Version 1.0
Events
Thursday 10:00
Thursday 11:00
Berlinisch: Geschichte und Grammatik
Prof. Dr. Jürg Fleischer
Hier könnt ihr die Folien der Keynote finden: https://box.hu-berlin.de/f/6339abef5fee4f98a93a/
Thursday 12:15
Why do we find it accurate?
Since early 2025, the use of DeepSeek for Bazi divination in China has been on the rise. Despite they are not objectively accurate, many still perceive the texts as accurate, which is likely related to their linguistic features. This thesis investigates why users perceive DeepSeek-generated Bazi divination texts as accurate through a genre-specific computational discourse analysis. Drawing on previous studies of horoscope and divination discourse, the research examines four dimensions of the texts: word frequency, pronoun usage, hedging expressions, and metaphorical patterns. Existing literature suggests that horoscope texts often rely on second-person pronouns, psychological vocabulary, vague or hedging language, and metaphors that encourage self-identification and interpretation.
The findings reveal that specialized Bazi terminology plays a central role in constructing an atmosphere of expertise and professionalism. While the polite second-person pronoun "您 (you)" frequently appears, users are more commonly addressed through highly genre-specific nouns such as "命主 (Life Master)" and "日主 (Day Master)", creating a semi-detached analytical perspective that both personalizes a...
Thursday 14:00
Context effects in language comprehension: When and to what extent does context play a role?
Prof. Dr. Pia Knöferle
Thursday 15:15
Conjugation class mixing in the diachrony of German
German distinguishes between two conjugation classes. While strong verbs like "lesen" 'to read' typically show a root vowel change (ablaut) in the synthetic past tense and no overt suffixal realizations of past tense and agreement for the first/th...
Stai leggendo questo testo? Stai leggendo questo test.
Prosodie spielt eine entscheidende Rolle bei der Vermittlung kommunikativer Absichten (Villani et al. 2025: 1). In den letzten Jahren wurden zunehmend Studien zur Identifikation von Intonationskurven in verschiedenen Sprachen veröffentlicht (bspw....
Petrone, Caterina & Mariapaola D’Imperio (2011): From Tones to Tunes: Effects of the f0 Prenuclear Region in the Perception of Neapolitan Statements and Questions. In Sónia Frota, Gorka Elordieta & Pilar Prieto (Hrsg.), Prosodic Categories: Production, Perception and Comprehension, 207–230. Dordrecht: Springer Netherlands. doi:10.1007/978-94-007-0137-3_9.
Petrone, Caterina & Oliver Niebuhr (2014): On the Intonation of German Intonation Questions: The Role of the Prenuclear Region. Language and Speech 57(1). 108–146. doi:10.1177/0023830913495651.
Villani, Caterina, Isabella P. Boux, Friedemann Pulvermüller & Rosario Tomasello (2025): The time course of speech act recognition conveyed by speech prosody: a gating study. Language, Cognition and Neuroscience. Routledge. 40(8). 1065–1084. doi:10.1080/23273798.2025.2506641.
The "Explain Me This" Puzzle
This study investigates why L2 learners persistently produce ungrammatical structures like “explain me this” despite conflicting input. While native speakers utilize statistical preemption to favor conventional forms, L2 learners are often hindere...
Why do L2 learners of English persistently produce structures like “Explain me this,” despite never encountering them in native input? To approach the so called Explain me this puzzle proposed by Adele Goldberg 2019, a study will be conducted to explore the mechanism of statistical preemption, a cognitive process by which speakers learn the most conventional ways to express intentions based on cumulative exposure. While native speakers use vast statistical data to intuitively reject ungrammatical structures as the double object structure for verbs like explain, L2 learners often perceive these patterns as acceptable due to native language interference (Robenalt & Goldberg 2015), such as the German “Erkläre mir das”.
The central hypothesis is that the effective use of statistical preemption is heavily dependent on working memory (Green 1998). It is argued that L2 speakers fail to benefit from statistical preemption because the high mental load required to suppress their L1 and engage in constant self-monitoring hinders their ability to make real time structural predictions. Without these expectations, the error-driven learning mechanism cannot function effectively. This study ...
Thursday 15:45
El Arte del aplicativo
Presentation of the methodology and results of my Bachelor's thesis.
Thursday 16:15
Thursday 16:45
Thursday 17:15
Friday 10:00
The anatomy of a word: lessons from language mixing
Prof. Dr. Dr. h.c. Artemis Alexiadou
Friday 11:10
On the Multifunctionality of the Turkish -lIK
This talk investigates the multifunctionality of the Turkish derivational suffix -lIK, proposing a three-way functional classification that reduces previous seven-category analyses to three core derivational functions.
While -lIK exhibits remarkable semantic diversity, deriving abstract nouns, relational adjectives, and concrete nouns, this study demonstrates that this apparent polysemy reflects three systematic morphosyntactic operations: Abstraction (producing abstract/collective nouns from adjectival or nominal bases), Attributivization (producing relational adjectives from nominal bases), and Association (producing concrete place or container nouns from nominal bases). By organizing -lIK derivatives according to morphosyntactic patterns and semantic root-derivative relations rather than referent types, this approach aims to reveal consistent structural principles underlying surface semantic diversity. The analysis further further wishes to explore the possibility that some Association uses, particularly container nouns, may involve morphemic decomposition into -lI + -K rather than monomorphemic -lIK.
With this talk I wish to present the findings of my Bachelor's thesis and to open the gate into new work concerning some open puzzles in this thesis.
Friday 11:50
Friday 14:00
Contiguity and discontinuity in nominal phrases — some lessons, implications and challenges
PD Dr. Andreas Blümel
Friday 15:10
When ‘Bad’ Means Good: Reframing Negative Lexicon in Online Communities
Virtual space is one of the domains undergoing rapid development and transformation. Given that it is characterized by a fairly free expression of emotions and creativity, within the Internet discourse certain linguistic phenomena are more emphasi...
Friday 15:40
Does your chatbot hold grudges against you?
"Climate Change is God's Punishment for the Sins of Humanity"
In the face of ecological challenges such as climate change, religions play a significant role; as religious beliefs can inspire but also discourage environmental protection, they function as value systems that may alleviate or deteriorate environmental problems. While various disciplines have already researched religions in the context of climate change, investigating discursive representations of climate change with a linguistic approach is still a relatively new endeavour. This is however a salient target of inquiry, as people shape and organise their behaviour in many areas of life through discursive practices. Considering climate discourses in the Catholic Church, scholars predominantly give one-sided attention to the environmental messages disseminated on an institutional level. It therefore remains to be investigated how Catholic practitioners respond to this messaging and how they discursively represent climate change in relation to their faith. Focusing on the Catholic Church’s environmental communication on Instagram, this interdisciplinary study explores how Catholics respond to officially issued statements on climate change in the comments, examining how these reacti...
Friday 16:15
Saturday 08:30
Saturday 11:00
Intro to Distributed Morphology
This workshop introduces the theoretical framework of Distributed Morphology (Halle & Marantz 1993; Embick 2015; Alexiadou et al., to appear) and provides participants with the opportunity to practice applying its analytical tools to concrete data...
Aber geht da nicht immer etwas verloren?
Saturday 11:30
Saturday 12:10
From optional to ... optional? Case marking and information structure in Thadou (Trans-Himalayan)
This presentation provides an insight into the interplay of core case marking and information structure in Thadou, a Trans-Himalayan language of Northeast India. At first glance, the use of the ergative marker in Thadou seems to be more (though no...
Mehr als nur Kritik: Die Übersetzungskritik in der Ausbildung von Übersetzer:innen
Saturday 14:00
Saturday 14:30
Saturday 15:20
Can AI-Generated Participants Replace Humans in Linguistic Research?
Collecting data from human participants is slow and costly. Recruitment is difficult, participants are often un- or underpaid, and results are affected by biases such as the Observer’s Paradox and the Hawthorne Effect. In addition, many sample gro...
Collecting data from human participants is a demanding and time-consuming process. Recruiting participants can be difficult, they are often un- or underpaid, and results are not always reliable due to effects such as the Observer’s Paradox and the Hawthorne. Moreover, participant pools are often WEIRD (coming from Western, educated, industrialized, rich, and democratic societies). So why not use Artificial Intelligence (AI) to simulate human participants? AI-generated data would be faster to obtain and significantly more cost-efficient than traditional data collection with humans.
This case study addresses precisely this question. To investigate it, I generated a sample of 600 AI participants (N = 600) using the LLaMA 3.1 model (8B parameters). The sample approximated the demographic distribution of Germany’s population, based on 2024 data from the Federal Statistical Office (Statistisches Bundesamt). These simulated participants completed a set of questionnaires, and their responses were then compared to data collected from human participants (N = 37).
The results generated by LLaMA 3.1 tended to be stereotypical and did not align with human-collected data, particularly i...
Saturday 16:00
What Counts as Language? Uncovering Written Bias in Language Documentation
Dr. Mandana Seyfeddinipur