Parole de présidents (1958-2022)

En plus de soixante ans, huit pr\'esidents se sont succ\'ed\'e \`a la t\^ete de la Ve R\'epublique fran\c{c}aise (de Gaulle, Pompidou, Giscard d'Estaing, Mitterrand, Chirac, Sarkozy, Hollande, Macron). Apr\`es avoir pr\'esent\'e le corpus de leurs discours -- soit 9202 textes et plus de 20 millions de mots \'etiquet\'es -- le style de chacun des pr\'esidents sera caract\'eris\'e \`a l'aide de leurs vocabulaire (vocables et cat\'egories grammaticales). Une analyse plus approfondie r\'ev\`ele les s\'equences typiques de chaque locataire de l'\'Elys\'ee. Bas\'ee sur les distances entre l'ensemble des allocutions, une figure illustre les similitudes et diff\'erences entre les diff\'erents pr\'esidents. Over the past sixty-six years, eight presidents successively headed the Fifth French Republic (de Gaulle, Pompidou, Giscard d'Estaing, Mitterrand, Chirac, Sarkozy, Holland, Macron). After presenting the corpus of their speeches -- 9,202 texts and more than 20 million labelled words -- the style of each of them will be characterized by their vocabulary (lemmas and part-of-speech). A deeper analysis reveals the typical sequences of each tenant of the Elys\'ee. Based on an intertextual distance between all presidential speeches, a synthesis can be drawn reflecting the similarities and differences between presidents.

ChatGPT as speechwriter for the French presidents

Generative AI proposes several large language models (LLMs) to automatically generate a message in response to users' requests. Such scientific breakthroughs promote new writing assistants but with some fears. The main focus of this study is to analyze the written style of one LLM called ChatGPT by comparing its generated messages with those of the recent French presidents. To achieve this, we compare end-of-the-year addresses written by Chirac, Sarkozy, Hollande, and Macron with those automatically produced by ChatGPT. We found that ChatGPT tends to overuse nouns, possessive determiners, and numbers. On the other hand, the generated speeches employ less verbs, pronouns, and adverbs and include, in mean, too standardized sentences. Considering some words, one can observe that ChatGPT tends to overuse "to must" (devoir), "to continue" or the lemma "we" (nous). Moreover, GPT underuses the auxiliary verb "to be" (^etre), or the modal verbs "to will" (vouloir) or "to have to" (falloir). In addition, when a short text is provided as example to ChatGPT, the machine can generate a short message with a style closed to the original wording. Finally, we reveal that ChatGPT style exposes distinct features compared to real presidential speeches.

Stylistic Analysis of the French Presidential Speeches: Is Macron really different?

Presidential speeches indicate the government's intentions and justifications supported by a dedicated style and rhetoric oscillating between explanation and controversy. Over a period of sixty years, can we observe stylistic variations by the different French presidents of the Fifth Republic (1958-2018)? Based on official transcripts of all their allocution, this paper illustrates the stylistic evolution and presents the underlying main trends. This study shows that de Gaulle's rhetoric is not mainly dedicated to his own person, or that the two terms of J. Chirac are not fully similar. According to several overall stylistic indicators, Macron's style does not appear as complex compared to his predecessors (F. Hollande or N. Sarkozy) but a more careful analysis clearly demonstrates his noticeable new style. Compared to the recent US presidents, the French ones present some similarities (e.g., similar mean sentence length) and dissimilarities (more I-words, less we-words). In this comparative analysis, Macron's style is also clearly distinctive from both the US and former French presidents. Opting for a more abstract discourse, less anchored in space, using less numbers, E. Macron tends to use long sentences. These various stylistic and rhetorical features could explain his being misunderstood by the French people and his recurrent low approval ratings.