Des corpus audiovisuels en Humanités. Méthodes, expériences, résultats

Audiovisual Corpora in Humanities: Methods, Experiences, Results.

30-31 May 2024 MSH Paris Nord (France)

FR EN

Presentation

The design, analysis, traffic flow or archiving of films and videos are at the heart of a growing number of research projects in Arts, Humanities, and Social Sciences. The emergence of digital tools has speeded up this process: annotating, preserving and sharing audiovisual content has become a common practice for many researchers, despite the technical and scientific raised by these corpora. This international conference focuses on different approaches and methods applied to the study of natively digital or digitized audiovisual content. It aims to bring together researchers from different fields (History, Sociology, Media Studies, Education sciences, Film Studies, Performing arts, Anthropology, etc.) to share case studies and concrete experiences.

Event description

The filmmaking, the analysis, the online publishing and the archiving of films and videos are long-standing issues in the field of Arts, Literature, Humanities or Social Sciences. These practices are at the heart of Film studies (Metz, 1977 ; Aumont & Marie, 1989) and Visual studies (Bartholeyns, 2016 , Pinotti et al., 2022). In History or Anthropology, they have been widespread and even institutionalized for decades (Ferro, 1974 ; Rousso, 1987 ; Colleyn, 1994). Then, they gradually entered other fields, such as Sociology (Sorlin, 1977; Tessier, 1999, Sebag & Durand, 2020), research in Education (Jacquinot, 2012), Media Studies (Odin, 2011; Soulez, 2011; Scopsi, 2010) or research in Performing arts (Bardiot, 2020; Chantraine-Braillon, 2022; Le Marec & Sauret, 2016).

The rise of computers, digital technology and Internet have both introduced and strengthened these practices. This digitization has not only made it easier the sharing of content: it has also facilitated activities such as video making and editing that were previously for professionals only. The increasingly massive production of digital audiovisual content (DAC), as well as their increasing exposure on Youtube, Vimeo and social networks, such as TikTok or Instagram, have opened up new horizons for researchers. As a result, they have modified their ways of working, taking advantage of digital tools to build new corpora and shape new working methods (Bourgatte, 2012; Ouvrard & Farge, 2022; Sensevy, Tiberghien, Veillard, 2022).

The software Mediascope developed by the INA (French Audiovisual Institute) has provided a solution adapted to the exploration and the analysis of the collection of this institution; other softwares have also been developed as part of research projects, such as Advene, Lignes de temps or Annoto, and were either partial or temporary responses to researchers' needs (Aubert, Prié & Schmitt, 2012; Puig & Sirven, 2007; Blau, 2021). AI experiments - as with Distant Viewing - has also emerged in recent years (Arnold & Tilton, 2023). But neither a single method nor a single technological solution has become the norm in Humanities and Social Sciences (Melgar-Estrada & Koolen, 2018) or among the community of digital humanities researchers, even though one of the main challenges of the latter is precisely to take hold of all forms of literacies "outside the book”, whether they be new forms of oral, visual, audiovisual or multimedia culture (Clivaz & Vinck, 2014).

When it comes to audiovisual corpora, there are still a number of "do-it-yourself" approaches, which contrastwith the clearer advances made in the field of instrumented text analysis (Pulizzotto, 2019). In addition to a lack of familiarity with the tools or hesitations regarding their use, research focused on DAC remains caught between moments of enthusiasm (the emergence of new methods or technologies) and others of disappointment (unsatisfactory results, difficulty in getting to grips with the tools, recurring technical problems, etc.) (Bourgatte & Tessier, 2017; Besson & Lavorel, 2023). As a result, researchers thus often continue to work by taking notes on paper or in a word processor, which doesn't facilitate film mining, let alone its valorization, promotion or circulation to a wider audience.

If the study of the DAC raises issues, other questions also need to be considered, such as those of the preservation, mediatization and sharing of research materials and results, whether it's the audiovisual content itself, metadata or annotations (Fleckinger, 2011). Researchers are increasingly concerned about how to archive, promote and exchange their resources. Here again, several services or platforms offer solutions that can be mobilized (Youtube, Archive.org), but none of them succeeded in establishing itself definitively, for various reasons: inability to adapt digital tools to specific needs, data standards or formats, security, legal and ethical issues, etc. On all these topics, a collective scientific instrumentation movement should open up new perspectives by simplifying procedures, enabling more systematic research, perpetuating results and encouraging sharing, in a process of open science. The question of the free and open-source nature of the technologies deployed or used for DAC exploration is, in this sense, a key issue.

Based onthese observations, the consortium CANEVAS (translation of the French acronym for: Consortium for the annotation, analysis and archiving of video applied to scientific activities) was launched and certified by Huma-Num in 2022. Hosted by the Maison des Sciences de l'Homme - Paris Nord, its aim is precisely to instrument research in Humanities and Social Sciences around audiovisual content and video corpora, by facilitating actions such as: annotating, commenting, analyzing, searching, sharing or archiving. To this end, consortium members are helping to develop tools and standards (FAIR) and support the application of best practices. Moving forward on these issues means answering a key question that will be at the heart of the conference organized by the consortium CANEVAS: how do researchers proceed in practice? What methods and tactics do they use to work on audiovisual corpora and share the results of their investigations?

Suggested topics

Proposals should present an audiovisual corpus, the methodology used, the tools employed to analyze the corpus, and the research results. Among the dimensions addressed in these proposals, the following elements may (among others) be dealt with:

DAC annotation. Note-taking is a natural part of any research activity (whether or not it relates to the video): are specific softwares used or, on the contrary, bypass strategies or DIY strategies employed (Archat-Tatah & Bourgatte, 2014; Boutin & Laborde, 2023)?

Video archiving. This is necessary to ensure that DAC can be consulted and stored (Véray, 2011; Maeck & Steinle, 2016). What are the most appropriate technological choices? Can and should we rely on private services (Youtube, Vimeo) or, on the contrary, turn to public platforms? What role does a player like the Internet Archive play? What alternatives exist today?

Sharing DAC within a community of peers is sometimes necessary as part of a collaborative research project or educational activity (Fournout, Beaudouin & Ferrarese, 2014; Kaufman & Mohan, 2009). How can these collaborative projects be implemented, and under what conditions can they achieve their goals?

The publication and mediatization of research results in audiovisual formats or based on the use of audiovisual resources are also a necessity for many projects, whether to present a research field or to administer evidence (Lacković, 2018). In a context where the written publishing of research results remains central (papers, chapters or books), what solutions can be considered?

Questions related to rights (image or copyright) raised by DAC are configured in different ways depending on the type of video (produced or not within the framework of research) and they may also be addressed.

Finally, the question of the standards and formats chosen for the audiovisual content themselves, their metadata and the textual or multimedia data could be an area for further thought. How can we get to grips with this issue, and work towards the adoption of common standards and formats? Is this even necessary or desirable?

Participants in this conference are invited to answer one or more of these questions, drawing on their own fieldwork and experience. Beyond the presentation of their research results, this conference will also look at the ways in which they have proceeded, the difficulties they faced and the technologies they have experimented with.

References

Archat-Tatah, Caroline, Bourgatte, Michaël (2014). « Vers des formes instrumentées d’enseignement et d’apprentissage : le cas de l’analyse de contenus audiovisuels », Eduquer Former, no 46, 69‑92.

Arnold, Taylor, Tilton, Lauren (2023). Distant viewing: computational exploration of digital images, MIT Press

Aubert Olivier, Prié, Yannick (2005). « Advene: Active Reading through Hypervideo », in Proceedings of ACM Hypertext'05, 235-244.

Aumont Jacques, Marie, Michel (1989). L’analyse des films, Nathan.

Bardiot, Clarisse (2020). « Theatre Analytics: Developing Software for Theatre Research », Digital Humanities Quarterly, vol. 14-3. http://www.digitalhumanities.org/dhq/vol/14/3/000476/000476.html

Bartholeyns, Gil (dir.) (2016). Politiques visuelles, Les Presses du Réel.

Besson, Rémy, Lavorel, Marie (2023). L’annotation vidéo pour la recherche. Usages et outils numériques, MkF.

Blau, Ina, Tamar, Shamir-Inbal (2021). « Writing private and shared annotations and lurking in Annoto hyper-video in academia: insights from learning analytics, content analysis, and interviews with lecturers and students », Educational Technology Research and Development, 69-2, 763‑786.

Bourgatte, Michaël (2012). « L’écran-outil et le film-objet », MEI, 34, 103‑120.

Boutin, Perrine, Laborde, Barbara (2023). « La pratique de l’essai vidéo comme forme d’empowerment », in Camille Roelens & Chrysta Pélissier (dir.), Éthique, numérique et idéologies, Presses des Mines, 175-194.

Chantraine-Braillon, Cécile (2022), « L’École du spectateur : informatiser la recherche en arts de la scène », Humanités numériques, 5, https://doi.org/10.4000/revuehn.2849

Clivaz, Claire, Vinck, Dominique (2014). « Introduction. Des humanités délivrées pour une littératie plurielle», Les Cahiers du numérique, vol. 10-3, 9-16.

Fleckinger, Hélène (2011). Cinéma et vidéo saisis par le féminisme (France, 1968-1981), Thèse de doctorat, Paris 3 Sorbonne Nouvelle.

Fournout, Olivier, Beaudouin, Valérie, Ferrarese, Estelle (2014) « De l’utopie numérique à la pratique : le cas de l’annotation collaborative de films », Communication & Langages, 180, 95-120.

Kaufman, Peter B., Mohan, Jen (2009). Video Use and Higher Education: Options for the Future, New-York University. https://intelligenttelevision.com/files/42-intcccnyuvideo_and_higher_edjune_2009_2.pdf

Lacković, Nataša (2018). « Analysing videos in educational research: an “Inquiry Graphics” approach for multimodal, Peircean semiotic coding of video data », Video Journal of Education and Pedagogy, 3-1. https://doi.org/10.1186/s40990-018-0018-y

Le Marec, Joëlle, Sauret, Nicolas (2016). « Archivage de répétitions et médiations du spectacle vivant. Le cas du projet spectacle en ligne(s) », Les Cahiers du numérique 12-3, 139‑164.

Maeck, Julie, Steinle, Matthias (2016). L’Image d’archives. Une image en devenir, Presses universitaires de Rennes.

Melgar-Estrada, Liliana et Koolen, Marijn (2018). « Audiovisual media annotation using qualitative data analysis software: A comparative analysis », The Qualitative Report, 23-13 : 40-60.

Ouvrard, Louise & Farge, Odile (dir.) (2022). Corpus audiovisuels : Quelles approches ? Quels usages ?, Editions des Archives Contemporaines.

Pinotti, Andrea, Somaini, Antonio, Burdet, Sophie, Aubry-Morici, Marine (2022). Culture visuelle : images, regards, médias, dispositifs, Les Presses du Réel.

Puig, Vincent, Sirven, Xavier (2007). « Lignes de temps, une plateforme collaborative pour l’annotation de films et d’objets temporels », in Proceedings IHM’07.

Pulizzotto, Davide (2019). « L’analyse de texte assistée par ordinateur : introduction à l’un des champs fondamentaux de la sémiotique computationnelle », Cygne noir, 7, 17‑41.https://doi.org/10.7202/1089328ar

Scopsi, Claire, Gouet-Brunet, Valérie, Guillaume, Louis-Pierre, Nayrolles, Lise, Battisti, Michèle (2010). « Les nouveaux territoires de la vidéo », Documentaliste-Sciences de l’Information, 47-4, 42‑53.

Sebag, Joyce & Durand, Jean-Pierre (2020). La sociologie filmique. Théories et pratiques, Paris, CNRS, coll. « Sciences politiques et sociologie ».

Sensevy, Gérard, Tiberghien, Andrée, Veillard, Laurent (2022). « VISA : un outil de mutualisation des matériaux empiriques de la recherche », in Brigitte Albero & Joris Thievenaz (dir.), Enquêter dans les métiers de l’humain, Éditions Raison et Passions, 413‑426. https://doi.org/10.3917/rp.alber.2022.03.0413

Soulez, Guillaume (2011). Quand le film nous parle : rhétorique, cinéma, télévision, PUF.

Tessier, Laurent, Bourgatte, Michaël (2017). « Les outils d’annotation vidéo pour la recherche en Humanités numériques », in Etienne Cavalié, Frédéric Clavert, Olivier Legendre & Dana Martin (dir.), Expérimenter les humanités numériques, PUM.http://www.parcoursnumeriques-pum.ca/les-outils-d-annotation-video-pour-la-recherche.

Tessier, Laurent (1999). Méthodologies d’enquête en sociologie du cinéma, Mémoire de DEA, Université Sorbonne Paris IV.

Véray, Laurent (2011). Les images d'archives face à l'histoire. De la conservation à la création, Éditions SCEREN CNDP-CRDP.

Scientific committee

Taylor ARNOLD (Richmond University)

Olivier AUBERT (LS2N – UMR 6004, Université de Nantes)

Rémy BESSON (CRIalt-Univ. Montréal)

Jean-Noël BLOCHER (CREAD EA 3875, INSPE Bretagne)

Martin BONNARD (Labdoc-UQAM)

Adrien BOURG (Institut Catholique de Paris)

Michaël BOURGATTE (EA7403-Institut Catholique de Paris)

Perrine BOUTIN (EA185-IRCAV, Paris Sorbonne Nouvelle)

Peppe CAVALLARI (EA7403-Institut Catholique de Paris)

Cécile CHANTRAINE-BRAILLON (CRHIA, La Rochelle Université)

Hélène FLECKINGER (UMR 8238-LEGS, Paris 8)

Barbara LABORDE (EA185-IRCAV, Paris Sorbonne Nouvelle)

Marie LAVOREL (CHORN-Univ. Concordia)

Arghyro PAOURI (LAP, CNRS/EHESS)

Vincent PUIG (Institut de Recherche et d’Innovation)

Claire SCOPSI (EA7339 DICEN, CNAM)

Gérard SENSEVY (CREAD EA 3875, INSPE Bretagne)

Laurent TESSIER (EA7403-Institut Catholique de Paris)

Lauren TILTON (Richmond University)

Virginie TREMION (IFP, Université Catholique de Lille)

Online user: 2

Privacy