FAU MoD Lecture: Measuring productivity and fixedness in lexico-syntactic construction

Date: Wed. November 13, 2024
Event: FAU MoD Lecture
Organized by: FAU MoD, the Research Center for Mathematics of Data at Friedrich-Alexander-Universität Erlangen-Nürnberg (Germany)

FAU MoD Lecture: Measuring productivity and fixedness in lexico-syntactic constructions
Speaker: Prof. Dr. Stephanie Evert
Affiliation: FAU MoD member/vice-spokesperson | CCL – Chair of Computational Corpus Linguistics. Friedrich-Alexander-Universität Erlangen-Nürnberg (Germany)

Abstract. In cognitive linguistics, constructions are understood as pairings of form (i.e. a lexico-grammatical pattern) and meaning (as a parameterised function if the pattern contains variable elements), which constitute the fundamental building blocks of speakers’ linguistic knowledge. Between the extremes of purely syntactic constructions (such as the ditransitive) and purely lexical ones (individual words or multiword units), a large part of constructions fall somewhere in the middle of the lexis-grammar continuum. They often consist of multiple lexical and grammatical elements, which range from completely fixed lexical items to highly variable slots.

In this talk I argue that the variability of slots in a lexico-grammatical pattern forms a cline ranging from complete fixedness to full productivity. This cline cannot be quantified by a single integrated measure, but is a combination of three distinct, but overlapping aspects:
(i) fixedness is quantified by the frequency of an element (or rather, its conditional probability given the other items in the lexico-grammatical pattern);
(ii) at the opposite end of the cline, productivity is quantified by type-token measures and interpreted with the help of statistical LNRE models;
(iii) in the middle ground between productivity and fixedness, statistical association plays a central role in identifying salient, semi-fixed lexical items.

These methodological considerations are illustrated with a case study on shell noun constructions such as “It is a fact that you will have to listen to the entire talk.”

See poster

BIO.- Prof. Dr. Stephanie Evert is a Professor at the Chair of Computational Corpus Linguistics at Friedrich-Alexander-Universität Erlangen-Nürnberg. After studying mathematics, physics and English linguistics, she received a PhD degree in computational linguistics from the University of Stuttgart, Germany. Her research interests encompass the quantitative methodology of corpus linguistics, multivariate analysis and distributional semantics, applied corpus studies and digital humanities, tools for processing large text corpora, the combination of human interpretation with machine learning (digital hermeneutics), as well as language technology and its applications.

Prof. Evert is member of the Steering Committee and Vice-spokesperson of our FAU MoD.

SLIDES

FAU MoD Lecture | Slides by Stephanie Evert

VIDEO / RECORDING

Tip: Click the play button, then use the player controls to maximize the video player as needed.

AUDIENCE

This is a hybrid event (On-site/online) open to: Public, Students, Postdocs, Professors, Faculty, Alumni and the scientific community all around the world.

WHEN

Wed. November 13, 2024 at 14:30H (Berlin time)

WHERE

On-site / Online

[On-site] Friedrich-Alexander-Universität Erlangen-Nürnberg
Room H13 Johann-Radon-Hörsaal
Cauerstraße 11, 91058 Erlangen
GPS-Koord. Raum: 49.573764N, 11.030028E

[Online] FAU Zoom link
Meeting ID: 680 1463 6900 | PIN code: 222990

This event on LinkedIn

_
* Photo by Glasow