Research master thesis | Linguistics (research) (MA)
open access
This thesis investigates the distribution and syntactic structures of deverbal nominalizations in English, Hungarian, and Serbian. Deverbal nominalizations are distinct from regular nouns in that...Show moreThis thesis investigates the distribution and syntactic structures of deverbal nominalizations in English, Hungarian, and Serbian. Deverbal nominalizations are distinct from regular nouns in that they are derived from verbs and thus contain both nominal and verbal properties. The significant work by Grimshaw (1990) established specific diagnostics advocating a strict dichotomy between event and result nominals: event nominals emphasize the action denoted by the verb, and result nominals focus on the outcome of the event denoted by the verb. However, ample cross-linguistic evidence shows that the distinction between these two types of deverbal nominals is less clear-cut than thought before, and a lot of this variation is found in forming the plural of these nominals. There is also no uniform syntactic structure for these kinds of nominals. This means that all the existing syntactic proposals in the literature deviate from one another in multiple ways. Therefore, the primary objective of this study was to bridge the gaps in the literature on the usage of deverbal nouns and their syntactic structures. This thesis presents acceptability judgment data on deverbal nouns that occurred in the singular and plural alongside an additional diagnostic by Grimshaw (1990): type of reading, adverbial modification, argument structure, and aspectual modification. The results indicate that Hungarian and Serbian allow plural event nominals in all contexts, while English only allows them without any modifiers. These findings are contra Grimshaw (1990) and underscore the crucial role of contextual factors in (dis)allowing plurality with event nominals. In addition to its empirical contributions, this thesis evaluates the applicability and generalizability of Grimshaw’s (1990) theorem by analyzing languages from different language families, thereby enriching the understanding of deverbal nominalizations across diverse linguistic contexts.Show less
Research master thesis | Linguistics (research) (MA)
under embargo until 2026-07-31
2026-07-31T00:00:00Z
Adjectival modification in the nominal domain is one of the broadly discussed issues in the field. In the current thesis I build on the novel observation that apart from those cases where the...Show moreAdjectival modification in the nominal domain is one of the broadly discussed issues in the field. In the current thesis I build on the novel observation that apart from those cases where the adjectival modifier is adjacent to the noun, languages such as Dutch also allow to have an overt mediating element in between the noun and the modifier resulting in ‘een groene kleur auto’ (‘a green colour car’) (cf., Kayne 2005). I call this element a dimension classifier and argue that it heads DimensionP. I treat adjectival modifiers as phrases (cf., Cinque 2010) generated in the specifier of it. Based on the syntactic distribution of different classes of modifiers in similar constructions in Dutch and Russian and their interpretational properties I argue for three possible ways to merge the modifier in the structure. In two of these scenarios DimensionP is itself embedded in a complex structure leaving room for a functional preposition ‘of’ and creating possessive relation with the noun. Based on the feature specification of the dimension classifier I derive two possible interpretations out of this structure. The third scenario does not assume a possessive relation between the noun and the modifier and is achieved due to DimensionP only. I show that this model is able to capture the distributional properties of modification with and without an overt dimension classifier in attributive and predicative positions as well as in case of modifying an indefinite pronoun. I hold that every modifier should be introduced by a dimension classifier that can be realised as a free-standing, bound or zero morpheme. I show all three of these possibilities for both Dutch and Russian based on the discussion of modifiers of material, origin, colour, shape, size and subjective comment.Show less
Research master thesis | Linguistics (research) (MA)
open access
Harmful speech has evolved into more strategic mechanisms enabling speakers to implicitly spread controversial ideas without facing backlash. One such mechanism is through so-called dogwhistles....Show moreHarmful speech has evolved into more strategic mechanisms enabling speakers to implicitly spread controversial ideas without facing backlash. One such mechanism is through so-called dogwhistles. While there has been growing awareness of the potentially harmful consequences of dogwhistles, it is still unclear what dogwhistles exactly are and how they work. In order to combat the pernicious power of dogwhistles, we need to have a proper understanding of what we are dealing with first. Whereas previous work has mainly focused on the expressions used by speakers as dogwhistles, I propose a new speech-act theoretic account of dogwhistles, shifting the focus to the actions performed by speakers when using such expressions. This thesis sets out to formulate a theoretically sound conceptualization of dogwhistles, as well as account for the processes through which dogwhistles convey meaning. As will become clear, however, dogwhistles are a tricky phenomenon that cannot be adequately accounted for from a speaker-centric perspective on meaning. In this thesis, I will therefore introduce a new class of appellative speech acts, whose meanings are co-contributed by the hearer. Using this theoretical framework, I define dogwhistles as an appellative speech act with the illocutionary force of simultaneously appealing to multiple ideologically diverse audiences in their own right.Show less
Research master thesis | Linguistics (research) (MA)
under embargo until 2026-01-31
2026-01-31T00:00:00Z
Like other Bantu languages, Kîîtharaka (Guthrie E54, Central Kenya Bantu) allows indexing the object on the verb by inserting an object prefix in front of the verb stem. Object marking is...Show moreLike other Bantu languages, Kîîtharaka (Guthrie E54, Central Kenya Bantu) allows indexing the object on the verb by inserting an object prefix in front of the verb stem. Object marking is prohibited in case of an overt object. However, if the object is dislocated, this marker sometimes occurs obligatorily, while in other cases, it is prohibited. Hence, Kîîtharaka poses a classical case for differential object marking (DOM). The main outcome of this thesis is the finding that a complex interplay of a variety of factors is responsible for differential object marking in Kîîth araka. These factors include cross-linguistically well-known triggers for DOM such as animacy and accessibility, but also lesser-known factors such as predicate class and verum. In addition to proposing a hierarchy according to which the established factors apply, individuation is given as a possible theoretical explanation to account for the diversity of factors found.Show less
Research master thesis | Linguistics (research) (MA)
closed access
There are large individual differences among late second language learners. The present study investigated whether individual differences in late L2 acquisition could be explained by individual...Show moreThere are large individual differences among late second language learners. The present study investigated whether individual differences in late L2 acquisition could be explained by individual differences in musical experience, because a large number of prior studies showed that musicians have enhanced auditory processing abilities. Dutch late-learners of English performed an AXB identification task, a 4I2AFC discrimination task and a passive oddball task with the English vowel contrast DRESS-TRAP. Data was analyzed with a multilevel single-trial approach to focus on the explanation of individual differences. The results showed large individual differences in all three tasks, good behavioral discrimination of the vowel contrast, and a large mismatch negativity (MMN) effect. A relation between discrimination performance and the variables musical experience and MMN amplitude was found, indicating that better behavioral discrimination of small acoustic differences was related to more musical experience and larger MMN differences. The absence of a similar relation with identification performance was attributed to a ceiling effect and restricted range of results on this task. The fact that no significant relation was found between musical experience and MMN amplitude, was explained by the required discrimination difficulty to measure a positive effect of musical experience, which was not met by the stimuli in the oddball task.Show less
Research master thesis | Linguistics (research) (MA)
open access
This thesis investigates a recurring issue found in fieldwork on Niger-Congo and Nilo-Saharan languages: how can a linguist tell if a given language has the vowels [ɪ ʊ], or [e o], or both? These...Show moreThis thesis investigates a recurring issue found in fieldwork on Niger-Congo and Nilo-Saharan languages: how can a linguist tell if a given language has the vowels [ɪ ʊ], or [e o], or both? These two pairs of vowels – which differ in their height and ATR (‘advanced tongue root’) value – are notoriously difficult to tell apart acoustically, and their pronunciation varies considerably between languages and even speakers. Moreover, previous studies have only focused on how to distinguish these vowels in languages with both pairs; in languages with just one pair, how do we know if that pair is [ɪ ʊ] or [e o]? To address this gap, the main acoustic study of this thesis looks at data from Boa-Leboale, a Bantu language spoken in the Democratic Republic of the Congo that has historically been described as only having [ɪ ʊ], not [e o]. Using the results and acoustic methodology of Starwalt’s (2008) landmark study, I construct a number of hypotheses to test two research questions; research question (I) investigates whether Boa-Leboale has one or two pairs of vowels and research question (II) attempts to identify whether a given set of vowels is more likely to be [ɪ ʊ] or [e o]. The results of this study indicate that Boa-Leboale does indeed have only one pair of vowels, and those vowels are most likely [ɪ ʊ], as suggested by previous descriptions. Given this promising result, I hope that further studies will be able to improve on the study presented here and solve the issue of identifying these vowels acoustically not just for Boa-Leboale, but for any language.Show less
Research master thesis | Linguistics (research) (MA)
under embargo until 2025-07-31
2025-07-31T00:00:00Z
While multiple models of language production exist, it remains difficult to find conclusive evidence for one or the other model. The present study is one of the first to incorporate both behavioral...Show moreWhile multiple models of language production exist, it remains difficult to find conclusive evidence for one or the other model. The present study is one of the first to incorporate both behavioral data and EEG data in a native Dutch noun phrase production task to investigate the processing of grammatical gender. We used a picture-word interference paradigm in which speakers had to name a picture while being distracted by a superimposed word. The distractors were manipulated for grammatical gender congruency with and semantic relatedness to the picture names. In separate blocks, participants had to produce either bare nouns, or determiner noun constructions. We found a robust effect of the semantic relatedness effect in the behavioral data, in which targets belonging to the same semantic category as the distractor were named significantly later than unrelated targets. The EEG data on the semantic relatedness effect were less clear, possibly due to opposing ERP components in (partly) overlapping time windows. More importantly, we replicated the gender congruency effect in reaction times, showing that naming latencies were longer for gender-incongruent trials than for congruent trials. This behavioral effect disappeared in bare noun naming. The EEG data showed similar results, in which a more positive peak was found between 385-585 milliseconds for incongruent than for congruent trials, again only in the determiner noun block. The result seems to indicate a P300 effect rather than the expected enhanced N400, suggesting higher conflict resolution processes for incongruent trials. The absence of the gender effect in bare noun naming in both data types suggests that grammatical gender is not automatically accessed or selected in Dutch noun (phrase) production. This in turn has implications for speech production models.Show less
Research master thesis | Linguistics (research) (MA)
open access
2024-01-31T00:00:00Z
The existence of a bilingual advantage in cognitive processing is a popular research topic and is heavily debated. A seminal study by Kovács and Mehler (2009) provided evidence that there is a...Show moreThe existence of a bilingual advantage in cognitive processing is a popular research topic and is heavily debated. A seminal study by Kovács and Mehler (2009) provided evidence that there is a bilingual advantage in 7-month-old pre-verbal infants in a switching task, although replication findings are inconsistent (Dal Ben et al., 2022; D’Souza et al., 2020; Kalashnikova et al., 2021; Spit et al., 2023). Measuring the pupil dilation response (PDR), a physiological measurement linked to the locus coeruleus (LC) and as such, to cognitive processing load, could give us a more direct look into whether 7-month-old bilingual infants indeed have a cognitive advantage compared to monolinguals. For this pilot study, additional pupil size measurements were taken in the Leiden arm of the replication effort of Spit et al. (2023) to examine whether bilinguals have a smaller PDR from baseline. After hearing a syllable pattern (AAB or ABB) the infants had to predict on which side a visual reward appeared. The reward side would be the same for the first nine trials in the pre-switch block. The next nine trials had the other syllable pattern and the reward on the other side in the post-switch block. Finally, the last 18 trials mixed both syllable patterns, retaining their associated reward side. The results suggest no difference in cognitive load between the monolingual and bilingual groups when they needed to relearn to predict the target reward side in the post-switch block, nor was there a difference in mean PDR in the association block. This is in line with the results found in Spit et al. (2023) where anticipatory looking behaviour was examined. However, an exploratory analysis suggested there was a significantly larger PDR in monolinguals during stimulus presentation in the pre-switch block compared to the post-switch block, indicating monolinguals had a higher processing load in the first block of the experiment. The implications of this are unclear, but might be explained by an effect related to the unfamiliarity of the task stimuli seen in monolinguals only due to different attentional strategies between the groups. Future research should be done with larger sample sizes and more sophisticated statistical modelling.Show less
Research master thesis | Linguistics (research) (MA)
under embargo until 2025-06-30
2025-06-30T00:00:00Z
This study investigates the development of the following lateral fricatives and affricates (obstruents) in Nguni (S40) languages, spoken in Southern Africa. These lateral obstruents (/ɬ, ⁿɬ, ɮ, ⁿɮ,...Show moreThis study investigates the development of the following lateral fricatives and affricates (obstruents) in Nguni (S40) languages, spoken in Southern Africa. These lateral obstruents (/ɬ, ⁿɬ, ɮ, ⁿɮ, k͡ʟ ̝̊/) are rare in the Bantu language family, but occur in three subbranches of Southern Bantu: Sotho Tswana, Nguni, and Tsonga. Given the rarity of these sounds, in Bantu and crosslinguistically, the question arises how Southern Bantu languages have incorporated lateral obstruents in their phonologies, as they are not reconstructed for Proto Bantu. To answer this question, I analyze secondary data from 10 Nguni languages to study which lateral obstruents are used and how, leading to the conclusion that both inheritance and contact played a role in the development of lateral obstruents in Nguni languages. The data shows that alveolar lateral obstruents can be reconstructed to Proto Nguni and derive from a regular sound change from Proto Bantu palatals /t͡ʃ/ and /d͡ʒ/. The velar lateral affricate finds its origin in loan words. I evaluate existing theories that proposed several contact scenarios, which at this point are not plausible explanations for the development of Nguni lateral obstruents. The findings from this study make several contributions to the field. It is the only empirical study to date to collect and reinterpret data from a large number of secondary sources, leading to a better understanding of not only the distribution of lateral obstruents in Nguni languages, but also the hypothesized ancestor of these related languages, Proto Nguni. Further, this thesis lays the groundwork for future research into lateral obstruents in Southern Bantu. If we expand our focus from Nguni to the other subclades that feature lateral obstruents, we can combine that knowledge of lateral obstruents in Southern Bantu with other innovations and patterns to gain insight into the diversification of Southern Bantu languages and understand how the subgroups relate to each otherShow less
Research master thesis | Linguistics (research) (MA)
open access
Eye-tracking reading on bilinguals has found divergent results: some have found disadvantages for bilinguals of varying proficiency as compared to monolinguals, where language proficiency scores...Show moreEye-tracking reading on bilinguals has found divergent results: some have found disadvantages for bilinguals of varying proficiency as compared to monolinguals, where language proficiency scores and individual differences in cognitive control ability accounted for these differences. Others reported a bilingual advantage in cognitive control which also affected syntactic parsing beneficially as bilinguals scored higher on comprehension whilst processing garden-path sentences. However, bilingualism itself is often poorly defined, which can lead to unfair comparisons between, potentially, extremely different types of bilinguals. Therefore, we employ a strict definition of bilingualism, as well as clearly defining what language-pairing our bilinguals have and what the potential language interaction effects of the pairing could be. In this fashion, the current study assesses whether high-proficiency Dutch-English bilinguals show an advantage on sentence comprehension of garden-path sentences and whether lingering misinterpretations related to garden-path effects in Good-Enough parsing theories remain and follow the expected patterns. We employed eye-tracking with N = 20 Dutch-English bilinguals and N = 12 native English speakers, and compared their reading times and comprehension accuracy. Our results confirm the patterns suggested in recent adaptations made to Good-Enough parsing models, in which information structure and prediction are incorporated and help guide the parsing process. Additionally, we find evidence of a specific Dutch-English language interaction which surfaces as an advantage for the bilinguals in specific eye-tracking measures and sentence parts, but no further (dis)advantage between our bilingual and native English speaker group, neither in sentence comprehension nor cognitive control, was found.Show less
Research master thesis | Linguistics (research) (MA)
closed access
Event nominalizations in Iraqw, a Cushitic language spoken in Tanzania, have a number of interesting properties. Particularly, they do not license arguments; instead, thematically related nominals...Show moreEvent nominalizations in Iraqw, a Cushitic language spoken in Tanzania, have a number of interesting properties. Particularly, they do not license arguments; instead, thematically related nominals may occur as possessors, or as arguments of the main clause. This thesis, based partially on new data gathered during fieldwork in Tanzania, provides an overview of the morphosyntactic properties of event nominals in Iraqw, as well as an analysis of their internal structure and the syntactic status of their arguments, using the framework of Distributed Morphology.Show less
Research master thesis | Linguistics (research) (MA)
under embargo until 2025-01-31
2025-01-31T00:00:00Z
Vilela is a language isolate from El Chaco (Argentina) with just one semi-speaker left. Prof. Dr. Golluscio directed its documentation within the DOBES Programme (2003-2006) and is involved in its...Show moreVilela is a language isolate from El Chaco (Argentina) with just one semi-speaker left. Prof. Dr. Golluscio directed its documentation within the DOBES Programme (2003-2006) and is involved in its grammatical description. However, its phonology is understudied, and only a few observations (Llamas, 1910; Viegas Barros, 2001; Lozano, 1970/2006) and specific studies (Grawunder & Golluscio, 2014) have been made so far. In addition, and despite its endangerment level, only one article has focused on how language attrition affected the retrieved data (Golluscio & González, 2008), and only scarce mentions of phonological processes were made. Investigations of phonetics and phonology represent a relatively late arrival within the otherwise well-established field of language attrition studies (ca. the second half of the 2010s). Therefore, this thesis has a double aim. I will first describe the Vilela's (segmental) phonological system, analyzing the available acoustic and phonological information and my fieldwork data (bibliographic research and elicitation sessions in Resistencia). Next, I study how phonological and phonetic attrition has affected and is still affecting Vilela and what this can bring to current language attrition research.Show less
Research master thesis | Linguistics (research) (MA)
open access
This thesis investigates how the use of three futurates (will, shall and BE going to) is constrained in six varieties of World English (American, British, Canadian, Indian, Jamaican, and Philippine...Show moreThis thesis investigates how the use of three futurates (will, shall and BE going to) is constrained in six varieties of World English (American, British, Canadian, Indian, Jamaican, and Philippine). Using a Construction Grammar approach, it is assumed that the futurates are separate constructions, all part of a single network (Bergs 2010). They can then be analysed in terms of form (i.e. their morphosyntactic environment), function (i.e. their semantic and pragmatic meaning) and context (i.e. the types of texts they appear in). The six varieties of World English under investigation here present a cross-section of World Englishes as they are conceptualized by Strang (1970), Quirk et al. (1972), Kachru (1985) and Schneider (2007). The data used in this thesis comes from the ICE-corpora, a collection of corpora with the same methodology and which are thus optimally comparable. The impact of various variables related to form and function was explored using conditional trees and conditional random forests. The results show that the futurates under investigation show a high degree of overlap regarding the environments in which they can be found. The functional variables included do not play a significant role. On the other hand, some variables associated with the futurates’ form do – especially the person and animacy of the subject. Additionally, the futurates are conditioned differently in different varieties of English. However, no clear categorization in terms of the frameworks designed to capture World Englishes (e.g. Kachru’s (1985) Concentric Circles or Schneider’s (2007) Dynamic Model) can be uncovered. This first look at how the use of the futurates is conditioned in six varieties of World English shows that it is an exciting topic for further investigation, as there are interesting differences between the varieties under investigation.Show less
Research master thesis | Linguistics (research) (MA)
open access
Conceptual Engineering is the practice of improving the concepts we use for a specific purpose. However, despite involving words and their meanings, this practice has not been looked at from the...Show moreConceptual Engineering is the practice of improving the concepts we use for a specific purpose. However, despite involving words and their meanings, this practice has not been looked at from the perspective of linguistics. This paper takes a small, niche scientific community, namely the Royal Society, and investigates to what extent Newton’s proposed distinction between mass and weight, which can be thought of as an instance of Conceptual Engineering, was consistently used in scientific journal articles between 1700 and 1920. Before Newton’s Principia, the lemma weight referred to both the amount of matter that makes up an object and the force of gravity acting on the matter. In 1687, Newton proposed that the former concept should be referred to as mass, and the latter as weight. Success, for this project, is differentiation based on sense and not on any other extra-linguistic factors. To discover whether the project was successful, 1500 tokens of mass and weight from the Royal Society Corpus were annotated for their lemma, ‘sense’, ‘object’, ‘subfield’, ‘author’, ‘year’, ‘type’, ‘plurality’, and part-of-speech. This data was analysed by looking at the frequencies of the respective senses over time, along with Ctrees and Random Forests to identify annotations that were the most important in predicting the lemma, and Principal Component Analysis to visually inspect clustering and patterns over time. The results showed that sense was not an effective predictor of the lemma, but that the extra-linguistic factors of ‘object’, ‘author’ and ‘subfield’ had great predictive power. Furthermore, the Principal Component Analysis confirmed clustering based on ‘object’, ‘author’ and ‘subfield’ rather than based on ‘sense’. It was concluded that Newton’s Conceptual Engineering of mass and weight was unsuccessful. From the perspective of Conceptual Engineering more generally, this work showed future research on more diverse communities with less coherent language ideologies is necessary when investigating cases of Conceptual Engineering, and also that linguists can be important contributors to Conceptual Engineering research in the future.Show less
Research master thesis | Linguistics (research) (MA)
open access
Comparative Correlative (CC) constructions, such as “the bigger they are, the harder they fall” are a relatively understudied construction within English syntax. The existing research further...Show moreComparative Correlative (CC) constructions, such as “the bigger they are, the harder they fall” are a relatively understudied construction within English syntax. The existing research further points to a great deal of disagreement on the nature of this construction. The aim of my research was to conduct a diachronic study of the CC construction, using a corpus-based approach, in order to show that the construction has remained largely the same from its earliest attestation in Old English through to the Middle English period. Five hypotheses are explored in this work: (1) there are two terms which express the same construction; (2) CC constructions are hypotactic, not paratactic; (3) the date given for the lexical change of swa to the is earlier than previously proposed; (4) the current syntactic structure can be explained by a process (micle-deletion) that took place in the Old English period, and lastly (5) that the Latin CC construction influenced the syntax of the OE construction. The results from this analysis show that hypotheses (1), (2) and (4) are true, while (3) remains unclear and (5) proves false.Show less
Research master thesis | Linguistics (research) (MA)
open access
This thesis aims to describe the noun class system and its interaction with semantics in the Tanzanian Bantu language Kimbugwe (ISO 639-3). This thesis adds to the existing literature on Kimbugwe,...Show moreThis thesis aims to describe the noun class system and its interaction with semantics in the Tanzanian Bantu language Kimbugwe (ISO 639-3). This thesis adds to the existing literature on Kimbugwe, but also has typological value, and shows the linguistic diversity of the area the language is spoken in. The main research question of the thesis is; ‘How does the noun class system of Kimbugwe interact with semantics?’. The thesis shows that the noun class system interacts with semantics on a lexical level, through the inherent semantics of the noun classes, on a morphological level, through the derivational semantics of the noun classes, and on a syntactic level, through the semantic agreement of the noun classes.Show less