Lexical change

We present Sense Clustering over Time (SCoT), a novel network-based tool for analysing lexical change. SCoT represents the meanings of a word as clusters of similar words. It visualises their formation, change, and demise. There are two main approaches to the exploration of dynamic networks: the discrete one compares a series of clustered graphs from separate points in time. The continuous one analyses the changes of one dynamic network over a time-span. SCoT offers a new hybrid solution. First, it aggregates time-stamped documents into intervals and calculates one sense graph per discrete interval. Then, it merges the static graphs to a new type of dynamic semantic neighbourhood graph over time. The resulting sense clusters offer uniquely detailed insights into lexical change over continuous intervals with model transparency and provenance. SCoT has been successfully used in a European study on the changing meaning of ‘crisis’.

Даследуецца развіццё беларускай літаратурнай мовы 1920–1930-х гг. у яго сувязях з развіццём прэскрыптыўнага адгалінавання беларускай лінгвістыкі. З выкарыстаннем матэрыялаў перыядычнага друку і архіваў аналізуюцца акалічнасці фарміравання беларускага мовазнаўства, асэнсоўваюцца літаратурная мова як «рукатворная» з’ява і асаблівасці вывучэння яе гісторыі. Аглядаецца літаратура па гісторыі беларускай лінгвістыкі і літаратурнай мовы міжваеннага перыяду. Даследуецца моўнае ўжыванне 1920–1930-х гг. з асаблівай увагай да зменаў у моўным развіцці. Асэнсоўваецца рэформа беларускай мовы 1933 г. як паваротны пункт у беларускім мовазнаўстве і беларускай літаратурнай мове. Вывучаецца гісторыя шэрагу слоў з мэтай выяўлення зрухаў у рамках сінанімічных радоў і ўзнікнення новых, сучасных стылістычных іерархій.

This paper examines the mechanisms of semantic change in the creation of ten Cantonese slang words. It demonstrates with synchronic evidence that metaphorisation, metonymisation and (inter)subjectification are three principal driving forces behind the shift in meaning. It is argued that Traugott and Dasher's (2002) Invited Inferencing Theory of Semantic Change (IITSC), initially proposed for and widely used in the context of grammati-calisation, is equally useful for the study of neologisms-in this case, the relatively recent slang expressions in Cantonese. These monosyllabic lexemes are shown to have followed the same unidirectional pathway of semantic change-that is, the shift from non-subjective meaning to encoded (inter)subjective meaning-outlined in their model of semantic change.

Este artículo muestra que la introducción de términos del inglés especialmente desde el campo de la informática y la tecnología de la información y la comunicación ha dado como resultado el uso inapropiado de términos y expresiones que si bien es cierto son de uso común sin embargo, en la mayoría de los casos abusa de la semántica y la gramática del español. Verbos que normalmente son intransitivos en español pasan, mediante el uso (como acuñamiento, préstamo o calco) a ser transitivos con los concomitantes problemas de significación. Se proporciona una lista de estos términos algunos de los cuales (calco) ya son de uso necesario pues no existe un equivalente en la lengua materna.

Dintre cele trei momente din „viaţa” unui cuvânt, naşterea, vieţuirea şi moartea, descrise în detaliu în volumul introductiv al colecţiei Viata Cuvintelor (Marius Sala "101 cuvinte mostenite, imprumutate si create"), în cartea de faţă, am avut în vedere un moment din etapa „vieţuire”, adică una dintre întâmplările care intervin în perioada existenţei în limbă a unor cuvinte şi expresii. Acestea sunt surprinse în momente diferite ale vieţuirii: imediat după intrarea în limbă, când sunt primite cu ostilitate de lingvişti, dar cu multă prietenie de publicul larg, în momentul în care îşi primenesc sensurile sau preferinţele de combinare, după o (re)întâlnire spectaculoasă cu perechile lor din alte limbi, pe care adesea le imită, în momentul lungirii prin derivare sau al scurtării prin trunchiere, al combinării nefericite cu o rudă care înseamnă acelaşi lucru sau cu una care înseamnă exact opusul, atunci când vorbitorii le înlocuiesc cu alte cuvinte cu care seamănă, când le înţeleg şi le interpretează greşit, când le folosesc excesiv, astfel încât ele ajung să-şi piardă sensul.

Word-recognition tests for second language learners have become more common in recent years with increased awareness of their importance in lexical processing. However, the very short reaction times involved mean the associated technology is complicated to handle and set up in school-based testing situations. Our solution to this problem is a quick-and-easy test called „Q_Lex‟. The most innovative feature is that the stimuli are masked in nonsense strings; e.g. <pajlchanceacdut>. This approach slows down recognition time to a degree that PCs can reliably measure (1 or 2 seconds). Native speakers have little trouble identifying the hidden words and so their latencies can be used as a baseline to create norms for each item. Learners score points if they respond within these norms. In this way, Q_Lex produces a score rather than a response time result. The target words are all high-frequency and a large number (typically 50) can be tested in a few minutes.
In this thesis, the approach to word recognition testing described above is first set within the literature on second-language word recognition. 9 papers are reviewed. Two areas are addressed: the precise measurement of vocabulary recognition speed and the measurement of lexical accessibility. Beyond careful experimental design, it is useful to try to understand what makes some tests better than others. Simple design, rapid completion and adequate vocabulary coverage are very important factors if the aim is to reflect the state of development of the second language lexicon rather than test individual words.
In the next part, I will go through a cycle of testing the reliability and validity of Q_Lex from the prototype version 1.1 to version 3.01. The work focuses on areas including: the consistency of scores between two tests with varying intervals; the improvement of items and the test interface, specifically the effect of the mask on finding the hidden word; the comparison of similar item sets for evidence of parallel forms reliability; issues connected with the interpretation of test-takers‟ performance, especially with regard to the question of variability in reaction time. Gradually I will build up a picture of a test gaining in construct validity and reliability whilst taking heed of the work of experts on testing and theorists in second language acquisition theory. One weakness of Q_Lex is the recurrent evidence of test-taking strategies to account for some of the scores of many individuals. For this reason, in Chapter 9, the thesis moves to a more qualitative perspective to investigate exactly what happens as people take the test. The findings here provide additional useful pointers for further development of the format in future work.
In the discussion chapter, the threads are brought together. An important issue concerns the “co-efficient of variation” index for cognitive fluency which cannot be calculated from standard Q_Lex data. The reasons for this are elucidated, and an alternative is suggested.

Resumen
Nuestra intención es medir la riqueza léxica de los hablantes a partir de un planteamiento original que consiste en la estimación del tamaño virtual del vocabulario de los individuos. Para ello, usamos un algoritmo de optimización aleatoria llamado Simulated Annealing que nos permite representar una función a partir de la cual obtener un modelo de la estimación léxica muy fiable. Puesto que este método es compatible con otros procedimientos tradicionales usados para la estimación de la riqueza léxica, comparamos nuestros resultados con otros procedentes de diferentes investigaciones con la intención de obtener los patrones sociolingüísticos que condicionan la variación de la riqueza léxica.

Abstract
We intend to explore the use that speakers make of the Spanish lexical system in oral exchanges. Our intention is to measure the lexical richness of the population. We shall use an original approach which estimates the virtual size of vocabulary of individuals. We shall apply an algorithm for random optimization called Simulated Annealing. This procedure allows us to represent a function from which we obtain a very reliable pattern for lexical estimation. This method is compatible with other traditional procedures used for the estimation of lexical richness. We shall compare our results with others collected in previous investigations, thus establishing the socio-linguistic patterns which could have an influence on the variation of lexical richness.

The Albanian extended family includes more generations, elder people,
parental couples, all unmarried children and married sons with their wives
and children living in the same household. The organization and the function
of the extended family take a central place in the Albanian Customary Law
(often referred to as Kanun). Its origins reach far back in antiquity. In
Northern Albania it has survived in its classical form until the first half of the
20th century.
Alb. motër (older motërë) ‘sister’ has been deemed as a reflex of the Pre-
Indo-European word for MOTHER in nearly all other Indo-European
languages. The semantic change MOTHER → SISTER has been long discussed,
reaching no convincing conclusions. Recently M. Majer (2019) has put
forward a keen and well argued proposal, reconstructing a PIE derivative
*meh2t(e)reh2 ‘of the mother’ as the preform of motër(ë). Such a
reconstruction fits well in a hypothesis that takes into account the structural
frame of the Albanian extended family as described above, as well as the
disputed origin of vajzë ‘girl’ from PIE *swesor- ‘sister’.

Cet article analyse environ un millier d’innovations lexicales (IL) du français contemporain à base nom propre, validées sur un corpus de presse contemporain. Après les jalons théoriques, il présente le corpus, la méthode de validation et les outils automatiques pour découvrir les propriétés de ces unités. Il détaille enfin les IL, leurs domaines de référence, les mécanismes de formation néologiques, leurs caractéristiques linguistiques et socio-pragmatiques et leur évolution depuis 2015.

In this short article, our aim is to replicate Facebook's recent study on the varieties of laughter in new communication media (CMC). Even when taking into account the differences due to the different corpora (Facebook vs. SMS), the change in the use of e-laughter type is striking. In 2004, lol was, even for French-speaking people, the most common way of e-laughing (0.71). Nowadays, it is widely considered as "old fashioned" in both English and French-speaking regions.

I show that several quantitative properties of various Russian constructions with "dative subjects" have been changing during the last three centuries. These properties include explicitness of the dative subject, its position in the clause, coocurrence with embedded infinitives, etc. Although these properties are logically independent (and only obliquely related to the problem of subjecthood) I argue they conspire to a unitary trend towards higher salience of the dative argument. Interestingly, this trend covers heterogeneous constructions with dative subjects. Ultimately, I argue that this is a scenario of grammatical change that is propelled by lexical mechanisms.

Vlashki/Zheyanski, or Istro-Romanian, is a severely endangered Eastern Romance language spoken on the Istrian peninsula in Croatia. The language is now spoken by less than 120 fluent and active speakers, most of them over the age of 50. The Vlashki/ Zheyanski-speaking communities are in the midst of the language shift to Croatian. Since cohabitation, population mix and bilingualism have characterized these communities for centuries, as a result of intense and long-term contact pressures, the language has undergone many lexical and structural changes under the influence of Croatian. In the lexical domain, the language has adopted many cultural borrowings, but also a significant number of core borrowings. The paper explores the extent of contact influence in the basic vocabulary of body part terms, a section of the lexicon assumed to be one of the most resistant to contact interference. Lexical equivalents of 46 body part meanings found in three different historical sources and in aut...

In this contribution we present a use case of the application of big language data and digital methods such as natural language processing, machine learning, and network analysis in the fields of digital humanities and linguistics by characterizing and modeling the diachronic dynamics of lexical networks. The proposed analysis will be based on two corpora containing 20 years of data with billions of tokens. 2012 ACM Subject Classification Human-centered computing → Social network analysis; Computing methodologies → Natural language processing; Computing methodologies → Machine learning

The aim of this paper is to explore the importance of the ludic dimension for linguistic innovations by combining synchronic and diachronic analyses of lexicographic sources from French and by reinterpreting the data from a usage-based perspective. I will discuss the possibilities and methodological challenges in tracing ludicity in the lexicon, taking into account contemporary and historical dictionaries, most importantly Le Petit Robert 2016 and different editions of the Dictionnaire de l’Academie francaise. Moreover, I will analyse how innovations are introduced and perceived by speakers, distinguishing different subtypes of innovation based on structural, semantic, and pragmatic features. Finally, I will turn to the diachronic evolution of ludic innovations in order to identify general tendencies and pathways of evolution and argue that markedness plays a key role for ludic innovation, which represents an important, albeit so far neglected, domain of lexical dynamics.

El análisis diacrónico e histórico del léxico verbal relacionado con el movimiento desde una perspectiva semántica es un ámbito de estudio que ha demostrado ser enormemente productivo (vid. Company 2004, Crego 2002, Dworkin 1992, Eberenz 1998, Melis 2006, y Paz en prensa, entre muchos otros).
No obstante, no todos los verbos de movimiento han sido igualmente abordados. Mientras que ir o venir han sido ampliamente analizados, otras unidades de este grupo no han recibido tanta atención por parte de las investigaciones diacrónicas. Este el caso de entrar, la unidad lingüística en la que se centra este trabajo, el cual persigue dos objetivos principales. Por un lado, se pretende realizar una aproximación a la realidad semántica de este verbo en el siglo XIII, la etapa más antigua del español de la que se dispone de un número suficientemente significativo de documentaciones y un periodo clave para el desarrollo léxico y semántico del español como lengua. Por otro lado, este estudio busca poner de manifiesto los vínculos existentes en esa época entre los diferentes valores semánticos del verbo con el apoyo del marco teórico de la semántica cognitiva y, más concretamente, mediante la Teoría de los prototipos (Geeraerts 1997). A través del análisis de 205 documentaciones procedentes del Corpus del Nuevo Diccionario Histórico (CNDH), es posible extraer conclusiones reveladoras acerca de la presencia y organización de los significados nucleares de entrar frente a los de carácter más periférico y su relevancia para el posterior desarrollo del verbo, tal y como han demostrado estudios previos dedicados al catalán, (Montserrat 2007), el portugués (Soares da Silva 1999) o al propio español (Fernández 2006, Paz 2012).
De esta forma, la presente investigación pretende arrojar algo de luz sobre la historia de una voz patrimonial y básica del español como es entrar a la que no se han dedicado estudios monográficos en profundidad y, a su vez, demostrar cómo una herramienta teórica de corte cognitivo como la Teoría de los prototipos puede aportar información de gran relevancia para la investigación diacrónica e histórica del léxico.

Classical Arabic forms the basis of Arabic linguistic theory and it is well understood by the educated Arabic reader. It is different in many ways from Modern Standard Arabic which is more simplified in its lexical, syntactic, morphological, phraseological and semantic structure. King Saud University Corpus of Classical Arabic is a pioneering corpus of around 50 million words of Classical Arabic. It is initially constructed for the purpose of studying distributional lexical semantics of the Quran and Classical Arabic, however, it is designed in a general way making it also appropriate for other researches in Linguistics and Computational Linguistics. In this paper, we will briefly describe the structure of our corpus, and then we will demonstrate how it can be used to depict some aspect of Arabic language change between the classical and the modern periods.

The Albanian extended family includes more generations, elder people, parental couples, all unmarried children and married sons with their wives and children living in the same household. The organization and the function of the extended family take a central place in the Albanian Customary Law (often referred to as Kanun). Its origins reach far back in antiquity. In Northern Albania it has survived in its classical form until the first half of the 20th century. Alb. moter (older motere) &#39;sister&#39; has been deemed as a reflex of the Pre-Indo-European word for MOTHER in nearly all other Indo-European languages. The semantic change MOTHER → SISTER has been long discussed, reaching no convincing conclusions. Recently M. Majer (2019) has put forward a keen and well argued proposal, reconstructing a PIE derivative *meh2t(e)reh2 &#39;of the mother&#39; as the preform of moter(e). Such a reconstruction fits well in a hypothesis that takes into account the structural frame of the Alba...

Vlashki/Zheyanski, or Istro-Romanian, is a severely endangered Eastern Romance language spoken on the Istrian peninsula in Croatia. The language is now spoken by less than 120 fluent and active speakers, most of them over the age of 50. The Vlashki/ Zheyanski-speaking communities are in the midst of the language shift to Croatian. Since cohabitation, population mix and bilingualism have characterized these communities for centuries, as a result of intense and long-term contact pressures, the language has undergone many lexical and structural changes under the influence of Croatian. In the lexical domain, the language has adopted many cultural borrowings, but also a significant number of core borrowings. The paper explores the extent of contact influence in the basic vocabulary of body part terms, a section of the lexicon assumed to be one of the most resistant to contact interference. Lexical equivalents of 46 body part meanings found in three different historical sources and in authors' own field data are compared. The comparison shows that a significant number of body part words has been borrowed from dialectal Croatian, that the number of loanwords has doubled since the beginning of the 20 th century, that standard Croatian has become a factor of contact influence recently, and that the Zheyanski dialect of the language has preserved a greater number of native body part words.

Classical Arabic forms the basis of Arabic linguistic theory and it is well understood by the educated Arabic reader. It is different in many ways from Modern Standard Arabic which is more simplified in its lexical, syntactic, morphological, phraseological and semantic structure. King Saud University Corpus of Classical Arabic is a pioneering corpus of around 50 million words of Classical Arabic. It is initially constructed for the purpose of studying distributional lexical semantics of the Quran and Classical Arabic, however, it is designed in a general way making it also appropriate for other researches in Linguistics and Computational Linguistics. In this paper, we will briefly describe the structure of our corpus, and then we will demonstrate how it can be used to depict some aspect of Arabic language change between the classical and the modern periods.

The aim of this paper is to explore the importance of the ludic dimension for linguistic innovations by combining synchronic and diachronic analyses of lexicographic sources from French and by reinterpreting the data from a usage-based perspective. I will discuss the possibilities and methodological challenges in tracing ludicity in the lexicon, taking into account contemporary and historical dictionaries, most importantly Le Petit Robert 2016 and different editions of the Dictionnaire de l'Académie française. Moreover, I will analyse how innovations are introduced and perceived by speakers, distinguishing different subtypes of innovation based on structural, semantic, and pragmatic features. Finally, I will turn to the diachronic evolution of ludic innovations in order to identify general tendencies and pathways of evolution and argue that markedness plays a key role for ludic innovation, which represents an important, albeit so far neglected, domain of lexical dynamics. * * 1 Introduction: Ludicity as an important factor for lexical innovation If we compare the target language equivalents of English V.I.P. (very important person) in French and German, we can observe an interesting difference: while the French dictionary Le Petit Robert 2016 (= PR 2016) indicates that this form is used in familiar French and in ludic contexts (see the lexicographic marks "fam[ilier]" and "plaisant"), the German dictionary Duden does not indicate any special value for this item. (1) (French) V. I. P. [veipe; viajpi] nom invariable ETYM. avant 1959 ◊ sigle anglais de Very Important Person «personne très importante» FAM. et PLAISANT Personnalité de marque. Une V. I. P. (PR 2016) || * Some of the reflections of this paper and the following paper were presented in a "tandem talk" given together with Claudine Moulin at a linguistic colloquium at Trier University in December 2016. We would like to thank our colleagues at Trier for their valuable feedback. Moreover, we would like to thank Peter Kühn and two anonymous reviewers for their very helpful comments and suggestions. In addition, our thanks go to Martina Bross and Angela Oakeshott for the stylistic revision of our papers.

We present Sense Clustering over Time (SCoT), a novel network-based tool for analysing lexical change. SCoT represents the meanings of a word as clusters of similar words. It visualises their formation, change, and demise. There are two main approaches to the exploration of dynamic networks: the discrete one compares a series of clustered graphs from separate points in time. The continuous one analyses the changes of one dynamic network over a time-span. SCoT offers a new hybrid solution. First, it aggregates time-stamped documents into intervals and calculates one sense graph per discrete interval. Then, it merges the static graphs to a new type of dynamic semantic neighbourhood graph over time. The resulting sense clusters offer uniquely detailed insights into lexical change over continuous intervals with model transparency and provenance. SCoT has been successfully used in a European study on the changing meaning of 'crisis'.

We present Sense Clustering over Time (SCoT), a novel network-based tool for analysing lexical change. SCoT represents the meanings of a word as clusters of similar words. It visualises their formation, change, and demise. There are two main approaches to the exploration of dynamic networks: the discrete one compares a series of clustered graphs from separate points in time. The continuous one analyses the changes of one dynamic network over a time-span. SCoT offers a new hybrid solution. First, it aggregates time-stamped documents into intervals and calculates one sense graph per discrete interval. Then, it merges the static graphs to a new type of dynamic semantic neighbourhood graph over time. The resulting sense clusters offer uniquely detailed insights into lexical change over continuous intervals with model transparency and provenance. SCoT has been successfully used in a European study on the changing meaning of ‘crisis’.

Vlashki/Zheyanski, or Istro-Romanian, is a severely endangered Eastern Romance language spoken on the Istrian peninsula in Croatia. The language is now spoken by less than 120 fluent and active speakers, most of them over the age of 50. The Vlashki/ Zheyanski-speaking communities are in the midst of the language shift to Croatian. Since cohabitation, population mix and bilingualism have characterized these communities for centuries, as a result of intense and long-term contact pressures, the language has undergone many lexical and structural changes under the influence of Croatian. In the lexical domain, the language has adopted many cultural borrowings, but also a significant number of core borrowings. The paper explores the extent of contact influence in the basic vocabulary of body part terms, a section of the lexicon assumed to be one of the most resistant to contact interference. Lexical equivalents of 46 body part meanings found in three different historical sources and in aut...