Shaping stories: the process characteristics of literary revisions

Floor Buschenhenke

Samenvatting

Revision is a key aspect of professional literary writing. (Sullivan, 2013) Using the
keystroke logger Inputlog (Leijten & Van Waes, 2013), eleven Dutch and Flemish
authors recorded their processes for our study into born-digital literary writing. To
discover how they use revision and what they revise during the composition of short
stories, the revisions from three of these processes were manually classified using a
genre-specific taxonomy. Existing taxonomies generally follow Faigley & Witte's
(1985) broad distinction between meaning-changing, meaning-preserving and
surface revisions. Extensions have been created to subdivide these categories.
These extensions are usually genre-specific (e.g. Yang et al, 2017 for Wikipedia
edits, or Conijn & Schultz, 2020 for academic essays). Existing taxonomies did not,
however, allow for a subdivision within meaning-changing revisions that are relevant
for literary narrative texts. This is why I opted for a new semantic categorisation
which acknowledges that both stylistic and narrative changes can affect meaning.
For this collection of nearly 5000 revisions, the process characteristics were also
registered. In specific, I modelled the potential connections between preceding
pause time, linguistic level, the revisions' locations (contextuality) as well as their
timing within the multi-session processes. Significant correlations were found
between process characteristics and the semantic labels of the revisions. These
indicate that cognitive effort is higher for both stylistic and narrative revisions when
compared to surface revisions. Different linguistic units were affected for the different
categories of revisions. At pre-contextual locations (within incomplete sentences, see
Lindgren et al, 2019) writers focussed on different aspects of their stories then at
contextual locations. Also, the writers had unique revision profiles, in particular with
regards to their pause behaviour, but also with regards to the amount and the
proportion of revisions with a semantic impact. Over time, they focussed more and
more on stylistic aspects rather than narrative changes.

Originele taal-2	Engels
Pagina's	52-53
Aantal pagina's	1
Status	Gepubliceerd - 26 jun. 2024
Evenement	Sig Writing 2024: ways2write - University of Nanterre Paris, Paris, Frankrijk Duur: 26 jun. 2024 → 28 jun. 2024 Congresnummer: 20 https://www.earli.org/sig-writing-2024-conference-programme

Conferentie

Conferentie	Sig Writing 2024
Land/Regio	Frankrijk
Stad	Paris
Periode	26/06/2024 → 28/06/2024
Internet adres	https://www.earli.org/sig-writing-2024-conference-programme

Toegang tot document

https://drive.google.com/file/d/1Y9TiUEH-0KX9TGgcsxp4TWmXRbQHPhX_/viewLicentie: Unspecified

Document onder embargo

Shaping stories
Final published version, 148 KB
Kopie aanvragen

Citeer dit

@conference{2f05a5a0b32242b7bab2a3a0df606b0f,

title = "Shaping stories: the process characteristics of literary revisions",

abstract = "Revision is a key aspect of professional literary writing. (Sullivan, 2013) Using thekeystroke logger Inputlog (Leijten & Van Waes, 2013), eleven Dutch and Flemishauthors recorded their processes for our study into born-digital literary writing. Todiscover how they use revision and what they revise during the composition of shortstories, the revisions from three of these processes were manually classified using agenre-specific taxonomy. Existing taxonomies generally follow Faigley & Witte's(1985) broad distinction between meaning-changing, meaning-preserving andsurface revisions. Extensions have been created to subdivide these categories.These extensions are usually genre-specific (e.g. Yang et al, 2017 for Wikipediaedits, or Conijn & Schultz, 2020 for academic essays). Existing taxonomies did not,however, allow for a subdivision within meaning-changing revisions that are relevantfor literary narrative texts. This is why I opted for a new semantic categorisationwhich acknowledges that both stylistic and narrative changes can affect meaning.For this collection of nearly 5000 revisions, the process characteristics were alsoregistered. In specific, I modelled the potential connections between precedingpause time, linguistic level, the revisions' locations (contextuality) as well as theirtiming within the multi-session processes. Significant correlations were foundbetween process characteristics and the semantic labels of the revisions. Theseindicate that cognitive effort is higher for both stylistic and narrative revisions whencompared to surface revisions. Different linguistic units were affected for the differentcategories of revisions. At pre-contextual locations (within incomplete sentences, seeLindgren et al, 2019) writers focussed on different aspects of their stories then atcontextual locations. Also, the writers had unique revision profiles, in particular withregards to their pause behaviour, but also with regards to the amount and theproportion of revisions with a semantic impact. Over time, they focussed more andmore on stylistic aspects rather than narrative changes.",

author = "Floor Buschenhenke",

year = "2024",

month = jun,

day = "26",

language = "English",

pages = "52--53",

note = "Sig Writing 2024 : ways2write ; Conference date: 26-06-2024 Through 28-06-2024",

url = "https://www.earli.org/sig-writing-2024-conference-programme",

}

TY - CONF

T1 - Shaping stories: the process characteristics of literary revisions

AU - Buschenhenke, Floor

N1 - Conference code: 20

PY - 2024/6/26

Y1 - 2024/6/26

N2 - Revision is a key aspect of professional literary writing. (Sullivan, 2013) Using thekeystroke logger Inputlog (Leijten & Van Waes, 2013), eleven Dutch and Flemishauthors recorded their processes for our study into born-digital literary writing. Todiscover how they use revision and what they revise during the composition of shortstories, the revisions from three of these processes were manually classified using agenre-specific taxonomy. Existing taxonomies generally follow Faigley & Witte's(1985) broad distinction between meaning-changing, meaning-preserving andsurface revisions. Extensions have been created to subdivide these categories.These extensions are usually genre-specific (e.g. Yang et al, 2017 for Wikipediaedits, or Conijn & Schultz, 2020 for academic essays). Existing taxonomies did not,however, allow for a subdivision within meaning-changing revisions that are relevantfor literary narrative texts. This is why I opted for a new semantic categorisationwhich acknowledges that both stylistic and narrative changes can affect meaning.For this collection of nearly 5000 revisions, the process characteristics were alsoregistered. In specific, I modelled the potential connections between precedingpause time, linguistic level, the revisions' locations (contextuality) as well as theirtiming within the multi-session processes. Significant correlations were foundbetween process characteristics and the semantic labels of the revisions. Theseindicate that cognitive effort is higher for both stylistic and narrative revisions whencompared to surface revisions. Different linguistic units were affected for the differentcategories of revisions. At pre-contextual locations (within incomplete sentences, seeLindgren et al, 2019) writers focussed on different aspects of their stories then atcontextual locations. Also, the writers had unique revision profiles, in particular withregards to their pause behaviour, but also with regards to the amount and theproportion of revisions with a semantic impact. Over time, they focussed more andmore on stylistic aspects rather than narrative changes.

AB - Revision is a key aspect of professional literary writing. (Sullivan, 2013) Using thekeystroke logger Inputlog (Leijten & Van Waes, 2013), eleven Dutch and Flemishauthors recorded their processes for our study into born-digital literary writing. Todiscover how they use revision and what they revise during the composition of shortstories, the revisions from three of these processes were manually classified using agenre-specific taxonomy. Existing taxonomies generally follow Faigley & Witte's(1985) broad distinction between meaning-changing, meaning-preserving andsurface revisions. Extensions have been created to subdivide these categories.These extensions are usually genre-specific (e.g. Yang et al, 2017 for Wikipediaedits, or Conijn & Schultz, 2020 for academic essays). Existing taxonomies did not,however, allow for a subdivision within meaning-changing revisions that are relevantfor literary narrative texts. This is why I opted for a new semantic categorisationwhich acknowledges that both stylistic and narrative changes can affect meaning.For this collection of nearly 5000 revisions, the process characteristics were alsoregistered. In specific, I modelled the potential connections between precedingpause time, linguistic level, the revisions' locations (contextuality) as well as theirtiming within the multi-session processes. Significant correlations were foundbetween process characteristics and the semantic labels of the revisions. Theseindicate that cognitive effort is higher for both stylistic and narrative revisions whencompared to surface revisions. Different linguistic units were affected for the differentcategories of revisions. At pre-contextual locations (within incomplete sentences, seeLindgren et al, 2019) writers focussed on different aspects of their stories then atcontextual locations. Also, the writers had unique revision profiles, in particular withregards to their pause behaviour, but also with regards to the amount and theproportion of revisions with a semantic impact. Over time, they focussed more andmore on stylistic aspects rather than narrative changes.

M3 - Abstract

SP - 52

EP - 53

T2 - Sig Writing 2024

Y2 - 26 June 2024 through 28 June 2024