Academic interests
- Digital humanities: distant reading/text mining of large text collections, machine learning, and generative artificial intelligence
- Literature and political and social changes: discourse analysis across fiction, non-fiction, newspapers, etc.
- Norwegian and French 19th-century literature
- Contemporary literature
- Public sphere and freedom of expression, with a focus on nation-building, gender and whistleblowing
- Power relations, including individual agency and experiences of exhaustion
- Philosophical practice
Courses taught
- https://www.uio.no/studier/emner/hf/iln/ILN4100/index.html
- https://www.uio.no/studier/emner/hf/iln/ILN2100/index.html
Background
- Master in Philosophy
- Master in French Literature
- Ph.D in Nordic Literature
Tags:
Digital Humanities,
Discourse Analysis,
Literary studies
Publications
-
-
Leclaire-Karlsen, Heidi
(2025).
Varsleren Vinje.
European Journal of Scandinavian Studies.
ISSN 2191-9399.
55(2),
p. 310–329.
doi:
10.1515/ejss-2025-2022.
Full text in Research Archive
Show summary
In 1868, Aasmund Olavsson Vinje was dismissed from his position in the Department of Justice. The minister dismissed him on the grounds of the criticism of the government Vinje had expressed in his newspaper, Dølen . Vinje also reported the dismissal conversation in his newspaper. He defended himself by referring to the right to freedom of speech. This article analyses Vinje’s texts in the “Vaar Politik” column in Dølen on February 9 and March 1, 1868. My reading aims to examine the texts as political expressions, partly in a legal-historical context, and to explore the portrayal of the criticism of the authorities and of the national in them. He viewed the fact that his critical expressions cost him his job in relation to his judgment of the Norwegian government. The article shows that the texts can be seen as early whistleblowing. Through a mix of humor – which I demonstrate is an element in the genealogy of whistleblowing – and seriousness, intertextual play, natural tropes, and wordplay, the texts appeal to the idea of a threatened national instinct and mobilize concern on behalf of the nation. This interpretation sheds new light on the poet, journalist, and also the lawyer and departmental employee Vinje, aspects which have not been previously explored in literary research.
-
Karlsen, Heidi & Alm, Kristian
(2025).
Has law and social science trivialised the concept and practice of whistleblowing in Norway 2007–2023?
In Grandis, Giovanni De & Blanchard, Anne (Ed.),
The Fragility of Responsibility : Norway’s Transformative Agenda for Research, Innovation and Business.
De Gruyter Brill.
ISSN 9783111397719.
p. 215–244.
doi:
10.1515/9783111397719-010.
Full text in Research Archive
Show summary
This chapter examines the modern evolution of whistleblowing legislation and practice in Norway and its implications for Corporate Social Responsibility (CSR). Our discourse analysis spans legal provisions, scholarly research, and national surveys and reports on whistleblowing. A finding is that the whistleblowing institution is tied to a power inter-est in encouraging employees to speak out, predominantly to prevent severe public interest damage to the population and economy. Yet, we have found that the law’s broad whistleblowing concept has led to the predominance of commonplace personal-related cases within organisations’ whistleblowing channels. Our analysis further aimed to dis-cern whether this prevailing broad understanding of whistleblowing serves the public interest or rather prioritises individual grievances at its expense. Both organisations’ whistleblowing routines and research surveys exploring the landscape of whistleblowing tend to be based on or are bound by the broad whistleblowing concept in the legislation. Consequently, we have limited tools for assessing whetherwhistleblowing serves the public interest and is an efficient mechanism for detecting when CSR fails. We examine recent research indicating an increase in retaliation against whistleblowers. However, the broad whistleblowing concept makes it difficult to discern whether this increase targets public interest reporting, individual grievance reporting, or both. Consequently, our ability to assess the effectiveness of whistleblowing to restore CSR is impaired. The Norwegian Supreme Court recently interpreted the whistleblowing provisions, solidifying a low threshold for what constitutes whistleblowing. A dissenting minority highlighted the provisions’ intent: whistleblowing in the public interest. Against this background, we conclude by providing suggestions for improving the whistleblowing institute and its potential as a CSR mechanism.
-
Larsen, Eirinn & Karlsen, Heidi
(2025).
Den digitale vending som mulighet og utfordring.
Historisk tidsskrift.
ISSN 0018-263X.
104(2),
p. 183–189.
doi:
10.18261/ht.104.2.7.
Full text in Research Archive
Show summary
Siden 2022 har Universitetet i Oslo (UiO) gitt historiestudenter og studenter fra andre humanistiske fag tilbud om å lære å ta i bruk digitale forskningsmetoder i arbeidet med tekster. Under merkelappen «digital humaniora» og med litteraturviter Heidi Leclaire-Karlsen som emneansvarlig har til sammen om lag 50 studenter fulgt kurset. Ikke mange, vil du som leser dette, kanskje tenke, og kun en femtedel av disse var historiestudenter. Samtidig hadde ingen av dem, hverken historiestudentene eller de andre, noe særlig problem med å reflektere rundt begrepet «digital humaniora». Men svarene de ga, overrasket likevel oss undervisere, for mens det digitale var noe de hadde mye å si om, viste humaniora seg å være mer ukjent. I tillegg møtte historiestudentene utfordringer når de ble satt til å reflektere over hvilke metoder og epistemologiske tradisjoner historikerne bruker og virker innenfor, og hvordan digital teknologi og maskinlæring skulle kunne bidra med ny kunnskap og nye perspektiver på fortiden.
Den digitale vendingen gir oss som fagpersoner i historieopplæringen både nye muligheter og utfordringer. Vi har med en studentgenerasjon å gjøre som er født inn i det digitale og derfor behersker sider ved vår tids teknologi som vi selv ikke alltid gjør. At nåtidens historiestudenter synes det er merkelig å bli presentert for digital historie som noe eget, eller som forgreninger av det som en stund er blitt omtalt som digital humaniora, er derfor kanskje ikke så merkelig. I dag foregår historisk forskning og oppgaveskriving, som studentene er mest opptatt av, med utstrakt bruk av digitale søkemotorer, algoritmer, tekst og arkivressurser på nett, skriveprogrammer og KI. Likevel opplever vi at det i historikerfagmiljøet har vært lite felles diskusjon omkring det digitale skiftet, ikke minst opp mot utdanningsoppdraget vårt. Hvordan utdanner vi dagens og morgendagens historiestudenter når vilkårene for kunnskapsproduksjon har endret seg så radikalt?
Denne artikkelen er et forsøk på å innlede til debatt og kunnskapsdeling om det digitale skiftet i historiefaget på norsk ved å dele våre erfaringer fra det ovennevnte kurset i digital humaniora og den øvrige teori- og metodeundervisningen for historiestudenter ved UiO
-
Karlsen, Heidi
(2022).
Foucault’s archeological discourse analysis with digital methodology—Discourse on women prior to the first wave women’s movement.
Digital Scholarship in the Humanities.
ISSN 2055-7671.
38(1),
p. 195–208.
doi:
10.1093/llc/fqac022.
Full text in Research Archive
Show summary
In L’archéologie du savoir (1969), Foucault specifies the method of discourse analysis as the identification of statements. Few digital humanities projects have set out to conduct discourse analysis in this tradition. A certain degree of confusion relates to what the Foucauldian statement consists of and how it can be operationalized in order to identify discourse, using digital methods. This article demonstrates, however, that such an operationalization becomes possible if we take Foucault’s own distinction between statement (énoncé) and enunciation (énonciation) into account. As an example, the article shows how discourse on women prior to the first wave women’s movement in Western countries has been identified in almost 7,000 books digitized by the Norwegian National Library. This article concludes that a digitized Foucauldian discourse analysis is possible, using a combination of digital methodology and close reading.
-
Karlsen, Heidi
(2021).
Kvinner i Nasjonalbibliotekets digitaliserte samlinger i perioden 1830–1880: etableringen av en minoriserende diskurs,
Minoritetsdiskurser i norsk litteratur.
Universitetsforlaget.
ISSN 9788215045320.
p. 93–113.
doi:
10.18261/9788215045320-2021-05.
Full text in Research Archive
Show summary
En analyse av ca. 7000 bøker i Nasjonalbibliotekets digitaliserte samlinger fra 1830 til 1880 vitner om at spørsmålet om hva en kvinne er og bør være, i økende grad tematiseres på tvers av sjangre og i både norske og oversatte bøker. Kvinnespørsmålet og fremveksten av en kvinnelighet som kan ta plass i den offentlige debatten, tiltar parallelt og henger sammen med at kvinnens «plass» i det private hjem sementeres, noe som utfordrer oppfatningen om at kvinnefrigjøringen først kunne komme med et brudd med det religiøse og det borgerlige kvinneidealet de første tre fjerdedelene av 1800-tallet.
-
Karlsen, Heidi
(2019).
Sub-corpus topic modeling og diskursanalyse: gruvearbeid, tråling eller kokekunst?
Samlaren: tidsskrift för forskning om svensk och annan nordisk litteratur.
ISSN 0348-6133.
140,
p. 281–304.
Full text in Research Archive
Show summary
This article presents the data-mining technique sub-corpus topic modeling and argues that it is well suited for capturing data that can be used for discourse analysis. Epistemological implications of digital methods and the related concept of ”data-mining” in the humanities are discussed based on Emely Apter’s 2017 essay ”Overburden.” ”Mining” as a metaphor may indicate that knowledge is regarded as something that is extracted and refined, as metals are from mines. In this case, data-mining may appear incompatible with the ”surface reading” on which discourse analysis depends. This article argues, however, that sub-corpus topic modeling, with some adjustments, particularly preserves the ”surface” of the research objects, which is demonstrated by means of a detailed, but accessible presentation of the method. Sub-corpus topic modeling uses themes modeled from a ”sub-corpus” to search for the same themes in a larger corpus. With the Norwegian National Library’s ”Bokhylla” (the ”Digital Bookshelf ”), a large number of texts are available digitally. As an example of how sub-corpus topic modeling can be used to capture data for discourse analysis, the Digital Bookshelf constitutes the larger corpus, while texts by the Swedish author Fredrika Bremer constitute the sub-corpora.
View all works in NVA
-
Karlsen, Heidi & Johnsen, Lars G.
(2022).
A Digital Discourse Analysis of the Norwegian National Library’s Collections - The Idea of a Feminine Essence in the First Half of the Twentieth Century.
Full text in Research Archive
-
-
-
-
Karlsen, Heidi
(2019).
Oppdag meningsfylte avsnitt med Sub-corpus topic modeling - "Tråling" eller kokekunst"? - med Bokhylla.
Full text in Research Archive
-
Karlsen, Heidi
(2019).
Women’s place in Norwegian Society 1830- 1880 - Discourse Analysis with the Digital Bookshelf.
Full text in Research Archive
-
-
-
Karlsen, Heidi
(2019).
Le mouvement des femmes en Norvège au XIXème siècle avait-il besoin de George Sand pour sortir de son silence ?
DESHIMA, revue d'histoire globale des pays du Nord.
ISSN 1957-5173.
13,
p. 195–211.
Full text in Research Archive
-
-
-
Karlsen, Heidi
(2018).
Interdisciplinary Advancement through the Unexpected: Mapping Gender Discourses in Norway (1840-1913) with Bokhylla.
Full text in Research Archive
-
-
Karlsen, Heidi
(2018).
The Discursive Context of Nora’s Slamming Door: Gender Discourses in Norway in the Second Half of the Nineteenth Century and Nora’s Decision to Leave Home.
Full text in Research Archive
-
Karlsen, Heidi
(2018).
The Stage as an Intersection across Discursive and Non-Discursive Gender Practices in Norway in the Second Half of the Nineteenth Century.
Full text in Research Archive
-
Karlsen, Heidi
(2018).
"Data-mining the Digital Bookshelf" Presentasjon av metodologiske muligheter og utfordringer knyttet til litteraturhistorisk arbeid med NBs digitaliserte samling.
Full text in Research Archive
-
-
-
Karlsen, Heidi
(2020).
A Discourse Analysis of Woman’s Place in Society 1830-1880 through Data-Mining the Digital Bookshelf.
Universitetet i Oslo.
Full text in Research Archive
Show summary
My main question is how woman's place in society is established and negotiated by the fact that one
speaks about woman, and that a growing number of women themselves get access to discourse and
publish texts that thematize gender and women. Furthermore, I have explored how digital
methodology can be used to generate material for such an analysis. My main finding is that these
negotiations concern what virtues women were to cultivate and live in accordance with, what kind
of education she needed for this, and what degree of freedom woman should be granted. Through
these negotiations woman's place in society is established as a function and calling she has — in
virtue of her gender and its sensitive nature— to enhance the moral condition of the population.
Most of my captured material consists of religious texts, including devotional house books with
which it is likely that many women were familiar. Several of these address woman directly. Woman
is encouraged to realize what are the virtues of her gender and to live by them. In many cases,
woman's subordination in marriage is emphasized as a virtue, while also arguing that woman has
and should have freedom. It is a given that woman has to have awareness of her specific function
and that this awareness must be shaped and strengthened due to the complexity of her submitted
tasks. Even the most restrictive texts on women's freedom are part of a discursive practice that is
susceptible to negotiating woman's right to freedom, as long as one argues that it strengthens her in
her gender-specific function of promoting morality in society. Women's discourse production
unfolds along two main tracks, both of which are characterized by zooming close in on woman's
inner life. First, it negotiates extended freedom for women, mainly because they can thus more fully
fulfill their calling, and second, it seeks to influence women to be convinced that cultivating and
living by feminine virtues is morally good.
I have mainly used the data mining technique sub-corpus topic modeling (STM) and a “word bag”
tool in order to identify text passages in the corpus that speak about woman. Briefly put, STM
works the following way: “topics” — that is in this context “word clusters” — are modeled in
selected texts. These topics are then applied to a larger corpus to search for texts that contains the
same topics. In this project, selected texts primarily by women writers form my sub-corpora. Topics
modeled in these texts are then applied to the Digital Bookshelf corpus 1830-1880. In line with
Heidi Karlsen
Michel Foucualt's method for discourse analysis, I operate with a distinction between enunciations
(énunciations) and statements (énoncés). I consider my results as concrete enunciations about
gender and women, while the analytical work consists of identifying the statements, that is, the
regularity in terms of what is said, how it is said, in what contexts one speaks, and what it takes to
take a position as a discourse producer. This project contends that STM in combination with a word
bag tool is apt for conducting an archival discourse analysis in Foucault’s tradition.
View all works in NVA
Published
Jan. 25, 2022 3:09 PM
- Last modified
Mar. 10, 2025 2:55 PM
