The Great Read

Here are more fascinating tales you can’t help reading all the way to the end..

When an illegal smoke shop opened across the street, an Upper West Side councilwoman, vowed to close it. What happened next was “like a Fellini movie.”

The diabetes drug Ozempic has become a phenomenon, and its inescapable jingle — a takeoff of the Pilot song “Magic” — has played a big part in its story .

A man’s five-year stay at the New Yorker Hotel cost him only $200.57. Now it might cost him his freedom .

Researchers are documenting deathbed visions , a phenomenon that seems to help the dying, as well as those they leave behind.

Around 2020, the “right” pants began to swing from skinny to wide. But is there even a consensus around trends anymore ?



  1. World War I: Propaganda Essay Example

    propaganda thesis paper

  2. Samples Of Thesis

    propaganda thesis paper

  3. Argumentative Essay

    propaganda thesis paper

  4. Wwii propaganda thesis

    propaganda thesis paper

  5. Apa Format Thesis Paper Sample

    propaganda thesis paper

  6. Sample of front page of thesis

    propaganda thesis paper


  1. Propaganda in USSR during WW2

  2. Thesis Paper Support in BD

  3. Rating Chinese Propaganda Paper 9/10 #china #propaganda #ccp #friendly #graphicdesign #politics

  4. Propaganda "Calling On Moscow"

  5. Thesis and Dissertation Evaluation Format in All Ethiopian Universities(በአማርኛ)

  6. How to (not) be Popular: The Making of a 1950's Emo (vs Hipster Punk)


  1. Propaganda, misinformation, and histories of media techniques

    This essay argues that the recent scholarship on misinformation and fake news suffers from a lack of historical contextualization. The fact that misinformation scholarship has, by and large, failed to engage with the history of propaganda and with how propaganda has been studied by media and communication researchers is an empirical detriment to it, and

  2. Protecting the image of a nation: Jim Crow propaganda

    functioned as a propaganda machine on behalf of the United States government at the dawn of the Cold War and the Civil Rights Movement. Drawing from literature on propaganda, public relations, and public diplomacy, this thesis connects 20th century American propaganda to its roots in public relations, communication studies, and psychology.

  3. PDF Understanding Media Propaganda in the 21st Century

    Here Simon Foley turns the Herman/Chomsky propaganda thesis round from its more usual targets and deconstructs the Irish Times and the Guardian's coverage of recent foreign policy controversies. Some of the conclusions about, for example, Syria will ruffle feathers. This book provides a counterpoint to challenge

  4. (PDF) Propaganda

    Abstract. Propaganda is sponsored information that uses cause‐ and emotion‐laden content to sway public opinion and behavior in support of the source's goals. Propaganda utilizes mass media to ...

  5. Defining propaganda: A psychoanalytic perspective

    This essay proposes to define propaganda through psychoanalytical research pioneered by Erich Fromm on symbiotic relations. Symbiotic relations, when transferred from biology to psychology and sociology, describe a process of allowing a person to merge with something big and important, therefore creating meaning beyond an individual's life. ...

  6. The Thin Line Between Propaganda and Persuasion

    This Open Access Thesis is brought to you for free and open access by the Theses and Dissertations at OpenSIUC. ... "THE THIN LINE BETWEEN PROPAGANDA AND PERSUASION" (2013).Theses.Paper 1322. THE THIN LINE BETWEEN PROPAGANDA AND PERSUASION by Ryan Jenkins B.A., Southern Illinois University, 2005 A Thesis Submitted in Partial Fulfillment of the ...

  7. (PDF) The Propaganda Model: Theoretical and ...

    This article provides an overview of the Propaganda Model and rehearses central theoretical considerations concerning the model's overall understanding of media behaviour. The article then ...

  8. Fake News and Propaganda: A Critical Discourse Research Perspective

    propaganda, and suggests ways in which such a typology may help us make sense of post-truth and the recent spread of fake news. According to Ellul (1973), "vertical" propaganda refers to the effort of a higher-order entity (such as a State or organization) to subvert an existing government, political enemies within

  9. PDF Online Jihadism: Propaganda, Recruitment and Homegrown Radicalization

    PROPAGANDA, RECRUITMENT AND HOMEGROWN RADICALIZATION by Gregory Shuck A thesis submitted to Johns Hopkins University in conformity with the requirements for the ... This paper examines this threat of Jihadist use of the internet and notes several themes. The first chapter addresses a broad base of contested academic theories, such

  10. Propaganda Analysis Revisited

    Propaganda Analysis Revisited. This special issue is designed to place our contemporary post-truth impasse in historical perspective. Drawing comparisons to the Propaganda Analysis research paradigm of the Interwar years, this essay and issue call attention to historical similarities between patterns in mass communication research then and now.

  11. Characterizing networks of propaganda on twitter: a case study

    The paper aimed at providing new insights into the dynamics of propaganda networks on Twitter. The results of our study are partly in line with existing research. Modularity-based clustering, applied to retweet graphs, pictured a wide panorama of communities of users with strong homophily/affiliation and polarized position.

  12. (PDF) Fake News and Cyber Propaganda: A Study of ...

    development of 21 st century mediascape has been perceived as the key reason for the scale and. power of propaganda to grow in an easy and smart way. In the recent times, cre ating and planting ...

  13. Nazi Propaganda Visual Essay

    This visual essay includes a selection of Nazi propaganda images, both "positive" and "negative.". It focuses on posters that Germans would have seen in newspapers like Der Stürmer and passed in the streets, in workplaces, and in schools. Some of these posters were advertisements for traveling exhibits—on topics like "The Eternal ...

  14. How persuasive is AI-generated propaganda?

    As shown in Fig. 1, the original propaganda was highly persuasive. While only 24.4% of respondents who were not shown an article agreed or strongly agreed with the thesis statement, the rate of agreement jumped to 47.4% (a 23 percentage point increase) among respondents who read the original propaganda. Thus, the original propaganda nearly ...

  15. PDF Propaganda, misinformation, and histories of media techniques

    This essay argues that the recent scholarship on misinformation and fake news suffers from a lack of historical contextualization. The fact that misinformation scholarship has, by and large, failed to engage with the history of propaganda and with how propaganda has been studied by media and communication

  16. Roman Propaganda in the Age of Augustus

    Augustus' control of the famed Roman Legions was critical in his ability to maintain. power, but it was also the manipulation of the Roman people at all levels of society that enabled him to maintain that power. Augustus ushered in an unprecedented age of peace and stability.

  17. PDF Comic Books As American Propaganda During World War Ii

    heroics, the use of scrap, rubber, paper, or bond drives back on the homefront to provide resources on the frontlines, to a general sense of patriotism. This research looks to the motivations behind such storytelling in the background of comic book writers and artists as well as involvement from government agencies such as the War Writer's Board.

  18. Propaganda

    Propaganda, dissemination of information—facts, arguments, rumors, half-truths, or lies—to influence public opinion. Deliberateness and a relatively heavy emphasis on manipulation distinguish propaganda from casual conversation or the free and easy exchange of ideas. Learn more about propaganda in this article.

  19. (PDF) The Great War: Cinema, Propaganda, and The ...

    E-mail: [email protected]. Abstract. The relation between war and cinema, propaganda and cinema is a. most intriguing area, located at the intersection of media studies, history and ...

  20. Propaganda During World War II

    Propaganda During World War II Essay. The Second World War was a complicated time for both the general public and the authorities since while the former worried for their safety, family, and homeland, the latter needed to maintain the national spirit and support the soldiers at the front. For such purposes, posters were implemented involving ...

  21. Dissertations / Theses: 'Propaganda'

    The thesis work develops the long history of Chinese animated cinema over the period from 1920 to 1977, from the beginning of the 20th century until the end of the great proletarian cultural revolution. The subject of the thesis is: ideology, propaganda and animated Chinese cinema between the years 20-70.

  22. Propaganda and Its Use During Wartime Thesis

    Thesis. Pages: 7 (1918 words) · Bibliography Sources: 4 · File: .docx · Level: College Senior · Topic: Drama - World. Graphic Propaganda: Posters Used by the United States during World War II. Most Americans today are undoubtedly familiar with the ubiquitous and firm visage of "Uncle Sam" pointing a finger at them with the compelling ...


    The origins of the war film, 1897-1902 In this thesis I present the first detailed treatment of war and early cinema, describing the representation of conflicts in film from the Greco-Turkish War ...

  24. They Used to Award Olympic Medals for Art?

    The founder of the modern Games thought they should honor both body and mind. But the tradition died years ago, and the winning artworks are largely forgotten.