Search

Causality and subjectivity are relevant cognitive principles in the categorization of coherence relations and connectives. Studies in various languages have shown how both notions can explain the meaning and use of different connectives. However, the Spanish language has been understudied from this perspective. Also, most of the existing research on connectives has used manual analyses. This paper explores the use of automatic analyses of subjectivity in causal connectives. The goal is to determine the degree to which Spanish causal connectives encode subjectivity across different text types, by carrying out automatic analyses. A corpus was constructed to identify causal connectives in journalistic texts (news and editorials) and academic texts (essays, research articles and textbooks in education and psychology). A Spanish lexicon of subjectivity was used to automatically identify the frequency of subjective words in the texts and the segments linked by the most frequent causal connectives. Our assumption is that supposedly subjective connectives will occur in more subjective environments, that is, a context containing relatively many subjective words. The results show a statistically significant relationship between the use of connectives and text type, and also between the text type and subjectivity. From a methodological point of view, the use of automatic analyses appears not to be without difficulties. However, it allowed us to explore various text types, to analyze the degrees of subjectivity in each of them and to identify tendencies related to the use of connectives in Spanish. More interestingly, the combination of automatic and manual analyses can result in a promising methodology for the study of discourse coherence and connectives.

Most children listen to speech as their primary source of communication. Yet which language they learn depends on where they grow up (Cutler, 2012): for instance, babies growing up in the Netherlands will grow up speaking Dutch while those growing up in China might master Mandarin. All these languages are markedly different: they differ in the repertoire of speech sounds, in the use of suprasegmental information (whether acoustic variations above the level of phonetics/phonemes can differentiate between words, such as word stress and tone), and in phonotactic information (which phoneme combinations are permissible in a language or not). When infants are born, their natural preferences and abilities in speech perception are hardly shaped by their native language. In other words, newborns are considered 'universal listeners' (Kuhl et al., 2008). Yet through repeated exposure to their native language, cross-linguistic differences between infants soon become apparent, which suggests that the first year of life sets the scene for language-specific listening. This chapter makes clear that speech perception is not a trivial task: speech is like a stream of sounds embedded in words combining into phrases, with no pauses that reliably signal where words begin or end (See Figure 1). Fortunately, speech contains and conveys cues to many linguistic elements simultaneously. Speech perception is the process of extracting cues from the speech stream, to recognise the message that a speaker is conveying. This process is further complicated by the fact that all speakers are different and therefore produce the cues slightly differently. In what follows next, we will first describe the input that children are exposed to (Section 15.1), before we turn to how children learn to recognise their native language from other acoustic signals (Section 15.2) and to decompose into meaningful units: into sounds (Section 15.3), into suprasegmental units (Section 15.4), and finally, into words (Section 15.5). In Section 15.6 we discuss the development of speech perception in relation to speech exposure and brain maturation. In our concluding section (Section 15.6) we underscore the relevance of early speech perception skill as crucial for language acquisition. 15.1 What kind of speech do children hear? The primary source of speech is vocal fold vibration, resulting in voiced sounds with a fundamental frequency, which is perceived as pitch. Speech can also vary in amplitude, 1 1 = Utrecht University, 2 = University of Postdam, 3 = Macquarrie University 2 perceived as fluctuations in loudness, and in the duration of segments and phrases, which may signal speaking rate, among many other things. Figure 1: A plot to indicate that speech is a stream of sounds: a sound spectrogram of a Dutch utterance 'in speech, all words are glued together', with time on the x-axis, spectral frequencies from 0 -8000 Hz (signaling vowel and consonant properties) on the left y-axis, and fundamental frequencies (pitch information signaling intonation and word stress by means of the rising and falling line in the spectrogram) on the right y-axis. The tiers below the spectrogram show the speech stream segmented into relevant subunits of speech: the top tier shows how harmonics cluster to correspond to specific speech sounds segments. The second tier groups these segments into Dutch words. Note that pauses in the speech signal the onset of plosives (/k/, /p/, /t/); they do not align with the onset or offset of words. The third tier offers a translation of the words into English. Frequency, amplitude and duration are essential cues to suprasegmental structure, such as lexical stress, tone, and intonation of an entire utterance (Fry, 1955; Pierrehumbert, 1980; Beckman, 1986). To give an example from Figure 1: the second vowel in the Dutch word /əәlkaːr/ is typically higher, louder, and longer than the first, because the second but not the first syllable carries lexical stress. These suprasegmental cues also play a role in distinguishing between the two main segmental classes, as vowels are typically voiced, and louder and longer than consonants.

Vereenvoudiging van teksten is complex. In dit corpusonderzoek gaan we na welk gevolg het streven naar begrijpelijkheid heeft voor de markering van samenhang, in het bijzonder de talige uitdrukking van causale coherentierelaties. Daartoe vergelijken we in een corpusanalyse twee recente Bijbelvertalingen: de Nieuwe Bijbelvertaling (NBV) voor relatief ervaren lezers en de Bijbel in Gewone Taal (BGT) voor lezers met een lage geletterdheid. De resultaten wijzen uit dat de BGT veelvuldig causale inferenties expliciteert die NBV-lezers zelf moeten maken, maar anderzijds ook zinnen herordent waardoor causale relaties verdwijnen. De BGT kiest vooral voor causaliteitsmarkering tussen korte finiete zinnen (want, daarom); de NBV gebruikt daarnaast varianten met een hogere informatiedichtheid: om-te-constructies en zinnen met causale woordgroepen (vanwege/ door…). Verder voert de BGT regelmatig syntactische vereenvoudiging door (meer daaromverbindingen), zonder dat dit altijd tot conceptuele vereenvoudiging leidt. Een effectstudie moet uitwijzen of genoemde ingrepen daadwerkelijk een gunstig effect hebben op de begrijpelijkheid.

Voting Advice via Internet: Political literacy in a digital information society The central aim in our NWO ‘Comprehensible Language’ project (2012-2016) was to investigate to what extent Voting Advice Applications (VAAs) intentionally and unintentionally affect political knowledge and political attitudes. In this article, we present an overview of four years of research. First, we investigated reasons for use of VAAs, distinguishing three types of users: checkers (well-informed, enjoying to check the VAA), seekers (looking for political information to base their vote on) and doubters (looking for information but cynical about politics). The proportions of these groups differ for first vs. second order elections. Second, we investigated whether VAAs increase users’ political knowledge. We found that users report an increase of internal efficacy due to their VAA use, but we did not find an increase in actual political knowledge. Third, a field experiment showed systematic effects of framing variation on the answers to VAA assertions, which might suggest different underlying knowledge representations. Finally, think aloud research showed that users experience considerable problems with understanding the assertions semantically and pragmatically, as well as with interpreting the results screen. Additionally, we found that users view the result screen as an end point rather than as a starting point for deliberation. We discuss some implications for theory and practice.

In this study, the influence of parental input on the acquisition of discourse connectives was investigated. Three factors were hypothesized to play a role in contributing to the course of language acquisition: first, an increase in age, and hence, an increase in conceptual abilities; second, short-term frequency effects (effects of parental input in the space of one recording); and third, long-term frequency effects (effects of the cumulative parental input over a longer period of time). The authors developed a growth curve analysis and used this to analyze data from a dense longitudinal corpus of a German boy aged 1;11.12—2;11.27. Results show that each factor has a significant effect on the acquisition of the German connectives aber `but,' damit `so that,' und `and,' weil `because,' and wenn `when' and should always be taken into account when studying connective acquisition. Furthermore, growth curve analysis promises to be an innovative tool to study factors influencing the course of language acquisition.

OCP-Place, a cross-linguistically well-attested constraint against pairs of consonants with shared [place], is psychologically real. Studies have shown that the processing of words violating OCP-Place is inhibited. Functionalists assume that OCP arises as a consequence of low-level perception: a consonant following another with the same [place] cannot be faithfully perceived as an independent unit. If functionalist theories were correct, then lexical access would be inhibited if two homorganic consonants conjoin at word boundaries—a problem that can only be solved with lexical feedback. Here, we experimentally challenge the functional account by showing that OCP-Place can be used as a speech segmentation cue during pre-lexical processing without lexical feedback, and that the use relates to distributions in the input. In Experiment 1, native listeners of Dutch located word boundaries between two labials when segmenting an artificial language. This indicates a use of OCP-Labial as a segmentation cue, implying a full perception of both labials. Experiment 2 shows that segmentation performance cannot solely be explained by well-formedness intuitions. Experiment 3 shows that knowledge of OCP-Place depends on language-specific input: in Dutch, co-occurrences of labials are under-represented, but co-occurrences of coronals are not. Accordingly, Dutch listeners fail to use OCP-Coronal for segmentation.

Respondents are more likely to disagree with negative survey questions (This text is boring. Yes/No) than to agree with positive ones (This text is interesting. Yes/No). The size of this effect, however, varies largely between word pairs. A semantic classification of adjectives in closed scale/absolute and open scale/relative types was predicted to explain this variation. To classify survey adjectives, a judgment experiment was conducted. Language users (N = 173) rated sentences in which an adjective was modified by the maximizer completely or the approximator almost: it should be possible to combine closed scale/absolute adjectives with these modifiers, in contrast to open scale/relative adjectives for which this is not the case. Results show that language users agree on which adjective and degree modifier combinations are acceptable and which combinations are unacceptable. Moreover, the two methods, almost and fully, show convergent validity. However, the rating of the same combination of a specific adjective and a specific degree modifier varies across contexts. This suggests that neither of the two methods allows for an unambiguous classification of adjectives. Hence, the distinction between closed scale/absolute and open scale/relative adjectives cannot explain variation in survey response effects. For semantics and pragmatics results indicate that context plays a crucial role in the linguistic behavior of adjectives and degree modifiers.

Social constructivism suggests that learning is an inherently social phenomenon. The ap- proaches presented in this dissertation emphasize the social nature of information, knowledge and language, using a combination of ontologies, reference repositories, folksonomy analysis, graph-based disambiguation and topic modeling. Differences between the vocabulary of an individual and that of a Community of Practice present both a problem and an opportunity. The problem is that it can impede access to appropriate resources for learning, e.g. keyword-based search can be very sensitive to subtle lexical differences. I have shown that poor alignment of an individual’s vocabulary with that of a community can be addressed using enriched ontologies in combi- nation with semantic search. The opportunity is that analysis of the vocabulary and ‘lexical competence’ of an individual reveals his or her conceptual knowledge and allows computer algorithms to assist individuals with reflection and critical thinking. Uptake of the proper community vocabulary is a good predictor of an individual’s integration with a community, because it signifies the acceptance and acquisition of the community’s conceptualization and vocabulary. The tight integration between conceptual knowledge and vocabulary usage is useful in the context of assessment and personalization in order to tailor resources and individuals to the appropriate level of knowledge by means of topic models.

For decades, survey researchers have known that respondents give different answers to attitude questions worded positively (X is good. Agree-Disagree), negatively (X is bad. Agree-Disagree) or on a bipolar scale (X is bad-good). This makes survey answers hard to interpret, especially since findings on exactly how the answers are affected are conflicting. In the current paper, we present twelve studies in which the effect of question polarity was measured for a set of thirteen contrastive adjectives. In each study, the same adjectives were used so the generalizability of wording effects across studies could be examined for each word pair. Results show that for five of the word pairs an effect of question wording can be generalized. The direction of these effects are largely consistent: respondents generally give the same answers to positive and bipolar questions, but they are more likely to disagree with negative questions than to agree with positive questions or to choose the positive side of the bipolar scale. In other words, respondents express their opinions more positively when the question is worded negatively. Even though answers to the three wording alternatives sometimes differ, results also show that reliable answers can be obtained with all three wording alternatives. So, for survey practice, these results suggest that all three wording alternatives may be used for attitude measurement., Survey Research Methods, Vol 7, No 3 (2013)

This dissertation is concerned with the deictic and grammatical category person and its associated linguistic expressions: indexical, i.e. first and second person, pronouns. Under the hypothesis that sentence meaning is computed from how words and their associated lexical information combine with each other, indexical pronouns present an interesting challenge: their actual referent can only be determined once interpreted with respect to the utterance context they are being used in. Formulated within the Chomskyan framework of generative syntax (cf. e.g. Chomsky 1995, 2000), this thesis develops a novel approach to the relationship between the morphosyntactic content, the interpretational ranges, and the associated crosslinguistic variation of these pronouns. Primarily based on data from Dutch, English, German (Indoeuropean), and Blackfoot (Algonquian), the main conclusion of this thesis is that person is derivative of spatiotemporal information. Adopting ideas developed in Musan (1995), the temporal component is proposed to restrict the interpretation to specific moments in time, i.e. it picks out a particular temporal stage of the individual denoted by the pronoun. Secondly, spatial information contributes the necessary contextual anchoring and is responsible for identifying the speaker or the hearer. Adapting Déchaine and Wiltschko’s (2002) analysis of the internal structure of pronouns, these spatiotemporal pieces of information are linked to separate projections in the underlying syntactic structure of indexical pronouns. Empirically, this dissertation draws on data from the domains of genericity, possession, tense, and modality. It is shown that the spatiotemporal information underlying indexical pronouns is also encoded in the morphosyntax of indexical pronouns. From a broader perspective, this dissertation addresses questions concerning the relation between context and syntax, morphology and syntax, as well as syntax and semantics.

Coherence relations and their linguistic markers play a significant role in the study of discourse processing and comprehension. A number of studies have shown that the presence of coherence markers, such as connectives, in a text facilitates the processing and representation of discourse. Therefore, connectives are assumed to function as processing instructions (e.g. Britton, 1994; Gernsbacher, 1997; Murray, 1995; Sanders & Spooren, 2007); they help the readers to arrive at the intended interpretation of the text. The main purpose of the current study was to get a better understanding of how the information encoded in connectives affects online discourse processing. More specifically, we focused on causal connectives, which in Dutch provide cues about the fine-grained distinction between subjective and objective causal relations. Objective causal relations express causality between events in the real world. In subjective causal relations, however, the causal relations does not exists in the real world but in the mind of the speaker who is drawing conclusions on the basis of events in the world. There are two reasons why it is relevant to study the notion of subjectivity in relation to causal connectives. First, in many languages of the world causal connectives seem to specialize in either subjective or objective causal relations (Stukker & Sanders, 2012). And second, subjective causal relations are assumed to be more complex than objective causal relations (Sanders, Spooren, & Noordman, 1992). This complexity is associated with longer processing times in reading experiments (Noordman & De Blijzer, 2000; Traxler, Bybee, & Pickering, 1997; Traxler, Sanford, Aked, & Moxey, 1997). In the current dissertation we present a series of eye-tracking experiments that were designed to investigate the influence of Dutch causal connectives on discourse processing. With respect to Dutch backward causal relations, the connective want is a prototypical marker of subjective causal relations, whereas omdat is a prototypical marker of objective causal relations (Degand & Pander Maat, 2003; Pit, 2003; Sanders & Spooren, 2009, 2012; Verhagen, 2005). In forward causal relations dus typically expresses subjective relations, whereas daarom is typically used in objective relations (Pander Maat & Degand, 2001; Pander Maat & Sanders, 2000, 2001; Stukker, 2005; Verhagen, 2005). Our results show that both backward and forward subjective causal connectives induce an immediate delay in processing times compared to their objective counterparts. These results reveal that causal connectives do more than just inform the reader that a causal coherence relation needs to be constructed between two pieces of text. They also provide information about the relative degree of subjectivity of that causal relation. On the basis of additional experiments we conclude that the processing complexity of subjective causal connectives and relations can be explained by the notion of speaker involvement (Pander Maat & Degand, 2001; Pander Maat & Sanders, 2000, 2001; Pit, 2003). Subjective causality requires the representation of the Mental Space of the thinking subject -be it the speaker, author, or a third person character in the discourse- who is responsible for the presented information. This representation comes at a processing cost.

One of the core features of language is that words in a sentence often fulfill a dual task. For instance, in sentences with raising verbs like seem, what appears on the surface as the subject of seem, is interpreted as the subject of the embedded proposition. That is, in the sentence Tommy seems to have eaten the chocolate cake, Tommy appears as the subject of seem, but he’s not interpreted as the ‘seemer’. What is evaluated by seem is his property of being the eater of the cake. There is thus a discontinuity between the surface position and the position of interpretation; the subject has been displaced. In this thesis I investigate why subject displacement is difficult for our brain to process, and why it can go together with delays in language acquisition. Interestingly, raising verbs like seem not only display subject displacement, but also allow us to talk about realities that differ from our current one; they have an evidential meaning. As such, not only subject displacement can cause trouble in acquisition, but also the precise interpretational effects these raising verbs have. Another structure that displays subject displacement is the unaccusative structure (e.g. Tommy fell). The subject of unaccusative verbs has properties typically associated with syntactic objects, although it appears in subject position. Unaccusative verbs differ in this from unergative verbs for which the subject displays subject properties. Processing and acquisition of raising verbs is compared to processing and acquisition of unaccusative verbs to tear apart the separate effects of subject displacement and the specific interpretational requirements of raising verbs. In an innovative visual world experiment, we show that unaccusative structures with subject displacement delay interpretation compared to unergative structures. There is thus a close match between grammatical computation and processing. Furthermore, children from the age of five already distinguish between unaccusatives and unergatives. Apparently, children from that age are already aware of the subject displacement in unaccusative verbs. Raising verbs on the other hand, are acquired at a later point in acquisition. Moreover, depending on the precise interpretational effects, differences in the timing of acquisition occur. Similarly, different raising verbs have a different effect on processing in adults. This thesis has therefore shown that in order to understand the behavior of raising verbs in acquisition and processing, we have to distinguish on the one hand the effect they have in their capacity of raising verbs and in addition the effect in terms of their specific contribution to semantic interpretation. Their contribution to semantic interpretation affects processing and the course of acquisition. The precise interpretational effects do not come for free. It is therefore not only the syntactic computation, but also the computations required to establish the right semantics that influence the acquisition path. The language-acquiring child who has acquired subject displacement, does not necessarily possess all the tools that are required for an adult-like interpretation of the raising verbs. Linking the specific interpretation in terms of the reality under discussion to the particular raising verb takes somewhat longer.

In many countries with multiparty systems, a decline in class voting has increased volatility and the need for comprehensive information about the political landscape among voters. Voting Advice Applications (VAAs) are online tools that match users to political parties and, as such, they hold a promise of reinforcing informational transparency and democratic representation. The current research investigated whether VAAs live up to this expectation by investigating to what extent VAAs affected users' political knowledge and vote choice in the Dutch national elections of 2012. Results show that VAA users feel that the VAA improved their political knowledge. In addition, those groups of VAA users who experienced a large knowledge increase, also relatively often indicated that their vote choice had been affected. This suggests that VAAs contribute to informational transparency by increasing knowledge among a potentially wide audience, and also that VAAs might increase democratic representation to the extent that VAAs persuade people to vote for the candidate that best represents their opinions. On the other hand, we found discrepancies between behavioural and perceptual measurements of the effect of VAAs on vote choice. This raises doubts about whether VAAs shape actual voting behaviours and knowledge, or rather perceptions of that.

The dissertation investigates the possible innate perceptual biases that may shape the phonological rule in Mandarin, T3 sandhi. T3 sandhi restricts the co-occurrence of two T3s and requires the first T3 to change to a T2. A second goal of the dissertation is to examine the cross-domain perception of pitch from infancy to adulthood. Mandarin and Dutch adult listeners participated in a categorical perception experiment and a speeded AX discrimination task. It is found that Mandarin listeners perceive Mandarin the T2-T3 contrast categorically while Dutch listeners perceive this contrast in a psycho-acoustical manner. Nevertheless, both groups of listeners are more accurate in discriminating T2 and T3 if T3 occurs first in the to-be-discriminated pair. This consistency provides strong evidence that there are innate perceptual biases favoring T3 sandhi to occur in such a positional and asymmetrical way. Dutch infants show easier categorization of T3 than of T2. Hence, the discrimination asymmetry can be a result that T3 works as a better referent in discrimination. Positive significant correlation between musical pitch processing and Mandarin lexical tone discrimination is observed among Dutch adult listeners but not Mandarin adult listeners. Dutch young infants (4 and 6 months old) show facilitated processing of musical pitch while they keep showing difficulties in discriminating Mandarin T2-T3 contrast, which suggests separate music and speech processing from early infancy. Mandarin infants do not outperform Dutch infants in either musical pitch processing or lexical tone discrimination. No evidence has been found for the correlation between the perception of musical pitch and the perception of lexical tones in early infancy. This dissertation is of interest to psycholinguists, phonologists, and those who work in early perception development in infancy.

Causal connectives are often considered to provide crucial information about the discourse structure; they signal a causal relation between two text segments. However, in many languages of the world causal connectives specialise in either subjective or objective causal relations. We investigate whether this type of (discourse) information is used during the online processing of causal connectives by focusing on the Dutch connectives want and omdat, both translated by because. In three eye-tracking studies we demonstrate that the Dutch connective want, which is a prototypical marker of subjective CLAIM–ARGUMENT relations, leads to an immediate processing disadvantage compared to omdat, a prototypical marker of objective CONSEQUENCE–CAUSE relations. This effect was observed at the words immediately following the connective, at which point readers cannot yet establish the causal relation on the basis of the content, which means that the effect is solely induced by the connectives. In Experiment 2 we demonstrate that this effect is related to the representation of the first clause of a want relation as a mental state. In Experiment 3, we show that the use of omdat in relations that do not allow for a CONSEQUENCE–CAUSE interpretation leads to serious processing difficulties at the end of those relations. On the basis of these results, we argue that want triggers a subjective mental state interpretation of S1, whereas omdat triggers the construction of an objective CONSEQUENCE–CAUSE relation. These results illustrate that causal connectives provide subtle information about semantic-pragmatic distinctions between types of causal relations, which immediately influences online processing.

This dissertation investigates the acquisition of particle verbs in Dutch. Based on longitudinal child data it shows how children acquire lexical categories V, A and P and how they acquire the complex predicate status of particle verbs. It shows that the acquisition of categories A and P is dependent on the acquisition of category V and that the acquisition of category V is tightly connected to the acquisition of verb-second. It is the acquisition of this verb-second that also leads to the acquisition of the complex predicate status of particle verbs. It further shows that the acquisition process inevitably leads children into overgeneralisation of preposition stranding, but that the same acquisition process also inevitably leads children out of it. This dissertation further presents data from experiments with adult speakers that shed an interesting light on the role this overgeneralisation still plays in the adult grammar. Investigating the acquisition of particle verbs cannot be done without taking a closer look at the acquisition process itself. As such, this dissertation also addresses more general questions concerning the acquisition process, such as the role of the lexicon and of generalizations. It shows an acquisition process in which the child makes use of pragmatic bootstrapping and syntactic cueing to build layer upon layer of grammar without the need of inherently, specifically linguistic knowledge. This dissertation will be of interest to anyone working on language acquisition, the syntax of particle verbs and preposition stranding, and the interaction between syntax and information structure.

My thesis studies communication systems that arise in the absence of linguistic conventions: restricted linguistic systems. Home sign and the language of unsupervised adult second language learners are examples of such systems. Jackendoff (2002) observed that utterances in these systems are governed by semantic principles, such as Agent First (a principle responsible for the intuitive difference between the two simple utterances HIT-TREE-FRED and HIT-FRED-TREE), and argued that they reflect ancient linguistic structures. As such, studying restricted linguistic systems offers a window into the earliest stages of language evolution, as well as offering insights into the cognitive biases which act to shape modern linguistic systems. In my thesis I related restricted linguistic systems to a novel experimental approach, improvised communication. I showed that gesture sequences in the improvised communication experiment are governed by semantic principles which are essentially the same as those governing restricted linguistic systems: we can therefore use improvised communication experiments as a source of evidence for semantic principles in evolutionarily early language. Moreover, I used improvised communication experiments to reveal a new semantic ordering principle governing the expression of intensional meaning. In an improvised communication experiment, participants are asked to describe events using only gesture and no speech. The methodology was first presented by Goldin-Meadow et al. (2008), who showed that participants by-pass the grammar rules of their native language and use a gesture order that is consistent with SOV (i.e. Subject-Object-Verb). Goldin-Meadow et al. conclude that this is the order in which we naturally represent events. In one study I contrasted extensional (e.g., ‘Man kicks ball’) and intensional (e.g., ‘Man searches for ball’) events. These two kinds of events differ semantically (e.g., ‘ball’ is a concrete object when it is an argument of ‘kick’, and an abstract object when it is an argument of ‘search for’), and participants use different orders to communicate about them: they consistently use SOV order for extensional and SVO order for intensional events. In a follow-up study I show that this flexibility in word order has a communicative function: ambiguous SOV gesture sequences are interpreted extensionally, while ambiguous SVO sequences are interpreted intensionally. These findings suggest added subtlety to the Goldin Meadow account: word order in improvised communication is flexible and determined by fine-grained properties of the meaning to be conveyed.

The benefits of evaluating websites among potential users are widely acknowledged. There are several methods that can be used to evaluate the websites’ quality from a users’ perspective. In current practice, many evaluations are executed with inadequate methods that lack research-based validation. This thesis aims to gain more insight into evaluation methodology and to contribute to a higher standard of website evaluation in practice. A first way to evaluate website quality is measuring the users’ opinions. This is often done with questionnaires, which gather opinions in a cheap, fast, and easy way. However, many questionnaires seem to miss a solid statistical basis and a justification of the choice of quality dimensions and questions. We therefore developed the ‘Website Evaluation Questionnaire’ (WEQ), which was specifically designed for the evaluation of governmental websites. In a study in online and laboratory settings the WEQ has proved to be a valid and reliable instrument. A way to gather more specific user opinions, is inviting participants to review website pages. Participants provide their comments by clicking on a feedback button, marking a problematic segment, and formulating their feedback. There has been debate about the extent to which users are able to provide relevant feedback. The results of our studies showed that participants were able to provide useful feedback. They signalled many relevant problems that indeed were experienced by users who needed to find information on the website. Website quality can also be measured during participants’ task performance. A frequently used method is the concurrent think-aloud method (CTA), which involves participants who verbalize their thoughts while performing tasks. There have been doubts on the usefulness and exhaustiveness of participants’ verbalizations. Therefore, we have combined CTA and eye tracking in order to examine the cognitive processes that participants do and do not verbalize. The results showed that the participants’ verbalizations provided substantial information in addition to the directly observable user problems. There was also a rather high percentage of silences (27%) during which interesting observations could be made about the users’ processes and obstacles. A thorough evaluation should therefore combine verbalizations and (eye tracking) observations. In a retrospective think-aloud (RTA) evaluation participants verbalize their thoughts afterwards while watching a recording of their performance. A problem with RTA is that participants not always remember the thoughts they had during their task performance. We therefore complemented the dynamic screen replay of their actions (pages visited and mouse movements) with a dynamic gaze replay of the participants’ eye movements. Contrary to our expectations, no differences were found between the two conditions. It is not possible to draw conclusions on the single best method. The value of a specific method is strongly influenced by the goals and context of an evaluation. Also, the outcomes of the evaluation not only depend on the method, but also on other choices during the evaluation, such as participant selection, tasks, and the subsequent analysis.

The two key questions recurring in the literature on loanword studies are: are there any general patterns of loanword adaptation, and if so, which forces drive the process(es) that result in these patterns. This dissertation readdresses these two questions on the basis of a study of loanword phonology in which Mandarin Chinese (MC) is the recipient language (RL) and English the source language (SL). It argues that factors that have not received much attention in the field have to be taken into consideration. They are input types and adapter types. In this dissertation, the roles of the two factors are put to the test. By investigating both loanword corpus data and extensive experimental results, this dissertation presents an analysis of MC loanword phonology that can account for the intricate adaptation patterns attested in a corpus of loanwords borrowed from English into MC. It moreover shows that, depending on input and/or adapter types, loanword adaptation is driven by one or more of the following grammars: the native RL phonological production grammar, the native RL perception grammar, the RL/SL interlanguage phonological production grammar and the RL/SL interlanguage perception grammar. This dissertation will be of interest to all those working on loanword adaptation, Mandarin Chinese phonology, perception and production in second language acquisition and the role of orthography in linguistics.

Standardized surveys are used in many contexts to measure people’s opinions and attitudes. Although it is widely assumed that survey answers represent the ‘true values’ of the concepts measured, a large body of research has shown that seemingly irrelevant question characteristics influence how respondents report their attitudes. The research presented in this dissertation revolved around on one of these characteristics: whether the question is worded positively (This is an interesting book. Yes/No.) or negatively (This is an uninteresting book. Yes/No.). A first aim of the current dissertation was to investigate whether the choice for a positive or a negative wording systematically affects survey answers. Results show that there is an effect of question polarity on survey answers: irrespective of the word pair used, the answering scale offered, and the question asked, respondents are more likely to answer no or disagree to negative questions than to answer yes or agree to equivalent positive ones. Hence, respondents express their opinions more positively when the question is worded negatively (Chapter 2, 4, 5). Even though the choice for a positive or a negative wording was shown to affect survey answers systematically, we also observed variation between word pairs in the size of the effect. A second aim of the current dissertation was to explain this variation using a semantic classification of antonyms in absolute versus relative types (Kennedy & McNally 2005). Unfortunately, we were unable to explain variation between word pairs using this typology (Chapter 3). The third and last aim of the current dissertation was to explain cognitively why respondents, in general, answer positive and negative attitude questions differently. To do so, we used eye-tracking as a method to measure the temporal aspects of the question-answering process for positive versus negative questions. In order to relate the observed time differences to cognitive processes of question answering (Tourangeau, Rips & Rasinski 2000), we used a combination of theoretical insights about the question-answering process, experimental designs in which that knowledge was used, and newly designed eye-tracking measures that are relevant for these specific purposes. With this combination of strategies we were able to demonstrate that positive and negative questions cause similar information to be retrieved from memory and a similar attitude to be formed. Therefore, contrastive questions measure the same underlying attitude, and hence, are equally valid. Wording effects arise only late in the question-answering process, when respondents translate their own opinion into the response options to the question. This is due to the fact that respondents assign a meaning to the response options that is not absolute, but relative to the polarity of the evaluative term used in the question. Hence, although response options like yes and no are straight opposites, the meaning of yes in answer to a positive question is not the same as the meaning of no in answer to a negative question. This dissertation concludes with a discussion of the implications of these results for survey theory and suvey practice, as well as the methodological implications about the use of eye-tracking for measuring cognitive processes

This thesis is about writing proficiency among students of secondary education. Due to globalization, the ability to express oneself in a language other than the first language (L1) is increasingly becoming a condition for educational success. In The Netherlands, this ‘other’ or second language (L2) is usually English. Although secondary school students are already quite able to express themselves in English, their L2 essays are often of lower quality than L1 essays, in terms of language use, but also in terms of organization. The research reported in this thesis was aimed at explaining this quality difference by comparing L1 and L2 relations between essay quality and writing processes. Analyses of writing processes involved cognitive activities such as reading the assignment, process planning, content planning, evaluating and revising. Results show that, in general, cognitive activities are relevant to essay quality at different stages of task execution during L1 and L2 writing. This means that writers need to distribute their attention differently across task execution during L1 and L2 writing. However, additional analyses show that if students’ general language proficiency levels are included in the analyses, this L1/L2 contrast disappears. For students with high general L2 proficiency, the demands of L2 writing in terms of how often cognitive activities are applied during certain stages of task execution are similar to the demands of L1 writing.

Sentences containing multiple quantified expressions often display scope ambiguity. However, in certain sentence-types scope interaction is unexpectedly restricted, for instance in double-object constructions containing an existentially quantified indirect object and a universally quantified direct object (The bear gave a fox every cake). In such cases only the existential quantifier can take wide scope. Inverse scope is blocked. This constraint is known as Scope Freezing (Larson, 1990; Bruening, 2001). Preschool children do not seem to have knowledge of this constraint, and in fact show a very distinctive performance pattern. Children reject the wide scope interpretation of the existential quantifier (a single fox is given every cake by the bear), but allow the inverse scope interpretation (every cake is given to a different fox by the bear). Their performance pattern is thus the opposite of the targetlike pattern, and therefore we named this pattern the Reverse-pattern. In this study we attempt to explain why children show the Reverse-pattern. We start out by investigating the possibility that children perform non-targetlike because they have not fully acquired the meaning of the existential quantifier(s) in question. We test predictions of both Su’s (2001) Lexical Factor Hypothesis, which argues that the non-targetlike performance is due to incomplete acquisition of the meaning of the indefinite article, and a new hypothesis called the Singleton Restriction Hypothesis which locates the problem in an inability on the part of children to use the discourse context to restrict indefinites to singleton sets. Neither hypothesis is able to capture the occurrence of the Reverse-pattern. We next turn our attention to the role of universal quantification. In a series of experiments we show that the Reverse-pattern is in fact restricted to sentences containing distributive universal quantifiers. We hypothesise that the basic interpretation of a distributive universal quantifier requires prototypical distributivity. We further hypothesise that children strongly adhere to this prototypically distributive interpretation and, unlike adults, cannot deviate from it. However, experimental evidence shows that children do not just rely on prototypical distributivity, but also require Full Set Linking (i.e. an exhaustive linking of the set of objects and recipients). This study finally argues that children get stuck in a verification strategy that adults use as a short-cut to interpretation. Whereas adults can switch interpretation strategies when required, children cannot. Rather, children attempt to accommodate situations which are not consistent with the short-cut verification strategy, leading to a reliance on Full Set Linking. All this can take place, because children are not constrained by Scope Freezing, which is argued to be an Information-Structural Constraint which is required late.

One of the most challenging tasks for language-learning infants and second language (L2)-learning adults is to segment the continuous stream of speech that surrounds them, and, following this, to acquire a lexicon. Both speech segmentation and lexical acquisition are known to be facilitated by phonotactics, i.e., language-specific restrictions on how phonemes may combine. This dissertation addresses questions regarding the representation and acquisition of such phonotactic knowledge in a native language and an L2. Five experimental studies are presented. The first three studies, using the artificial language learning paradigm, reveal that segmentation is influenced by structural phonotactic knowledge of OCP-PLACE, a restriction against pairs of consonants sharing the feature [Place]. It is shown that this knowledge is used only by native listeners or advanced L2 learners of a language restricted by the constraint. This suggests a language-specific acquisition from the input. The third study, with infants, shows that this input is continuous speech rather than the lexicon. The remaining two studies demonstrate that abstract phonotactic knowledge of syllable structure is represented separately from specific probabilistic knowledge, as the two have separate effects on lexical acquisition in a short-term memory recall task. Moreover, results from L2 learners suggest that probabilistic knowledge can be acquired independently of structural knowledge of the L2. While most studies have looked at the influence of specific representations of phonotactic probability, here it is shown that representations of abstract structural constraints also influence processing. Moreover, it is demonstrated that both types of phonotactic representations are acquired from the input

Phonotactic knowledge is knowledge of the sound combinations of a language. Previous research showed that frequent sound combinations are perceived more easily than illformed ones, while illegal combinations are often filtered out or adapted. Phonotactics can be described in categorical terms (i.e., a sound combination is legal or illegal), or in gradient terms (i.e., a combination has a certain level of wellformedness). In addition, phonotactics can both be described in negative or positive terms, by defining when combinations are legal vs.\ when they are illegal. This thesis present experimental research on the mental representation of phonotactics that is used for speech perception. A cross-modal priming experiment on second language listening showed that consonant clusters that are frequent in the second language are easier to process, in spite of perceptual epenthesis, an illusion caused by illegality of the same clusters in the first language. This disjunction between a categorical and a gradient effect suggest separate representations of both categorical and gradient phonotactics. However, the dichotomy between categoricalness and gradience is confounded with the dichotomy between negative and positive representations. The categorical filtering effects are based on negative knowledge, while the gradient facilitation effect is based on positive knowledge. Therefore, phonotactic effects on speech segmentation were investigated next. In segmentation, the effect of negative categorical knowledge is facilitatory, instead of inhibitory. Illegal combinations cue word boundaries, because illegal combinations cannot occur word-internally. An eye-tracking method was developed to test segmentation effects independently of word activation. An experiment with Dutch native listeners showed that phonotactic knowledge has an effect on segmentation in the absence of full lexical recognition. A follow-up experiment showed that illegal segmentations are inhibited; there might additionally be a bias toward more wellformed segmentations. In the same experiment, second language listeners of Dutch were more likely to split consonant clusters that are illegal in Dutch, but legal in their native language. Nevertheless, this effect was more categorical for native listeners. When other, more wellformed segmentations fail to yield words, second language listeners make segmentations that leave illegal clusters intact, which native listeners hardly do. These results show the existence of categorical effects of illegality and gradient effects of wellformedness. All second language learners studied before showed transfer of categorical phonotactic illegality from their first language. To test if categorical knowledge is always fossilised, the segmentation of English and Dutch by native Dutch listeners, proficient in English, was studied. The listeners applied English categorical phonotactics to English, showing that categorical phonotactics can be acquired. The empirical data presented in this thesis suggest that phonotactic representations can be intrinsically positive and negative, while their effects are categorical and gradient, depending on the task and the strength of the representations. A model of phonotactic grammar is proposed. This grammar is based on constraints representing positive or negative knowledge. It is applied to speech before lexical access takes place and can yield both gradient and categorical effects on speech perception.

One of the milestones in children’s language development is the ability to construct a discourse. In order to construct a discourse, children must learn to set up coherence relations between clauses. In this study, we focused on the acquisition of causal relations, and in particular on children’s development of causal connectives, such as because, that are used to mark causal relations. For example, children can use because to mark the objective cause-consequence relation in (1) or the subjective claim-argument relation in (2). (1) He’s so fat because there’s uh baby inside. (Shem, 3;00.05) (2) He’s a baby because he goes in a stroller. (Nina, 3;00.10) In this study we tracked young children’s production of causal connectives by running growth-curve analyses on longitudinal corpora. We investigated the role of two factors that have been hypothesized to influence the course of connective acquisition: parental input and cognitive complexity. We took a cross-linguistic approach by studying English because as well as its German counterpart weil and the Dutch equivalents want and omdat. In addition, we tested young children’s comprehension of causal relations in an innovative eye-tracking experiment that used the preferential looking paradigm. Results revealed that both cognitive complexity and parental input play a role during connective acquisition. We showed that the cognitive complexity of objective and subjective relations determines the acquisition order of these relations: the comprehension and production of the least complex objective causal relations develop ahead of that of the more complex subjective causal relations. We also showed that both parental connective input and parental scaffolding through why-questions influence the course of acquisition. In fact, our analysis of parental scaffolding through why-questions showed that effects of input and complexity are intertwined: parents adapt the complexity of these question-answer routines to their child’s cognitive ability. Our findings allowed us to construct a model of causal connective development. It consists of four phases that are cumulative in nature, each phase adding an extra ability to the previous one(s). First, children start out by developing an understanding of causality in which their comprehension of objective causality develops ahead of that of subjective causality. Second, parents start to scaffold connective production by asking why-questions. Children give a causal response to these questions, but do not use a connective to mark the relation. Third, children start to include connectives in their responses to why-questions, and they start to use connectives independently (on their accord). When we look at the type of relations they produce, we find that objective relations develop before subjective relations. Fourth, children start asking why-question themselves. While answering these questions, parents make use of causal connectives, and thereby provide input ‘on demand’. In this study we have used converging evidence – with different languages and different methods – to investigate causal connective acquisition. We have shown that connective acquisition is an intricate system in which parental input and cognitive complexity play an important role.

Background. We present a new theoretical model for word-finding difficulties in aphasic patients. We show how: (1) lexical complexity can be measured using the notion of entorpy; (2) the higher the entropy the more resources are needed to process it; and (3) lack of resources results in failure to access lexical items (inability to handle higher entropy values).

Roots and Affixes is an investigation into the primitives of syntax. It focuses on the lexical projection and the categorial head. Accordingly, it consists of two parts. The first part argues that the features of lexical vocabulary items (such as light and kiss) are not an active part of the syntactic derivation. The author provides empirical support for the claim that vocabulary items are inserted post-syntactically, adopting the view that syntax operates on UG-features only. She argues that the root terminal node is a by-product of the operation Merge that is characterized by the mere absence of features. It is further shown that functional structure determines subcategories of lexical items. In the second part of the thesis it is argued that categorial heads do not exist. As a result, derivational affixes do not realize categorial heads. The author proposes instead that derivational affixes are lexical vocabulary items which realize root positions. It is shown that the abandonment of categorial heads does not lead to a loss of explanatory adequacy. The general conclusion is that lexical categorial features are not a primitive of syntax.

During the first year of life, infants start to learn various properties of their native language. Among these properties are phonotactic constraints, which state the permissible sound sequences within the words of the language (e.g., Dutch words typically do not contain the sequence ‘pf’). Such constraints guide infants' search for words in continuous speech, thereby facilitating the development of the mental lexicon. An intriguing problem is how infants are able to acquire knowledge of phonotactics. This dissertation proposes a computational model of phonotactic learning, which is based on psycholinguistic findings. The model connects two learning mechanisms: statistical learning and feature-based generalization. Using these mechanisms, phonotactic constraints with varying levels of generality are induced from transcribed utterances of continuous speech. The constraints are subsequently used for the detection of word boundaries. The computational model, StaGe (Statistical learning and Generalization), induces biphone constraints through the statistical analysis of segment co-occurrences in continuous speech (Frequency-Driven Constraint Induction), and generalizes over phonologically similar biphone constraints to create more general, natural-class-based constraints (Single-Feature Abstraction). The model learns phonotactic constraints of two types: markedness constraints (*xy) and contiguity constraints (Contig-IO(xy)). The former type exerts pressure towards the insertion of boundaries into the speech stream, while the latter type militates against the insertion of boundaries. Conflicts between constraints are resolved using principles of Optimality Theory. The model is tested in various empirical studies. A series of computer simulations demonstrate that a crucial property of the model, feature-based generalization, improves the segmentation performance of the learner. In addition, StaGe provides a learnability account of a constraint from theoretical phonology, OCP-Place, and of its effect on human segmentation. A detailed analysis of the constraint set that is induced by StaGe reveals that the model learns constraints that resemble OCP-Place through abstractions over statistical patterns found in continuous speech. Moreover, these simulations show that both specific and abstract phonotactic constraints are needed to account for human segmentation behavior. The psychological plausibility of the phonotactic learning approach is addressed in a series of artificial language learning experiments with adult participants. It was found that human learners can induce novel phonotactics from continuous speech. The results were limited to constraints on specific consonants. No evidence was found for feature-based generalization to novel consonants (i.e., consonants not presented during training). This null result was possibly due to perceptual confusion. This dissertation demonstrates that phonotactic constraints can be learned from continuous speech by combining mechanisms that are available to language learners. The computational model provides a better account of speech segmentation than models that rely solely on statistical learning. With respect to human learning capacities, the dissertation shows that adult learners can induce novel phonotactic constraints from a continuous speech stream from an artificial language. By combining computational modeling with psycholinguistic experiments, this dissertation contributes to our understanding of the mechanisms involved in early language acquisition

In dit onderzoek is de rol van de moedertaal (T1) tijdens het schrijven in een tweede taal (T2) onderzocht. Twintig studenten schreven ieder vier opstellen in T1 (Nederlands) en T2 (Engels) onder hardop-denkcondities. Daarbij is gekeken naar de mate waarin zij hun T1 gebruikten tijdens het schrijven in T2, in hoeverre dit varieert tussen schrijvers en taken, en of T1-gebruik gerelateerd is aan tekstkwaliteit. Ook is nagegaan of de relatie tussen T1-gebruik en tekstkwaliteit beïnvloed wordt door T1-schrijfvaardigheid en T2-taalvaardigheid. T1-gebruik blijkt deels een negatieve invloed te hebben op tekstkwaliteit in T2, maar niet voor alle onderzochte cognitieve activiteiten. Daarnaast blijkt T1-gebruik meer voor te komen bij onervaren T1-schrijvers dan bij ervaren T1-schrijvers.

Survey designers have long assumed that respondents who disagree with a negative question (“This policy is bad.”: Yes or No; 2-point scale) will agree with an equivalent positive question (“This policy is good.”: Yes or No; 2-point scale). However, experimental evidence has proven otherwise: Respondents are more likely to disagree with negative questions than to agree with positive ones. To explain these response effects for contrastive questions, the cognitive processes underlying question answering were examined. Using eye tracking, the authors show that the first reading of the question and the answers takes the same amount of time for contrastive questions. This suggests that the wording effect does not arise in the cognitive stages of question comprehension and attitude retrieval. Rereading a question and its answering options also takes the same amount of time, but happens more often for negative questions. This effect is likely to indicate a mapping difference: Fitting an opinion to the response options is more difficult for negative questions.

This dissertation is a study of prosody and anaphora. Its basic aim is to study the syntax and semantics of anaphora by using data from the prosody of sentences with reflexive and pronominal anaphors. In order to achieve that I first lay out a theory of focus, i.e. a theory of the correlation between the prosodic properties of an utterance and the informational status of its parts. The main innovation of the focus theory I argue for is that it equates the domains of focus with cyclic nodes in the syntactic derivation. This theory is applied on a wide range of examples with reflexive and pronominal anaphors. In the case of reflexive anaphors it is shown that no analysis that treats them like elements of type e can capture their prosodic properties. Instead, the prosody of reflexive anaphors requires that i. the alternatives to a focused reflexive anaphor are argument structure operations rather than individuals, and that ii. reflexive anaphors are functions that take relations as arguments. The second requirement is also shown to apply in the case of pronominal anaphors. Both reflexive and pronominal anaphors, then, are not arguments of verbs like other definite DPs. Crucially, however, in order to explain their prosodic properties pronominal anaphors do have to share a property with full DPs; like full DPs they have descriptive content that interacts with focus semantics. Other descriptive content of pronominal anaphors, however, like gender, does not interact with focus. This is shown to reveal something about the interaction of focus values with presupposition, namely that focus values are blind to presuppositional content.

This thesis investigates the temporal and semantic alignment between verbal language and hand gestures. The two forms of alignment are considered as two fundamental components in the process of joint interpretation of speech and gestures. Regarding temporal alignment, the main hypothesis is that prosodic cues (together with other linguistic parameters such as constituency) guide the first step of the joint interpretation of language and gestures, determining the restricted set of verbal items to which the gesture is related. The alignment is assumed to be determined by quasi-simultaneous peaks in the contours of prosody (considered as a combination of pitch, rhythm and amplitude) and hand activity (roughly the quantity of observable kinetic activity). The hypothesis is tested with two experiments. The first experiment measures the subjects' sensitivity to the temporal alignment (or absence thereof) between speech and gesture. The second experiment is designed to isolate the effect of prosodic cues on the successful integration of the two communicative channels. The results of the experiments support the hypothesis that prosodic contour is the main component in the determination of the 'point of attachment' of gestures in the linguistic structures they accompany. The alignment between the meaning conveyed by verbal language and gestures is reconstructed on the base of a formal theory of meaning. The analysis is restricted to the class of iconic gestures, and in particular to those used to represent spatio-temporal properties or events. The meaning of gestures is taken to correspond to a function that monotonically restricts the meaning of the accompanying verbal expression. The restriction takes the form of an intersection between the spatio-temporal extensional reference of the verbal component and the class of spatio-temporal structures that are iconically equivalent to the virtual space described by the gesture. Iconic equivalence is defined as that relation that holds between two spatial structures that present a similar set of spatial features. This relation is formally defined in the thesis as a logical language that allows the description of different levels of 'similarity' between spatial structures, corresponding to the different modes of representation observed in gestures. The thesis describes the use of the theory for the design and implementation of a computational prototype capable of producing verbal and gestural descriptions of simple physical scenes. With the help of the prototype two experiments have been developed, aimed at testing the validity of the hypothesis regarding the semantic alignment between speech and gestures. The first experiment measures the extent to which the specific form of iconicity proposed in the thesis captures the modes of representation observable in natural gestures. The second experiment tests instead some specific predictions of the theory, connected to the combinatorial properties of the two communicative channels. In both cases the experimental results corroborates the proposed analysis.

The central topic of this thesis is microvariation in the syntax of nominal constructions in Dutch. Specifically, the focus is on quantificational constructions, among which constructs with numerals, quantitative er, universal quantifier alle and partitives. Using a vast array of novel data from Dutch dialects collected through fieldwork, this thesis on the one hand investigates what Dutch microvariation reveals about the syntax of such constructions. It is shown how Dutch microvariation gains insight into the architecture of the noun phrase: it reveals the role and position of several functional projections in a fashion that could not be established on the basis of the standard language only. On the other hand, it is investigated what Dutch microvariation in these constructions reveals about the locus of syntactic variation. It is concluded that the aforementioned functional projections play a major role here. Specifically, the detected points of variation are (i) the morphological realization of functional heads and (ii) the possibility to move to functional projections. This thesis is of interest to anyone working on the syntax of the nominal domain, numerals, (quantitative) pronouns, (micro)variation and ECP phenomena

The central topic of this thesis is the automatic extraction of definitions from text. Definition extraction can play a role in various applications including the semi-automatic development of glossaries in an eLearning context, which constitutes the main focus of this dissertation. A glossary provides definitions for the most important terms that are discussed in a text. The semi-automatic extraction approach presented in this study consists of two phases. As a first step, a method entirely based on lexico-syntactic patterns has been used to distinguish between definitions and non-definitions. A corpus consisting of 600 definitions has been employed to identify recurrent definition patterns. Since many of these patterns are not unique to definitions, a second step was employed to reduce the number of non-definitions identified. It has been investigated whether other textual characteristics can contribute to the correct classification of definitions, in addition to the lexico-syntactic patterns. The properties that have been examined vary from the importance of the defined word (phrase) within a text to the layout of the definition. Machine learning techniques have been employed to identify which are the most relevant (combinations of) definition properties. The results of this dissertation are relevant for researchers in linguistics and lexicography as well as for the development of language technology applications

This dissertation challenges the standard view on indefinite articles as argument markers and determiners that block bare nominal arguments. It argues for a more fine-grained view in which articles are poor in their semantics but rich in their pragmatic competition with bare nominals and other determiners. On the basis of new data, both from synchrony and diachrony, historical simulations and detailed analyses within formal semantics and pragmatics (especially Optimality Theory) it shows how the merits of functionalism can be exploited without giving up the formal precision of the generative approach to linguistics. On the empirical side, it presents new analyses of Spanish unos, French des, bare partitives and the use of Dutch een and English a in predicate position. This dissertation will be of interest to all those working on the semantics, pragmatics and/or syntax of DPs, language evolution and/or the interaction between semantics and pragmatics

This dissertation is about the syntax and semantics of non-local dependencies. It focuses on wh-displacement and operator scope and addresses the challenge they pose to theoretical linguistics: How are form and meaning related and to which extent do syntax and semantics operate in parallel? A common approach of formal grammar theories is to assume that displacement and scope construal go hand in hand, and consequently to impose a strict correspondence between syntax and semantics. This approach is challenged, however, by a considerable amount of cases where the syntactic position of an operator expression does not coincide with its semantic scope position. This dissertation therefore pursues the opposite approach. It argues that syntactic displacement and semantic scope-taking do not interact at all. The overall picture that is developed is that grammar consists of two parts: a core system for establishing local dependencies, with syntax and semantics operating in parallel, and extensions to this core system for establishing non-local dependencies, with syntax and semantics operating independently. In developing this picture, it is shown that a substantial amount of data (such as pronunciation in wh-clusters, order preservation with multiple displacement, remnant movement and freezing, as well as different scopal behavior) can indeed be accounted for with a syntax and semantics that are largely independent from one another

This thesis investigates two issues. It studies the interpretations of sentences with plural arguments and it analyzes anaphoricity triggered by the presence of semantically plural nouns. Regarding the first point, it is standardly assumed that sentences with plural arguments can have many interpretations. This thesis discusses various psycholinguistic experiments, as well as two new questionnaire studies, which show in detail how the preferences among these interpretations depend on the type of noun phrase, including preferences that are commonly not acknowledged in the semantic literature. One goal of this thesis is to offer a semantic and pragmatic account which captures all the interpretations but can also explain why some are preferred over others. The explanation is used for studying the anaphoric expressions `same', `different', `others' and reciprocals. What these expressions share is their ability to be anaphoric sentence-internally if there is a semantically plural noun in the clause. However, the availability of this reading depends on the type of noun phrase used, in a similar way that preferences in the interpretations do. Based on these and other data, the thesis offers a novel formal semantic analysis of reciprocals and argues for a particular semantics of `same', `different' and `others'. It is shown what consequences the analysis has for our understanding of anaphoricity and plurality. This study is relevant for scholars working on binding, anaphora, and the semantics and pragmatics of pluralities. More generally, it is of interest to scholars in the field of semantics and pragmatics, as well as for linguists working on the syntax-semantics interface

In this dissertation, I attempt at defining some properties of the cognitive deficit underlying dyslexia by looking at the ability of dyslexic subjects to comprehend language. Dyslexia is a language-based learning disability, defined as a failure to learn to read properly despite normal intelligence, normal hearing, adequate classroom exposure, and absence of physical, emotional and socioeconomic problems. There is wide consensus that dyslexia is caused by difficulties related to the processing of phonological representations. There is also experimental evidence that dyslexic subjects experience difficulties in the comprehension of certain aspects of language, among which tough-sentences (such as ‘the bird is tasty to bite’), object extracted relative clauses, pronouns in Condition B configurations cross-sentential anaphora, and scalar implicatures. In this dissertation, I provide further experimental evidence that dyslexic subjects experience comprehension difficulties. In particular, dyslexic subjects are shown to differ significantly from their unimpaired peers in the comprehension of sentences containing a pronoun ambiguous between a bound and a coreferential reading, imperfective aspect, and universally quantified noun phrases in quantifier spreading contexts. To account for the data, I propose that dyslexia is associated with a verbal working memory deficit, where the term ‘verbal working memory refers to the component of the Working Memory system that is responsible for the temporary storage of verbal information, when required in order to perform further linguistic computations. The proposal accounts bith for the reading difficulties and the language comprehension difficulties. It also accounts for the fact that the language related difficulties experienced by dyslexic subjects are mostly found in those domains of language that are also problematic for preschool children. The results achieved provide support for a type of linguistic theorizing that stresses the interdependence between linguistic computations and processing resources. In particular, I promote a type of research whose goal is to adequately describe linguistic computations as implemented in a resource-limited system. This line of research has significant implications for deciding between competing linguistic theories, individuating the cognitive engines that support linguistic computations, and defining the cognitive profile of subjects with a language impairment

In my study, I investigate the optional omission functional elements in children’s early production. Several findings indicate that functional categories are present in child grammar in the early stages of language acquisition (Bottari, Cipriani and Chilosi 1993/4; Gerken and McIntosh 1993). However, initially functional categories are only optionally realized (Guasti 2002). In my study, I target the acquisition of a specific set of function words: object clitics. I provide an extensive investigation of object clitic omission in early Italian. Initially, clitics are optionally omitted in obligatory contexts. Object clitic omission raises a number of questions concerning the nature of null objects in early grammars, the relation between argument structure and referentiality, the acquisition of syntactic and pragmatic aspects of referentiality, and the mapping between syntactic and prosodic structures. In particular, I address the following questions: 1. Does early Italian allow referential null objects? 2. Is clitic omission affected by the phonological context in which a clitic occurs? 3. Are clitics omitted because they refer to information that is easily retrievable from the preceding discourse? Three experiments aim at shedding light on these issues. The first experiment targets children’s interpretation of null objects in a receptive task, to investigate whether early Italian, differently from the adult grammar, allows referential null objects in the absence of an overt clitic. The results of my study do not support the hypothesis that children’s representation of null objects deviates from the target grammar. The second experiment explores the hypothesis that clitic omission is an interface phenomenon resulting from difficulties in the integration of syntactic and phonological information. I show that clitic omission can be partially accounted for by children’s tendency to omit unstressed syllables (such as clitics) when they do not belong to a (strong-weak) trochaic foot (Demuth 1995; 2007, Gerken 1991; 1996). The third experiment investigates object clitic omission at the syntax-discourse interface. This approach predicts that object drop should be restricted to pragmatically acceptable contexts, i.e. contexts in which the referent at hand is easily retrievable from the preceding discourse/context (Serratrice, Sorace & Paoli 2004). Clitics fulfill this requirement, since they are used for highly accessible referents (Ariel 1990). I show that children are sensitive to discourse in their omission patterns: they omit the object if the referent is highly accessible in the preceding discourse (in clitic contexts) but not when a more informative referring expression (a lexical object) is required. I propose that children have problems in integrating the syntactic requirement of an overt clitic with the pragmatic principle of “informativeness” (Greenfield and Smith 1976), which allows highly accessible information to be dropped. I conclude that clitic omission is an interface phenomenon triggered by the non adult-like integration of different aspects of linguistic knowledge.