The 50 most recently published documents
Um das Thema Gendern oder geschlechtergerechte Sprache hat sich eine hitzige gesellschaftliche Debatte entwickelt. Seit Anfang des Jahres ist die Diskussion um geschlechtergerechte Sprache medial wieder besonders präsent. Anlass ist u.a. die Überarbeitung der Bedeutungsbeschreibungen im Duden online. Vor kurzem widmete sogar Der Spiegel dem Thema den Hefttitel und einen Leitartikel (vgl. Bohr et al. 2021). Allerdings erschöpft sich die Diskussion leicht in Pro- und Kontra-Positionen, dabei gibt es eine ganze Bandbreite von Aspekten rund um das Thema ‚geschlechtergerechte Sprache‘ zu betrachten, die eine differenziertere Diskussion ermöglichen können. Ziel dieses Beitrags ist es, einige dieser Aspekte knapp und möglichst verständlich in die Debatte einzubringen.
Making research data publicly available for evaluation or reuse is a fundamental part of good scientific practice. However, regulations such as copyright law can prevent this practice and thereby hamper scientific progress. In Germany, text-based research disciplines have for a long time been mostly unable to publish corpora made from material outside of the public domain, effectively excluding contemporary works. While there are approaches to obfuscate text material in a way that it is no longer covered by the original copyright, many use cases still require the raw textual context for evaluation or follow-up research. Recent changes in copyright now permit text and data mining on copyrighted works. However, questions regarding reusability and sharing of such corpora at a later time are still not answered to a satisfying degree. We propose a workflow that allows interested third parties to access customized excerpts of protected corpora in accordance with current German copyright law and the soon to be implemented guidelines of the Digital Single Market directive. Our prototype is a very lightweight web interface that builds on commonly used repository software and web standards.
In this paper, we present our experiences and decisions in dealing with challenges in developing, maintaining and operating online research software tools in the field of linguistics. In particular, we highlight reproducibility, dependability, and security as important aspects of quality management – taking into account the special circumstances in which research software
is usually created.
Ungoliant: An optimized pipeline for the generation of a very large-scale multilingual web corpus
(2021)
Since the introduction of large language models in Natural Language Processing, large raw corpora have played a crucial role in Computational Linguistics. However, most of these large raw corpora are either available only for English or not available to the general public due to copyright issues. Nevertheless, there are some examples of freely available multilingual corpora for training Deep Learning NLP models, such as the OSCAR and Paracrawl corpora. However, they have quality issues, especially for low-resource languages. Moreover, recreating or updating these corpora is very complex. In this work, we try to reproduce and improve the goclassy pipeline used to create the OSCAR corpus. We propose a new pipeline that is faster, modular, parameterizable, and well documented. We use it to create a corpus similar to OSCAR but larger and based on recent data. Also, unlike OSCAR, the metadata information is at the document level. We release our pipeline under an open source license and publish the corpus under a research-only license.
Contents:
1. Julien Abadji, Pedro Javier Ortiz Suárez, Laurent Romary and Benoît Sagot: "Ungoliant: An Optimized Pipeline for the Generation of a Very Large-Scale Multilingual Web Corpus", S.1-9.
2. Markus Gärtner, Felicitas Kleinkopf, Melanie Andresen and Sibylle Hermann: "Corpus Reusability and Copyright - Challenges and Opportunities", S.10-19.
3. Nils Diewald, Eliza Margaretha and Marc Kupietz: "Lessons learned in Quality Management for Online Research Software Tools in Linguistics", S.20-26.
Kein Grund zur Panikmache
(2002)
Nachbemerkung
(1991)
The human ability to anticipate upcoming behavior not only enables smooth turn transitions but also makes early responses possible, as respondents use a variety of cues that provide for early projection of the type of action that is being performed. This article examines resources for projection in interaction in three unrelated languages—Finnish, Japanese, and Mandarin—in sequences where speakers make evaluative assertions on a topic. The focus is on independently agreeing responses initiated in early overlap. Our cross-linguistic analysis reveals that while projection based on the ongoing turn-constructional unit relies on language-specific grammatical constructions, projection based on the larger context seems to be less language-dependent. A crucial finding is that in the target sequences, stances taken toward the topic already during earlier talk, as well as other structural patterns, are among the resources that recipients use for projecting how and when the ongoing turn will end.
This article explores the relation between word order and response latency, focusing on responses to question-word questions. Qualitative (multimodal) and quantitative analyses of naturally occurring conversations in French—where question-words can occur in initial, medial, or final position within the question—show that variation in word order affects the timing of responses. It is argued that this is so because word order provides a differential basis for action ascription, creating different temporal opportunities for projecting the recipient’s next relevant action. The frequent occurrence of early responses to questions with an initial question-word, in particular, stresses the importance of the recognition point of an action under way for response timing and shows respondents’ pervasive orientation to sequential progressivity. Findings highlight how lexico-syntactic trajectories of emergent turns, prior talk and actions, material and bodily features of interaction, and participants’ shared expectations conspire in shaping the time-courses of action ascription and action projection.
This study examines head nods produced as embodied and silent answers to polar questions before a transition relevance place has been reached. It discusses the notion of “response” and the ways in which the literature conceptualizes head nods. The analysis of video recordings of ordinary and institutional multiparty interactions shows that answer-nods rely on mutual gaze and that affirmative head nods may co-occur with other facial expressions (e.g., eye blinks). By replying with a silent head nod, respondents may complete an unfolding adjacency pair without claiming speakership, thereby enabling the questioner to extend their turn-in-progress. Alternatively, respondents may expand their answer-nod with talk, in which case silent nodding may contribute to organizing the smooth transition of turns-at-talk. Head nods produced while a question is unfolding are described as a microsequential phenomenon that may affect the questioner’s turn-in-progress. Data are in French and Italian.
Focusing on request sequences, this article explores dynamics of projection and anticipation, enabling participants to produce early responses to requests. In particular, the analyses highlight the importance of multimodal formatting and the specific temporalities of multiple multimodal resources for the emergence of projections and the possibility to anticipate an ongoing action. Moreover, the analyses pinpoint the relevance of the local ecology and the praxeological context for the participants, enabling them to anticipate the next relevant action. These features characterizing the temporality of multimodal Gestalts, the relevance of the local ecology, and the details of the praxeological context make it possible for participants to produce very early responses and also to accomplish an action even before it has been actually requested.
Die traditionelle Sprachtypologie befasst sich hauptsächlich mit Standardvarietäten. Umgangssprache und Dialekte bleiben meistens außerhalb ihres Forschungsfeldes. Als extrem produktiv erweist sich ein integrierter typologisch-dialektologischer Ansatz, dessen Forschungsgegenstand die morphosyntaktischen Eigenschaften der Nicht-Standard-Varietäten sind. Die Anwendung typologischer Parameter und Hierarchien auf diese Varietäten ermöglicht eine vollständige Darstellung der wirklichen sprachlichen Lage eines Sprachgebiets, sowie eine Revision herkömmlicher typologischer Klassifikationen. Im Fall der Nicht-Standard-Relativsätze ergibt es sich, dass die Standardstrategie des Relativpronomens (der/die/das) in den deutschen Mundarten und in vielen romanischen und slavischen Nicht-Standard- Varietäten zu Gunsten anderer Strategien aufgegeben wird, die das Relativpronomen durch eine Relativpartikel bzw. die Kombination einer Relativpartikel mit einem Resumptivum ersetzen. Diese Strategien entsprechen den in den meisten Sprachen der Welt benutzten Verfahren zur Bildung von Relativsätzen: Ihr Gebrauch hängt vermutlich mit den Eigenschaften dieser Varietäten (z.B. Spontaneität, niedrige Ausbaustufe) zusammen.
Für die Analyse der gesprochenen Sprache wurden in den letzten Jahrzehnten computerisierte Sprachkorpora bereitgestellt, die qualitative und quantitative Untersuchungen erlauben. Die nächste „Herausforderung“ für die Korpuslinguistik stellt heute der Gebrauch des „World Wide Webs“ als unerschöpflicher Datenbank dar. Theoretischen Überlegungen über das Potenzial des WWWs folgt ein praktisches Beispiel: die Verwendung deutscher sprechsprachlicher Relativkonstruktionen in „Webforen“.
Der, die, das wird traditionell als Demonstrativpronomen eingestuft, obwohl es besonders in der gesprochenen Sprache zum Ausdruck der Referenz auf dritte Personen mit dem Personalpronomen er, sie, es in komplementärer Distribution auftritt. Im Beitrag wird das Verhältnis zwischen der und er zunächst auf dem Hintergrund anderer europäischer Sprachen kontrastiv untersucht. Anschließend wird die Frage der Didaktisierung aufgeworfen: an einem konkreten Beispiel wird gezeigt, wie man Deutschlernende auf den Gebrauch von der als Personalpronomen aufmerksam machen kann.
Migration wird oft sowohl mit einem Staats- als auch mit einem Sprachwechsel assoziiert. Es existieren aber auch mehrsprachige Länder, in denen eine interne Migration (Binnenwanderung), „nur“ einen Sprachwechsel erfordert. Ein Beispiel dafür ist die offiziell viersprachige Schweiz: Was passiert, wenn ein Tessiner in die Deutschschweiz oder ein deutschsprachiger Schweizer ins Tessin übersiedelt? Anhand des Begriffs vom „kommunikativen Raum“ wird im Beitrag die sprachliche Situation der schweizerischen Binnenwanderer näher beleuchtet; parallel dazu wird auf mögliche Probleme eingegangen, mit denen die Binnenwanderer zurechtkommen müssen.
This special issue investigates early responses—responsive actions that (start to) unfold while the production of the responded-to turn and action is still under way. Although timing in human conduct has gained intense interest in research, the early production of responsive actions has so far largely remained unexplored. But what makes early responses possible? What do such responses tell us about the complex interplay between syntax, prosody, and embodied conduct? And what sorts of actions do participants accomplish by means of such early responses? By addressing these questions, the special issue seeks to offer new advances in the systematic analysis of temporal organization in interaction, contributing to broader discussions in the language and cognitive sciences as to the social coordination of human conduct.
The purpose of this paper is to describe the functions of ‘where’-based relative elements' in six Balkan languages, paying particular attention to non-standard varieties.2 Relative elements based on an originally interrogative pronoun meaning ‘where’ are attested in all Balkan languages and, more generally, in all European languages. In accordance with the locative meaning of the original pronoun, ‘where’-based relative elements are primarily used to relativize locatives. However, it will be shown that in some Balkan languages, and especially in non-standard varieties, these elements have extended their functional domain. This process does not appear to be random, but rather to pattern with the following hierarchy: locative > unspecific connector > other syntactic positions (indirect/direct object, subject).3 Additionally, ‘where’-based relative elements will be compared with ‘what’-based ones in order to highlight common patterns of development.
This paper represents a report on an e-tandem project conducted at Freiburg University (Germany) from the winter term 2009/2010 on. It started with a German-ltalian pilot course organised in cooperation with Pavia University (Italy). In order to promote autonomous language learning, the authors used several web-based applications, relying on Skype to enable full (i.e. visual, auditive) interaction between learning partners and on e-mails to let participants practise writing and reading in the respective foreign language. Additionally, participants were asked to compile a weekly electronic portfolio (EPOS) to record their improvements as well as their difficulties. In the paper, the structure of the pilot course will be described and a first balance will be drawn.
Mit der rasanten Entwicklung einer ganzen Familie von XML-basierten texttechnologischen Standards seit mehr als zehn Jahren ist die Hoffnung verbunden, dass dadurch die technische Handhabung von linguistischen Daten enorm erleichtert werde und computerlinguistische Projekte davon in der Praxis stark profitierten. Dies um so mehr, als auch die meisten etablierten Datenbankmanagementsysteme (DBMS), Open Source-Lösungen sowie fachspezifische und/oder akademische Datenbankprodukte das Potential strukturorientierter Auszeichnungssprachen erkannt und Schritt für Schritt integriert haben. XMLStandards für die Modellierung der Mikrostruktur informationeller Angebote sowie für die Kommunikation mit externen Applikationen, kombiniert mit den bekannten Vorzügen einer datenbankgetützten Verwaltung und Weiterverarbeitung - aus dieser Perspektive heraus betrachtet sollte sich die computerlinguistische Forschung zukünftig vermehrt um das „Was?“ und immer weniger um das oft leidige „Wie?“ kümmern dürfen. In der Tat gibt es für viele Probleme, die vor einigen Jahren noch umfangreiche Spezialentwicklungen erforderten, inzwischen sehr hilfreiche Standard-Werkzeuge. Doch gilt dies durchgängig? Und welche Überlegungen sind vor Implementierungsentscheidungen nach wie vor unerlässlich? Hier soll der interdisziplinäre GLDV-Workshop „Datenbanktechnologien für hypermediale linguistische Anwendungen“, der im Rahmen der Konvens 2008 in Berlin stattfindet, für mehr Klarheit sorgen, indem er Erfahrungen mit hypermedialen Datenbanken aus unterschiedlichen texttechnologischen Projekten zusammenbringt. Die Vorträge und Systemdemonstrationen des Workshops liegen den Artikeln dieses Heftes zugrunde. Die Beiträge zeigen deutlich, dass nach wie vor nicht alle Werkzeuge alle naheliegenden Erwartungen erfüllen und jedes Projekt eine sorgfältige Bestandsaufnahme der verfügbaren technischen Lösungen erfordert. Die Berichte kommen aus höchst unterschiedlichen Anwendungsbereichen. Das Spektrum umfasst beispielsweise sowohl die Verwaltung unterschiedlicher Annotationsformen für geschrieben - wie für gesprochensprachliche Korpora; weiterhin werden Schnittstellen zu Online-Wörtebüchern und Wikis sowie die Verwaltung fachterminologischer Ontologien thematisiert.
Die Beiträge dieses Heftes gehen zurück auf einen Workshop des Arbeitskreises Hyper-media der Gesellschaft für Computerlinguistik und Sprachtechnologie (GSCL). Der Workshop fand im Rahmen der GSCL-Tagung 2009 in Potsdam statt und sollte den aktuellen Stand der Überlegungen zur Nutzbarkeit hypermedialer Systeme in den E-Humanities beleuchten.
Seit Jakob Nielsen Mitte der Neunzigerjahre die Kriterien für anwenderfreundliche Hypermediasysteme – Easy to learn, efficient to use, easy to remember, few errors, pleasant to use – dargelegt hat, beschäftigt sich die Usability-Forschung mit empirisch verifizierbaren Beurteilungskriterien und Erhebungsmethoden. Ziel ist die Steigerung der Nutzungsqualität hypermedialer Angebote, häufig mit den Schwerpunkten Internet/WWW bzw. Web 2.0 sowie in letzter Zeit verstärkt unter Berücksichtigung multimodaler Schnittstellen.
Die in diesem Heft zusammengestellten Beiträge beleuchten eine Reihe sehr unter-schiedlicher Aspekte von Nutzungsqualität an konkreten Anwendungen und aus theo-retischer Perspektive.
In this article, we provide longitudinal evidence for the progressive routinization of a grammatical construction used for social coordination purposes in a highly specialized activity context: task-oriented video-mediated interactions. We focus on the methodic ways in which, over the course of 4 years, a second language speaker and initially novice to such interactions coordinates the transition between interacting with her coparticipants and consulting her own screen, which suspends talk, without creating trouble due to halts in progressivity. Initially drawing on diverse resources, she increasingly resorts to the use of a prospective alert constructed around the verb to check (e.g., “I will check”), which eventually routinizes in the lexically specific form “let me check” as a highly context- and activity-bound social action format. We discuss how such change over the participant’s video-mediated interactional history contributes to our understanding of social coordination in video-mediated interaction and of participants’ recalibrating their grammar-for-interaction while adapting to new situations, languages, or media. Data are in English.
This study documents change over time and across proficiency levels in French second-language (L2) speakers’ practices for initiating complaints. Prior research has shown that speakers typically initiate complaints in a stepwise manner that indexes the contingent, moral, and delicate nature of the activity. Although elementary speakers in my data often launch complaint sequences in a straightforward way, they sometimes embodiedly foreshadow verbal expressions of negative stance or delay negative talk through brief positively valenced prefaces. More advanced speakers in part rely on the same initiation practices as elementary speakers. In addition, they recurrently use extensive prefatory work that accounts for and legitimizes the upcoming complaint, and they regularly initiate complaints jointly with coparticipants through a progressive escalation of negative stance expressions. I document interactional resources involved in this change and discuss the findings in terms of speakers’ development of L2 interactional competence. Data are in French with English translations.
We reflect on the affordances and challenges of interactional data in the analysis of long-term institutional change. To this end we draw on our studies of direct encounters between journalists and politicians in news interviews and presidential news conferences and in particular the use of question design as a window into the evolution of journalistic norms and press-state relations over time and the causal antecedents of such change. All analyses that incorporate a concern with environing contexts of interactional change impose certain burdens of empirical demonstration on the researcher. Here we consider three analytic issues that arise in the kind of historical-institutional analysis we have been pursuing: (a) controlling for the situational context, (b) pinpointing the locus of change, and (c) validating indicators of change. Data are in English.
Based on longitudinal audiovisual data from family interactions, we focus on how young children between 1;08 and 2;10 report trouble they are encountering in their current activity using the response cry oh in combination with other lexical items (e.g., “oh fell off”) and bodily displays. While at a very young age the children remain focused on their activity and try to solve the problem independently, at an older age they start to systematically use gaze directed toward the parent and suspension of the current activity to enlist the adult’s assistance. We argue that these bodily displays are among the resources whose presence or absence constrains whether the report of trouble leads to the recruitment of assistance or not. Regarding the developmental implications, it seems that during their third year of life, young children expand their repertoire for dealing with trouble interactively. Data are in German with English translations.
Semantic role labeling is traditionally viewed as a sentence-level task concerned with identifying semantic arguments that are overtly realized in a fairly local context (i.e., a clause or sentence). However, this local view potentially misses important information that can only be recovered if local argument structures are linked across sentence boundaries. One important link concerns semantic arguments that remain locally unrealized (null instantiations) but can be inferred from the context. In this paper, we report on the SemEval 2010 Task-10 on ‘‘Linking Events and Their Participants in Discourse’’, that addressed this problem. We discuss the corpus that was created for this task, which contains annotations on multiple levels: predicate argument structure (FrameNet and PropBank), null instantiations, and coreference. We also provide an analysis of the task and its difficulties.
Kontroversen wie die um gendergerechten Sprachgebrauch haben eindeutig eine politische Dimension. Das ist aber nur die eine Seite der Medaille. Jenseits der politischen Auseinandersetzung stellt sich die Frage, in welcher Weise die verschiedenen Positionen in der Gesellschaft verankert sind und warum die Kontroversen überhaupt entstehen. Die Analyse der postindustriellen Gesellschaft des Soziologen Andreas Reckwitz bietet dafür die Möglichkeit einer Erklärung.