Volume 17, Issue 2

We use algorithmic learning and statistical methods over a form frequency list (compiled from the Hungarian web corpus) to investigate variation in Hungarian verbal inflection. Our aims are twofold: (i) to give an adequate description of this variation, which has not been described in detail in the literature and (ii) to explore the range and depth of lexical attractors that potentially shape this variation. These attractors range from closely related ones, such as the shape of the word form or the behaviour of the verb’s paradigm, to broad ones, such as the behaviour of similar verbs or the phonotactics of related verb forms. We find that verbal variation is predominantly determined by similarity to related verb forms rather than by word shape or by word frequency. What is more, the effect of similarity is better approximated using inflected forms as opposed to base forms as points of comparison. This, in turn, supports a rich memory model of morphology and the mental lexicon.

This paper presents a diachronic analysis of the attributive uses of four synonymous adjectives which designate the concept of sweet-smelling ( fragrant, perfumed, scented , and sweet-smelling ) in the latter part of Late Modern and Present-day American English. By drawing on data from the Corpus of Historical American English ( COHA ) and applying a Hierarchical Configural Frequency Analysis (HCFA), it delineates the internal semantic structure of this set of synonyms, paying special attention to their noun collocates. The results show that the concept of sweet-smelling experiences major changes over the time span examined (1850–2009), from being used mostly to qualify entities which can exhibit a natural pleasant smell (e.g. flowers and trees) to modifying objects which are artificially sweet-smelling (e.g. oils and shampoos). Moreover, fragrant and perfumed , which initially were the most frequent adjectives, are gradually replaced by scented , thus reflecting a change in the relation between the synonyms over time. The study constitutes the first diachronic approximation to synonymy from the perspective of cognitive semantics and provides equally effective results as previous synchronic research in the field.

This paper is part of a larger research effort on language variability aimed at uncovering the relations between extra- and intratextual characteristics of Czech texts by means of multi-dimensional analysis. The palpable lack of prior art on quantitative register analysis of Czech led to several distinctive methodological decisions, concerning namely corpus design, feature selection and the parameters of factor analysis, especially the number of dimensions to extract. We report on these for their potential relevance to other researchers embarking on a similar journey. In order to demonstrate the viability of the model, we also present a brief interpretation of the resulting dimensions.

This article utilises an innovative, information-theoretic metric to assess complexity variation across written and spoken registers of British English. This is novel because previous research on language complexity mainly analysed complexity variation in typological data, single language case studies or geographical varieties of the same language. The measure boils down to Kolmogorov complexity which can be conveniently approximated with off-the-shelf compression programs. Essentially, text samples that can be compressed more efficiently count as linguistically simple. The dataset covers a wide range of traditional written and spoken registers (e.g. broadsheet newspapers, courtroom debate or face-to-face conversation), as sampled in the British National Corpus . It turns out that Kolmogorov-based register variation coincides with register formality such that informal registers are overall and morphologically less complex than more formal registers, but more complex in regard to syntax (defined here as rigid word order). Generally, the results show that written and spoken registers vary along a continuum, and significantly trade-off morphological against syntactic complexity (and vice versa). Finally, the findings support proposals to view language as a complex adaptive system and demonstrate how language adapts to the situational context of language production and functional-communicative needs of its users.

The Chinese antonymous compounds shangxia ‘up-down’, qianhou ‘front-back’ and zuoyou ‘left-right’ can convey similar meanings as approximatives, indicating ‘about’ or ‘around’. Based on data from the CCL (Centre for Chinese Linguistics) corpus and adopting a behavioral profile (BP) approach, this study investigates the semantic and usage differences between the three approximatives. The corpus analysis yields the following findings. First, the most frequently used of the three approximatives is zuoyou , followed by qianhou and shangxia , in that order. Second, qianhou mainly approximates time; zuoyou approximates quantity; and shangxia , age. Third, in approximating age, time and quantity, the three approximatives exhibit subtle behavioral preferences. Based on these corpus findings, I discuss the motivations underlying the similarities and differences between the three synonyms. I argue that their similarities are explained by the metaphor QUANTIFICATION IS SPACE and their differences are explained by the different kinds of quantification they are used to approximate.

This is a pilot study of usability of Context Specificity measure for stylometric purposes. Specifically, the word embedding Word2vec approach based on measuring lexical context similarity between lemmas is applied to the analysis of texts that belong to different styles. Three types of Czech texts are investigated: fiction, non-fiction, and journalism. Specifically, forty lemmas were observed (10 lemmas each for verbs, nouns, adjectives, and adverbs). The aim of the present study is to introduce a concept of the Context Specificity and to test whether this measurement is sensitive to different styles. The results show that the proposed method Closest Context Specificity ( CCS ) is a corpus size independent method which has a promising potential in analyzing different styles.

This paper presents a corpus-based quantitative study on linguistic proficiency of approx. 300 immigrant and heritage speakers of Danish in North America and Argentina, aiming at the question whether linguistic proficiency is connected to ‘immigrant generation’ (i.e. the difference between speakers who migrated as adults with a fully acquired language competence and foreign-born heritage speakers) or the sociocultural setting, or both. The large data base at hand provides a rare opportunity to compare developments within the same minority language in different places, representing different sociocultural settings for the immigrant or heritage speakers and, accordingly, different language ecologies. The study relies on the Corpus of American Danish (1.6 million tokens, including both words and non-word utterances). Based on this data set, the paper explores the distribution of 13 linguistic and non-linguistic variables representing linguistic proficiency (i.e. Danish words, L2 words, word-internal codeswitching, type-token ratio, empty and filled pauses, self-interruption, lengthening, speech rate, word length, runlength and the ratio of main and subclauses) by applying Factor Analysis as a statistical tool. On an empirically solid basis, the paper concludes that (a) the sociolinguistic setting is the crucial factor in the development of linguistic proficiency and (b) linguistic proficiency is a non-universal cognitive phenomenon.

One of the goals in Cognitive Linguistics is the automatic identification and analysis of constructions, since they are fundamental linguistic units for understanding language. This article presents DISCOver, an unsupervised methodology for the automatic discovery of lexico-syntactic patterns that can be considered as candidates for constructions. This methodology follows a distributional semantic approach. Concretely, it is based on our proposed pattern-construction hypothesis: those contexts that are relevant to the definition of a cluster of semantically related words tend to be (part of) lexico-syntactic constructions. Our proposal uses Distributional Semantic Models for modelling the context taking into account syntactic dependencies. After a clustering process, we linked all those clusters with strong relationships and we use them as a source of information for deriving lexico-syntactic patterns, obtaining a total number of 220,732 candidates from a 100 million token corpus of Spanish. We evaluated the patterns obtained intrinsically, applying statistical association measures and they were also evaluated qualitatively by experts. Our results were superior to the baseline in both quality and quantity in all cases. While our experiments have been carried out using a Spanish corpus, this methodology is language independent and only requires a large corpus annotated with the parts of speech and dependencies to be applied.