Language work of all kinds relies on primary data like recordings and transcripts, but the link to this data is rarely made transparent in finished products. This paper provides an introduction to The Tromsø Recommendations For Citation of Research Data in Linguistics (Author et al., 2019), a document that is designed to help language researchers properly cite language data, and give proper credit to everyone who contributed to a language documentation project. In this presentation we discuss the motivations behind the recommendations, and provide an introduction to the format of citation in the recommendations. We conclude with practical ways that you can improve data citation in your own practice and in your language or research community.

This paper aims to study on the impact of proper nouns and how they show their impact on usage or how they influence the society because of its bad collision in the environment. A few examples are taken for the study to see how proper nouns dealt with in some demanding situations. Also, it has been resourced in the OER that proper nouns influence the people and sometimes results in chaos when it causes disastrous conditions to the society as it has been experiencing by the society due to pandemic. This is to mainly focused on the following research questions:
Does a name show its impact in society?
Do the names of all the languages vulnerable to the impact?
How do the people percept about a word of a language?
Given the pandemic, strong evidences are provided and support has been sought from Millianism to people perception about language. Per Unger, 1983 the proper nouns are marginal elements of a language yet, they aren’t astray in intensifying the society at times. It has been proved that the semantic representation doesn’t contribute and control the present perceptions of people using multiple illustrations happened around the globe.

Este trabalho insere-se no projeto "Comparação entre Línguas via Frames e Construções: recursos multilíngues e tradução automática", desenvolvido no Laboratório FrameNet Brasil de Linguística Computacional – FN-Br – (SALOMÃO, 2009) em parceria com a Heinrich-Heine Universität Düsseldorf e a Leipzig Univerität. O projeto tem por objetivo desenvolver um sistema de tradução por máquina que utilize um recurso computacional léxico-construcional (TORRENT ET AL., 2018) para melhorar o resultado de sistemas de tradução por máquina. Nesse contexto, este trabalho tem por objetivo apresentar a descrição e a modelagem de construções de Pretérito Imperfeito em Português Brasileiro (PB) e em Espanhol.
A descrição e modelagem das construções de Pretérito Imperfeito em PB e em Espanhol se faz necessária de modo a garantir que as traduções buscadas pelo usuário entre essas línguas sejam mais adequadas do que as traduções fornecidas por outros aplicativos de tradução por máquina. No primeiro estágio da pesquisa fizemos o levantamento bibliográfico acerca das construções de Pretérito Imperfeito em PB e em Espanhol. Para atestar as ocorrências dessas construções, se fez necessário constituir um banco de dados específico desses tipos de construções para ambas as línguas, o qual apresenta dados escritos provenientes do uso linguístico. Dessa forma, criamos um corpus específico para as construções de Pretérito Imperfeito em PB e outro em Espanhol compostos por dados em língua natural. Feito o levantamento, passamos a uma proposta de descrição das construções de Pretérito Imperfeito em PB e em Espanhol com base na Gramática das Construções de Berkeley (KAY & FILLMORE, 1999), através da elaboração de Matrizes de Atributo e Valor. Posteriormente, tais análises foram convertidas em modelos linguístico-computacionais os quais foram armazenados na base de dados da FN-Br, na forma de construções. Para tanto, utilizamos a ferramenta FrameNet Brasil WebTool 3.0. Os dados armazenados serão futuramente incorporados ao sistema de pós-edição do tradutor.

In Aotearoa/New Zealand there are three official languages. One of them is an Indigenous language called Te Reo Māori (Eastern Polynesian language group). Since the 1980s there have been strenuous efforts to revitalise this endangered indigenous language (Te Taura Whiri i Te reo Māori, 2020). A recent census shows that 3% of the New Zealand population can speak Te Reo Māori (Keegan, 2013). Bilingual picturebooks are books featuring the text of the story in two languages. In New Zealand there are increasing numbers of English and Māori bilingual picturebooks. The Linguistic Landscape of these books have been analysed to explore the language hierarchies which are being presented to our children (Daly, 2017). Some books give English text first, and in larger font. This reflects the status quo of the English dominant society in Aotearoa/New Zealand. Other bilingual picturebooks give Māori first and larger to tautoko or support the revitalisation of Te Reo Māori and to change the existing language hierarchy. In this presentation I analyse a sample of Māori-English bilingual picturebooks, and suggest that the Linguistic Landscape of bilingual picturebooks are powerful tools for changing language hierarchies and for supporting the revitalisation of indigenous languages.

References

Daly, N. (2017). Audiences, referees and landscapes: Understanding the use of Māori and English in New Zealand Dual Language picturebooks through a sociolinguistic lens. In Marra, Meredith and Paul Warren (eds) Linguist at Work: Festschrift for Janet Holmes. Wellington: VUP
Keegan, P. (2017). Māori Langauge Information. Retrieved on 13 October 2020 from http://www.maorilanguage.info/mao_lang_faq.html#:~:text=A3.,all%20people%20living%20in%20NZ).
Te Taura Whiri i Te Reo Māori (2020). What we do. Retrieved on 13 October 2020 from
https://www.tetaurawhiri.govt.nz/en/services/

Na literatura, há dois tipos de nominalizações que são amplamente estudados:

(1) A construção da casa levou seis meses.
(2) A construção branca é muito alta.

Em (1), há um Nome de Eventualidade, com uma leitura de que o evento de construir a casa durou seis meses. Já em (2), temos um Nome de Entidade, que referencia um objeto no mundo.
Como visto, uma mesma palavra pode ser usada em diferentes contextos, com diferentes leituras. Os estudos construcionistas possuem grande interesse nesse fenômeno. Para explicá-lo, argumenta-se que os Nomes de Eventualidade são derivados de verbos, dos quais herdam a leitura eventiva e a grade argumental; e os Nomes de Entidade são nominalizações diretas da raiz.
Dados do português, contudo, confrontam essa visão. Isso porque há inúmeras palavras que contam com morfemas tipicamente verbais, mas apresentam leitura de entidade. Por exemplo, há os morfemas -fic- em “O guarda me entregou a notificação” e -iz- em “O infrator pagou a indenização”.
A partir disso, as perguntas que guiam este trabalho são as seguintes: (i) se Nomes de Entidade também podem derivar de verbos, por que parecem não herdar a estrutura argumental?; e (ii) é possível identificar alguma regularidade no processo de construção de referência desses nomes?
Para tentar responder a essas perguntas, analisamos as nominalizações em -ção do português brasileiro. Nessa empreitada, adotamos o arcabouço teórico da Morfologia Distribuída, do quadro da Teoria Gerativa. A principal proposta do modelo é de que as palavras não são formadas no léxico, por meio de regras específicas; na verdade, a formação de palavras está sujeita às mesmas operações envolvidas na formação de sentenças.
Como ponto de partida, analisamos o processo de criação envolvido nos Nomes de Eventualidade – transpondo um raciocínio já explorado pela literatura no campo verbal. Em “A construção da casa levou seis meses”, a casa é criada pelo processo de construção. Já em “A documentação dos pacientes levou duas horas”, o que é criado pelo processo de documentação não são os pacientes, mas sim os documentos. Portanto, no primeiro caso, há uma criação externa ao nome ‘construção’ (a casa); no segundo, interna ao nome ‘documentação’ (os documentos).
Olhando, agora, para os Nomes de Entidade deverbais, propomos, então, que haja sim estrutura argumental, mas que uma variável y, abstrata, ocupe a posição de complemento. Essa variável, então, pode estabelecer a referência dos nomes de duas maneiras distintas:

(3) A construção é muito alta.
(4) A documentação está na gaveta.

Em (3), a variável y é interpretada como a entidade construída, de modo que ‘construção’ é, justamente, essa entidade (externa ao nome). E mais: a variável y satura a posição de complemento, impedindo sua ocupação por outro elemento: “*A construção da casa é muito alta”. Já em (4), essa variável y é identificada com a raiz (em processo similar ao que ocorre com cópulas); ‘documentação’, então, é documento.

Migratory processes and the geography of Costa Rica have produced that a complex dialectal distribution. A particular case of these dialect differences is that of the alveolar vibrants, which have several allophones. Social processes and the influence of academic authoritarian figures and mass media have led to some of these variants being stigmatized. In this work we explore, in a preliminary way, the value judgments that are associated with different allophones of the vibrants.

This work focuses on the allophones of the alveolar tap /ɾ/ and trill /r/ in Costa Rican speech. The population of interest is groups of professionals living in the Greater Metropolitan Area (GAM) from various areas of the country. Eight of the speakers are female and four are male, aged between 23 and 46 years. The data was collected through two written questionnaires and WhatsApp recordings. In the first questionnaire, the speakers were asked for socioeconomic information, an assessment of their linguistic attitude about their own speech, possible negative linguistic experiences, and sentence readings. The second questionnaire was sent two weeks later. Each participant was asked to listen to the sentence reading audio of another of the speakers and determine if the speech is characteristic of the GAM, their educational level and socioeconomic group.

All speakers performed the tap as vibrant [ɾ] in the intervocalic and final position environments, as well as in the group [ɾt]. Similarly, its realization was as vibrant in the context of the groups [tɾ], [dɾ] and [bɾ]. Most of the speakers realized the group [pɾ] as vibrant. In the environment before nasal vibrant and retroflex realizations of the tap were observed.

Vibrant [r], fricative [ɹ̝], and retroflex [ɽ] variants were observed as allophones of the trill. No occurrences of the assibilated variant [ɹ̠] were found in sentence reading but were found in natural speech. Most speakers maintained a single variant throughout all sentences and in their natural speech.

Negative linguistics experiences reported included differences in the pronunciation of words or speed of speech, perception of the accent as a foreigner or from a different area than the area in which they are located, perception of different socioeconomic status.

In the majority of cases, the speakers were perceived as members of the middle class of the GAM, and in some cases they were perceived as members of a high socioeconomic group. Most of them were perceived as having at least a secondary education (7 of 12 cases).

None of the variants used by the speakers in the reading of sentences correspond to the allophones considered little prestigious. This is consistent with the perception of speakers as belonging to a middle or upper class socioeconomic group.
Some of the speakers presented assibilated realizations, however, these were not present in the reading of sentences, which could correspond to careful speech. It was not possible to determine if the perception of a lower educational level could be related to a perception of the voice as that of a younger person. This perception was found for both men and women of different ages so we do not consider that this is an effect related to the gender of the speaker.
The main limitation of this study is that the sample of speakers is small, and that the speakers belong to similar socioeconomic backgrounds.

The speakers showed allophonic realizations to the simple and multiple vibrant ones consistent with those found in people of the Costa Rican middle or upper class.

Os tweets apresentarão minha tese de doutorado em desenvolvimento. Partindo do princípio de que vivemos em um mundo com ampla aceitação do digital (de acordo com parte da literatura científica), gestos de interpretação sobre o digital são criados. Viver no digital hoje é interpretar o mundo usando uma tela como interface. Porém, onde há interpretação, há também polissemia, o que é conhecido na análise do discurso francesa Pechêuxtiana como “fuga de sentidos”. Em tais derivas, referências discursivas se deslocam historicamente, influenciando nossa percepção sobre o mundo e o real. Então, qual seria a deriva do sentido de se viver em uma era de maravilhas tecnológicas? A percepção de que o mundo era melhor no passado, principalmente o digital. O resultado de uma busca ao termo “a internet era melhor no passado” traz à tona uma série de artigos de opinião que ponderam sobre a qualidade da internet nos tempos atuais. Os enunciados trazem um sentido equívoco para o conceito de “internet” e “melhor”, se relacionados com a memória discursiva acerca do digital, que são 1. A internet é um símbolo de evolução tecnológica e evolução é algo positivo; e 2. A internet evoluiu de uma forma que não atendeu as necessidades e expectativas de seus usuários. O trabalho busca entender o papel da nostalgia no digital. Ela se apresenta como resistência ao discurso formador da ideologia capitalista, ou como sua aliada, buscando silenciar vozes dissonantes? Quais novos sentidos criados quando usamos a nostalgia nas condições de produção? Quais são os enunciados criados quando o discurso da tecnologia desliza e o novo é tido como indesejado ou desnecessariamente complicado? E como a adesão à essas novas condições de produção interpeladas pela nostalgia afeta o discurso em outros campos, como o político? Inicialmente, faremos uma investigação sobre o conceito de nostalgia ao longo da história e sua literatura. Em seguida, analisaremos a produção de conteúdo na internet, feita principalmente em agregadores de conteúdo de entretenimento e memes que apresentam um tom semelhante aos textos apresentados acima para então entendermos o processo de significação do sujeito por meio da nostalgia. Por fim, já munidos da compreensão de produção do discurso nostálgico, investigaremos o discurso corrente nas redes sociais e os interdiscursos presentes no enunciado, buscando relação com entre o espaço material da rede e a produção de discursos nostálgicos. Esperamos, ao final dessa reflexão, encontrar uma relação entre os enunciados produzidos nas redes sociais interpelados pelas condições de produção do discurso nostálgico e o campo do político.

For more than a hundred years, linguists have puzzled over questions regarding word borrowing. Empirical studies generally capture loanword use either by analysing types and tokens of borrowed words in a corpus, or by comparing type/token frequency with near-synonyms native to the receiver language. In such studies, the unit of measurement for investigating loanwords is frequency of use.

This talk will introduce an alternative approach to studying loanwords. Our method involves building networks of collocation by extracting sets of borrowings that co-occur within the same text. We refer to these sets as “intra-textual relationships”. Collocation is usually operationalised a priori with a specified window size (e.g. five words to the left or right of the keyword); however, the texts analysed are typically much larger than this window and may differ in length. Consequently, we extend the notion of collocation to what we term “collotextualisation”: capturing co-occurrence across the entire text, regardless of size.

We present a case-study of how collotextualisation can be used to complement conventional frequency-of-use measures when exploring loanwords. The data in our analysis consists of New Zealand English newspaper articles, which we use to study indigenous Māori words. Our corpus is themed around Matariki, the Māori New Year, and spans a period of ten years (2007-2016). The corpus comprises 91,958 words and 194 texts, with a borrowing rate of 29 loanwords per 1,000 words. After extracting 107 borrowings that occur at least five times in the corpus, we analysed the data by leveraging a special type of network (called a hypergraph) that preserves intra-textual relationships involving multiple loanwords. This allowed us to bypass the limitations of a standard network, which flattens the data into (less meaningful) pairwise co-occurrences.

We show that hypergraphs can uncover fresh insights into loanword use, especially when explored over time or by examining the (average) size of the intra-textual relationships. We report three main findings. First, most loanwords in our data occur with at least four others (i.e. loans occur in sets rather than in isolation). Second, there is an inverse relationship between intra-textual co-occurrence size and frequency of use, which means that newspaper articles are unlikely to contain an infrequent loanword and no frequently occurring ones. This is consistent with the idea that loanwords might occur in vocabulary frequency bands. Third, frequent loanwords take part in more distinct and recurrent relationships than infrequent ones, and are typically the first to occur in a given text.

El valor semántico que posee el ítem de libre elección “cualquier” ha sido discutido por más de 50 años (Vendler 1974, Dayal 1998, Horn 2000, etc.). Sin embargo, en lingüística hispánica dicha discusión es más reciente (Menéndez-Benito 2005, Arregui 2006, etc.). Se ha discutido si es un existencial, un universal, o ambos, y existen motivaciones independientes para cualquiera de las propuestas. Esto se debe a que puede aparecer en contextos que favorecen una lectura existencial, como 1, contra una universal, como 2:

1. Podés sacar cualquier auto.
2. Cualquiera puede caminar.

Aquí, defendemos que “cualquier” debe ser interpretado con valor universal, a partir del análisis de la interacción de la negación y frases formadas con "cualquier" del español rioplatense (del https://www.corpusdelespanol.org/web-dial) en la posición de objeto en oraciones episódicas negativas, como 3 y 4:

3. No nos referimos a cualquier tipo de prenda, sino aquellas exclusivas de alto nivel.
4. No queremos cualquier verdad.

Asumimos, siguiendo a Menéndez-Benito (2005) y Aloni (2019), que “cualquier” es un pronombre indeterminado universal que involucra dos operadores encubiertos, un universal y uno de exhaustividad. La exhaustividad, que puede ser aplicada tanto a nivel de proposiciones como de entidades, explica por qué, si bien “cualquier” es universal, no tiene el mismo comportamiento que
“todo”. Al combinar esta propuesta con un marco sintáctico para la negación (Etxepare and Uribe-Etxebarria 2011), mostraremos que la lectura que obtenemos en 3 y 4 de “algunas x no son y” se debe a que en ellas se niega el valor universal introducido por el indeterminado. Argumentaremos que la lectura obtenida se debe a que el agente tiene la libertad de elegir de todas las opciones posibles aquella o aquellas que se ajustan mejor a lo que quiere. Por eso, en 4, se quiere la verdad que nos permita entender la dictadura cívico-militar uruguaya y no toda ni cualquier verdad.

Aloni, M. (2019). Indefinites as fossils: The case of wh-based free choice/ Arregui, A. (2006). Cualquier, exception phrases and negation. Amsterdam Studies in the Theory and History of Linvuistic Science, S 4, 278:1./ Dayal, V. (1998). Any as Inherently Modal. Linguistics & Philosophy, 21(5):433–476./ Etxepare, R. & Uribe, M. (2011). Foco y negación de constituyentes. En Escandell, M. V., Leonetti, M., & Sánchez, C. (eds) 60 problemas de gramática. Akal, Madrid/ Horn, L. R. (2000). Pick a theory, not just any theory. In Horn, L. & Kato, Y., editors, Negation and Polarity. Syntactic and Semantic Perspectives, pages 147–192. Oxford University Press, Oxford/ Menéndez-Benito, P. (2005). The Grammar of Choice. University of Massachusetts Amherst Amherst, MA/ Vendler, Z. (1974). Each and every, any and all. In Linguistics in Philosophy. Cornell University Press.

This work makes use of the expansion of a corpus of spoken and written samples of English built to reflect how health information is conveyed among non-specialists. The newest component includes a sub-corpus of health memes, capturing informal peer-to-peer communication, widely transmitted, from multiple social media outlets.

Structurally, memes are single-move exchanges. This presentation focuses on their use in advice giving. As unsolicited advice, they reflect DeCapua & Huber (1995)’s description that advice-givers “take upon themselves the role of expert, authority and concerned person.” Computational heuristics for recognizing advice speech acts (Donaghy 1990) require a premise, implicit or explicit, that provides a reason for the agent to perform or avoid a future action, often through the post writer’s authority or experience. Here that authority can be either through graphics or words.

Though presented as authoritative, memes feature both pros and cons on the same topics (flu shots, diet foods, catching a cold), reflecting the conflicting health views in the larger culture.
This is examined through incorporating Placenica (2012)’s strategies of dis/affiliation between advisor and hearer, applied to the 130 health-related memes gathered through queries in keyword + Google image searches during 2018-19, a sub-set of CADOH--the Corpus of American Discourses on Health. Disaffiliation here works by shaming readers into an inference that their choice is wrong (e.g., Dr. Evil, Robert Downy eye rolling, Condescending Wonka, fat power rangers); Affiliation encourages readers to continue on the right health path (e.g., Success kid, Oprah’s You Get a Car, Winter is Coming, animals or babies). Both types of informal claims of authority advocate for the hearer’s future healthy action.

Results show that by tapping humor and celebrity reinforcement in tempering authority, internet memes can convey indirect speech acts of advice, conveying directive force through signaling different alignments with the reader. This provides a richer way to recognize and document positioning in vernacular online conversations. While this work frames authority as signaling affiliation or not with the reader, it also suggests further exploration of the impact of recognized vs anonymous characters as reinforcers of authority. Likewise, since topics and graphics in memes are closely tied to the cultures of their creators, future work that seeks to understand how health topics are conveyed in other speech communities could help reveal differences in local beliefs about healthy behavior. One application is for public health workers to track the framing of health advice conveyed through memes to help gauge a community’s understanding of expected health outcomes.

The study of place identity has a long history in variationist sociolinguistics. Over the years, the phenomenon has surfaced under various names, including “local loyalty” (Ito & Preston 1998), “cultural identity” (Hazen 2000), and “place orientation” (Carmichael 2017), and has been operationalized in countless ways, depending on the local context. Existing studies on this topic focus either on chain shifts or one or two variables in isolation (cf. Guy & Hinskens 2016). None of them examine if and how the features they investigate operate together. This is unfortunate given that speakers use a variety of features at any given point to index aspects of their identity (Pratt 2018). Moreover, the vast majority focuses on phonetic and phonological variables (but see Schlegl 2019), even though speakers use features at all levels of the grammar to position themselves in social space. This study will use evidence from Northern Maine, a largely unstudied part of the New England dialect region, to illustrate that some speakers actually index local affiliation through combining certain linguistic features. The focus will be on two features that are very common in vernacular varieties of English: was/were variation in non-existential constructions, as in (1), and demonstrative them for those, as in (2).

(1) We was/were really close.
(2) But she like talked to them/those guys.

The data for this project come from the County Corpus, a collection of sociolinguistic interviews with 84 native speakers of English born and raised in Aroostook County (AC), the largest and most sparsely populated county in Maine. For the present analysis, a subsample of 30 speakers was selected, stratified by age and gender. Following Sneller (2019), my measure of local affiliation is based on the idea that speakers’ orientation to place can be operationalized as their orientation towards the ideologies of that place as discussed during the interview. Based on a qualitative analysis of the data, I identified four common ideologies, which I translated into seven yes or no questions that were used to assign a local affiliation score to each speaker (Roberts 2016).
Results show that male-presenting speakers with high local affiliation scores are likely to use nonstandard variants for both features. In contrast, none of the female-presenting participants use both features. This suggests that combining these two features may be a strategy for men to “do local affiliation” linguistically. I argue that the most likely explanation for these gender differences lies in the fact that the ideologies of place in Northern Maine are deeply connected to ideologies of gender, class, and - while hardly ever said explicitly – ethnicity. I further suggest that these findings warrant paying closer attention to co-occurrence of features in studies of language and identity.

So-called impersonal verbs are predicates which occur without a nominative argument, as Old English hyngrian ‘to hunger’ in (1). Here, the accusative Experiencer hine codes the animate entity being inwardly affected by an event or characterized by a state.

(1) Se ðe cymes to me ne hyncgreð hine-ACC.
‘He who comes to me, he will not be hungry [lit. ‘[it] will not hunger him’]’.
(950, OED s.v. "hunger", v. †1.)

The loss of impersonal constructions has received much scholarly attention, as these tend to disappear in many of the languages that once made use of them (Elvira 2009). Across Germanic languages, with the notable exception of Icelandic (Barðdal 2004), impersonal verbs and constructions have been lost either completely (English, Dutch, Mainland Scandinavian) or partly (German). In the particular case of English, impersonal constructions went out of use in the transition from Middle English (1100-1500) to Early Modern English (1500-1700).
Previous research has shown that impersonal verbs followed several paths of development in the course of readjusting their syntax from impersonal to personal usage (Möhlig-Falke 2012). One of such paths was the replacement of the impersonal verb by an adjectival pattern formed by the combination of to be plus a past participle or a related adjective (e.g. 1992 Daddy wasn’t any too pleased about it […], OED s.v. please, v. 4. a.). Verbs which followed this trajectory include hunger, please and thirst. As representative of this general trajectory, this presentation undertakes the quantitative and qualitative study of the competition between the verb hunger and the adjectival construction be hungry.
The period examined is Early Modern English, and the data are extracted from EEBOCorp 1.0 (1473-1700, 525 million words). The study further considers the Late Modern English period based on data drawn from CLMET3.0 (1710-1920). Results reveal that the Early Modern English period witnesses a clear tendency for the adjectival pattern to increase at the expense of the verb hunger, especially in the literal sense ‘to feel hunger’ (see OED s.v. hunger, v.), and much less so in the metaphorical sense ‘to desire’. This tendency continues into Late Modern English, and can be accounted for by the interaction between the semantics of the adjectival construction, which assigns a stative and unvolitional interpretation to the event, and the sense ‘to feel hunger’, which is also inherently stative and unvolitional.

References
Bardðal, Jóhanna. 2004. “The semantics of the impersonal construction in Icelandic, German and Faroese”. In Werner Abraham (ed.), Topics of Germanic typology, 105-137. Berlin: Akademie Verlag.
CLMET3.0 = De Smet, Hendrik, Hans-Jürgen Diller & Jukka Tyrkkö. 2013. The Corpus of Late Modern English Texts, version 3.0. Leuven: KU Leuven.
Elvira, Javier. 2009. “El retroceso de la impersonalidad en español”. In Fernando Sánchez-Miret (ed.), Romanística sin complejos. Homenaje a Carmen Pensado, 123-145. Bern: Peter Lang.
EEBOCorp 1.0 = Petré, Peter. 2013. Early English Books Online Corpus 1.0. Leuven: KU Leuven.
Möhlig-Falke, Ruth. 2012. The Early English Impersonal Construction: An Analysis of Verbal and Constructional Meaning. New York: Oxford University Press.
OED = Oxford English Dictionary Online. https://www.oed.com