She refers therefore she is: Morphosyntax and pragmatics in referential communication


Post written by Aylin C. Küntay, Koç University, Istanbul & Utrecht University, Utrecht

Based on an upcoming keynote talk to be given at IASCL 2014 this week (14th – 18th July, Amsterdam)

Referential communication is talking about things and people, an essential ability upon which many human communicative interactions build. To be able to communicate effectively, speakers and addressees should concur on what they are talking about. Although this sounds trivial, even adults sometimes have trouble in pinpointing exactly what their interlocutor has in mind, or might fail to express their referential intentions in the clearest way.

The evidence we have about children’s referential abilities is mixed. An 18-month-old can be quite effective in making us pick the right diaper with the desired picture out from a heap of clean diapers. A 5-year-old, on the other hand, might lose us among the many characters he introduces in his retelling of a movie. Many factors distinguish the situation of the diaper-picker from the film-narrator. Yet in our methodological and analytical frameworks, we forget that the act of referential behavior is embedded in certain contexts and geared towards a particular type of interactive experience. My talk will focus on the contextual conditions that render toddlers and preschoolers referentially (in)effective.

For my keynote talk in the conference organized by the IASCL in 2014 to take place in Amsterdam, I will focus on the contributions interactive contexts, interactive goals, and interactive partners make to the development of referential communication. I will introduce data from narrative interactions and conversational discourse of children in addition to experimental studies.

These studies show how naturalistic interactions with others and their feedback impact (monolingual) children’s development of referential communication. Infants are presented by their caregivers with richly textured patterns of referential sets, where the referent remains constant across extended stretches of discourse. This constancy is accompanied by integration of nonverbal cues such as gestures, gazes, and touches in addition to linguistic expansions and reductions regarding the referent.

Preschool children display morphosyntactically more sophisticated and referentially clearer structures when they build their discourse structures conversationally rather than via being prompted by picturebooks, when they assume more audience-oriented interactive goals, and when they are trained on referential effectiveness. In brief, children need to learn to form a variety of (often language-specific) expressive devices in addition to learning how to use these devices for particular interactive contexts and discourse functions.

Discover more about the IASCL 2014 here


Uptalk: power and prejudice

ENG 30.2Post written by Editor of English Today Clive Upton based on his Editorial in the latest issue 

From time to time the media pick up on instances of English language use which do not carry over entirely happily from one variety to another. One of the most well-known of these must be the feature – variously known by such labels as ‘high rising terminal’ (HRT), ‘Australian question intonation’ (AQI), or ‘uptalk’ – which sees an upward inflection being introduced to utterances that are not actually questions. Long unremarkable in Australia, and increasingly unremarkable elsewhere in the English-speaking world, especially among younger speakers, this feature nevertheless annoys a lot of people who do not use it themselves. A recent online discussion of this phenomenon in the United States can be found here.

The latest in a catalogue of complaints recorded about uptalk has come as a result of a survey of business leaders carried out by Pearson, the findings of which were reported in the Daily Mail and other British newspapers in January this year. Under the headline ‘Want a promotion? Don’t speak like an AUSSIE [i.e. an Australian]’, the Mail explains that Pearson surveyed 700 men and women in managerial roles, and more than half said that the trait was a clear indicator of insecurity and would hinder employment prospects. Taking this up the following day, the Guardian newspaper featured a light-hearted debate between an Australian, Alex McClintock, and an Englishwoman, Rae Earl. Among many partisan claims by McClintock was that ‘far from indicating insecurity, some studies suggest that the AQI is often used by powerful people when speaking to their subordinates (thereby explaining why Australians use it when talking to Britons)’. Earl countered with equal spirit, observing that uptalk ‘makes tiny admissions of doubt sound like Pacific-sized adolescent insecurities’.

As well as provoking popular discussion, this one linguistic feature has a long and distinguished history of sociolinguistic analysis. But we might remark here on two very fundamental issues that touch on the everyday world. Firstly, it is clear that how we speak leads others to judge us: as G. B. Shaw wrote in his Preface to Pygmalion, ‘It is impossible for an Englishman [for this, probably read ‘any speaker’] to open his mouth without making some other Englishman [i.e. speaker] hate or despise him’. People do not generally hold back from being judgemental about the speech of others, even though they might be careful not to voice prejudices based on, for example, gender, race, or religion. Secondly, the media never tire of discussing English. Those of us who teach or research the language can expect a ready audience for what we have to say, making it essential that we ensure our facts are correct and our opinions objective.

Read the full Editorial here 

What could possibly be the relevance of nativelike hedging strategies in EFL instruction?

ENG 30.2Blog post written by Sergio Torres-Martínez based on an article in English Today

The answer seems to be obvious from the perspective of, say, ELF research. Yet things are less clear-cut in expanding circle regions where nativelikeness is increasingly a hallmark of language proficiency, success and status. This article hinges precisely on the assumption that specific speech functions such as hedging (used to express vagueness or non-assertiveness) are important assets in the construction of language proficiency in ELF classrooms. Thus, a case for hedging strings (HSs)-a set of formulaic sequences with pragmalinguistic relevance to the teaching of spontaneous speech functions in EFL- is presented.

Formulaic sequences such as sort of/kind of, usually disregarded as downtoners, are revisited, further classified into three main categories (utterance softeners, vagueness prompters, and conversational fillers) and analysed using samples from free-access corpora. Hedging strings are thus deemed to have a role in three different conversational models, namely NS/NS (L1-L1), NS/NNS (L1/L2), and NNS/NNS (L2-L2). In addition, this paper explores the concept of conversational grammar (CG) as a competence-driven construct and not simply as a performance, corpus-driven translation of digitised data brought into the classroom (corpus-driven learning, also referred to as data-driven learning [DDL]).

The connection with classroom is completed by the introduction of a constructionist framework to language acquisition (Goldberg, 1995; Goldberg, 2006) by means of concepts such as form-function pairings. Construction grammars depart form the assumption that general cognitive mechanisms, and not innate language endowment (as mainstream generative theory argue) are at play in language acquisition. This multidisciplinary theoretical framework is spiced up by a review of activities drawing on construction-noticing instructional strategies via enhanced types of input. These activities are aimed to consolidate the notion that conversational proficiency is at a large extent determined by learners’ ability to identify pragmalinguistic patterns of communication leading to the development of interactional competences.

Read the entire article ‘Introducing conversational grammar in EFL: a case for hedging strings’ here

Predicting risk for oral and written language learning difficulties in students educated in a second language

Post written by Dr. Caroline Erdos based on an article from Applied Psycholinguistics

Students who struggle with oral language and literacy are at increased risk for dropping out of school. The gap between struggling students and their typically-developing peers is smallest early on and therefore, the chances of bridging that gap are greatest in the early grades. However, more and more students have had little or no exposure to the language of schooling until their first day of school and this makes it difficult for school personnel to disentangle true risk for learning disability from incomplete second language acquisition. The result is that identification and intervention is often delayed in the case of second language learners, even those in immersion classes (ex: native speakers of English attending French immersion school), thus placing them at a significant disadvantage as compared to native speakers of the language of schooling (ex: native speakers of English attending English school) who often begin to receive help with oral language or (pre)literacy as early as kindergarten.

A promising avenue is to use student’s skills in oral language and literacy in their first language to predict how they will eventually perform in these areas in their second language. It is crucial to fully understand the possibilities and limitations of this method, however.

A second related issue is the importance of providing help that is most likely to have the greatest impact on student’s academic success. Numerous studies and clinical experience have shown that the more targeted the help, the more likely students will make gains. Therefore, once a child has been identified as presenting with oral language or literacy difficulties, it is imperative to identify the specific area of difficulty within each domain — in the area of oral language: vocabulary, grammar, phonology, discourse, or pragmatics; and in the area of literacy: phonological processing, letter-sound knowledge, decoding accuracy, decoding speed, lexical knowledge, or reading comprehension. Targeted intervention is key to making gains. For example, a child who struggles to understand what he reads is not likely to benefit from intervention targeting letter-sound knowledge, unless poor letter-sound knowledge was the primary cause of his inability to understand what he reads. Exactly how to provide targeted intervention is better understood for some areas, for example decoding accuracy or decoding speed,  than for others, for example oral language or reading comprehension. However, even in these less understood domains there is a general consensus that intervention that focuses on vocabulary (breadth and depth) and complex language skills would be useful.

Read the full article until July 31, 2014:

“Predicting risk for oral and written language learning difficulties in students educated in a second language” by Caroline Erdos, Fred Genesee, Robert Savage and Corinne Haigh

A time for Critical Language Awareness

ENG-Leiden-imagePost written by Robin Straaije

One of the aims of the research project Bridging the Unbridgeable: linguists, prescriptivists and the general public at Leiden University is to construct a mutually beneficial discussion involving those three groups.

My stance is that linguists should not shy away from being ‘critical’ in the sociological sense, nor from engaging with the public, with the practices of those whose language variety they are describing. And being specialists on language, linguists should explain the workings of language to the public, including normative aspects. Historical linguists are especially suited to being engaged in this discussion, since they can place it in a much needed historical and social context.

One of the things I am interested in is how involved linguists are – and how involved they wish to be – in public discussions about prescriptivism and language norms, and in raising critical language awareness. As far as I can can tell, linguists do not to lead these discussions, which seems strange since they are supposed to be the experts. But there are probably other opinions. In the second feature for the project in English Today, I ask you to join the discussion.

 Read more about the collaboration between English Today and Bridging the Unbridgeable here 

Variation in Faroese and the development of a spoken standard: In search of corpus evidence

NJL 2014Post written by Remco Knooihuizen based on an article in the latest issue of Nordic Journal of Linguistics 

A common theme in many European languages is that differences between dialects are becoming smaller. There are several in which this can happen: the strengthening of standard languages, frequent contact between people from different places (“dialect levelling”), or the spread of the dialect from an influential centre to other places. In many well-studied languages, these processes are difficult to tell apart because they interact and have been going on for a long while. But if we look at smaller languages, where these processes are more recent, we can find out more about the role of the different processes in dialect change.

One such language is Faroese, a Scandinavian language spoken by some 50,000 speakers in the Faroe Islands between Scotland and Iceland. The islands have known rapid social change after World War II, with improved transport links and Faroese-language media and education — three developments that map onto the processes of levelling, dialect spread and standardisation, respectively. There is evidence that dialect differences are decreasing in Faroese and that a more standard version of the language, called “Central Faroese”, is on the rise.

This study looks at a collection of recordings of spoken Faroese to chart the change in two linguistic variables, to see which mechanisms best account for the change. One variable is phonological (the pronunciation of the -ir and -ur endings which occur frequently in the language), the other is morphological (the occurrence of -st verb endings). Importantly, each mechanism would suggest a different course of change for these variables, which would allow us to tell the processes apart. We looked at two kinds of variables because Faroese language education focuses on ‘correct’ grammar, but no model pronunciation is taught in schools.

The analysis of the recordings confirms the geographical variation that we know exists in the language. However, there are very few differences between older and younger speakers from the same location, suggesting that there isn’t much language change going on. Speakers also don’t speak differently in informal conversations than they do in more formal interviews, where we would expect Central Faroese to appear. This is probably due to the close-knit Faroese society: because “everyone knows everyone”, there is less reason to use very formal speech styles and changes in these styles progress more slowly.

Read the full article ‘Variation in Faroese and the development of a spoken standard: In search of corpus evidence’ here

Place-names and the Norman Conquest

ENG - NormanBlog post written by David Trotter based on an article in the latest issue of English Today

Place-names are more than just names of places on a map or more often than not, in a satnav. Their history is the story of settlement and of human movement, of social organization and human beings’ implantation in a landscape. Just as cathedrals, churches, pubs, telephone boxes, straight Roman roads and winding English ones tell us who lived where and when, so too do place-names. English place-names show many layers of history: Celtic elements overlaid with Anglo-Saxon, Scandinavian, sometimes Latin components. All tell us something about who lived where. Yet in this linguistic mixture, one language is relatively – and surprisingly – absent: French, the language of the Norman Conquest, and of the elite of English society for a good part of the Middle Ages. William the Conqueror installed Frenchmen in leading positions in the church and in government; he issued laws in French, and Domesday Book (the land survey which found out who owned England and what the country produced) was compiled by Norman scribes. French became the language of law-courts, the royal court, literature, science (second to Latin), and remained for quite some time (at least 15o years after 1066) the language of the aristocracy, for whom in most cases Normandy was still their real home. Yet place-names remained intact as (mostly) Anglo-Saxon. In part, this is because places often do keep their names through thick and thin – all over Europe there are pre-Roman names, some of them traceable to Indo-European. Probably it was also easier just to carry on with the names already in use. At the same time, French is not as absent as has sometimes been concluded in the few studies published so far of this aspect of English place-names.

This study looked at the first volumes of a new compilation of all the place-names in England in alphabetical order, and at everything from two counties (Cumberland and Westmorland, most of modern Cumbria). Cumbria is not a region especially densely settled by the Normans yet there are quite a few place-names of obviously French origin there. Intriguingly, too, field-names are quite often French too. Fields are very “local” indeed (the name for a field cannot have had any significance at all outside the immediate locality) and it is quite a surprise to find that French is used for their names. Does this maybe mean that it was in wider spoken use than we usually think? That is hard to prove. What id does show is that the French of the Norman invaders was more widely known in the remoter parts of the English countryside than most people have so far allowed for. That may go some way to explaining why there are French elements in many English dialects; but that is another story.

Access the full article ‘Why are there so few French place-names in England?’ here 

The concept of identity in the East Midlands of England

ENG 30.2 NatalieBlog post written by Natalie Braber based on a recent article in English Today

Investigating feelings of identity in East Midlands adolescents

When considering language variation in the UK, linguists have frequently considered the North/South divide and the linguistic markers separating the two regions, for example by investigating the vowel speakers use in words like ‘grass’ and whether the words ‘put’ and ‘putt’ sound the same or not. But it has been noted that this is not a straightforward division and that this situation is more complex. There are clear stereotypes for the North and South – but how do areas like the East Midlands fit into the picture? The boundaries between North and South are defined in different ways and in linguistic studies the East Midlands have been described as belonging to the North and to the South. Linguistically, the question has been raised whether there is a clear North/South boundary or whether there is a transition zone in the Midlands. This paper revisits this question from the point of view of young people living in the East Midlands, to examine their sense of identity and whether this cultural divide is salient to them.

The East Midlands is a problematic area in its definition geographically as there is no overall agreement in which regions belong to the East Midlands, and people may have difficulty in relating this to their own sense of identity. It seems that for many the North/South divide is a natural one but what do non-linguists, and specifically young people, think? Although the East Midlands may be the geographical centre of England, it is not in any sense the perceived centre of England. It is an area which can be hard to locate perceptually and has been referred as ‘neither here nor there’ and as a ‘no-man’s land’. It seems that a definition of where the East Midlands is and what to call it is problematic, and this paper will deal with these issues to attempt to resolve them.

Read the entire article ‘The concept of identity in the East Midlands of England’ here

Chopping down the Syntax Tree

Word-tree-5Blog post by Remi van Trijp based on a recent article in Language and Cognition

One of the most notorious problems in linguistics is how to handle “long-distance dependencies”: utterances in which some elements seem to have been taken away from their original position and then moved to a different place. Typical examples are WH-questions such as “What did you see” in which the direct object (“what”) takes sentence-initial position instead of following the verb, as it would do in a declarative utterance (e.g. “I saw the game”).

But what makes long-distance dependencies so difficult? Most linguists assume a tree structure (or “phrase structure”) for analyzing utterances. As a data structure, trees consist of nodes that have at most one parent node, which means that information in a tree can only trickle down from a parent to its immediate children, or percolate upwards in the other direction. A tree structure is thus hopelessly inadequate for representing dependencies between nodes that are in the top of the hierarchy and nodes that are situated somewhere below. The most common solution to this problem is to say that there is a “gap” where we would normally expect a part of the utterance. Information about the gapped element then has to be communicated node-by-node upwards in the tree, until the “filler” of the gap is found.

In recent years, however, a cognitive-functional alternative has started to crystallize in which long-distance dependencies spontaneously emerge as a side effect of how grammatical constructions interact with each other in order to cater for the different communicative needs of language users. For example, the difference between “I like ice cream” and “Ice cream I like” can be simply explained as the tendency for speakers to put the most topical information in the front of the sentence – suggesting that word order should be decoupled from an utterance’s hierarchical structure.

While this view has for a long time been dismissed for being ad-hoc and not lending itself to proper scientific formalization, there now exists a formally explicit computational implementation of the cognitive-functional alternative in Fluid Construction Grammar, which works for both parsing and production. The implementation eliminates all formal machinery needed for filler-gaps by chopping down the syntax tree: rather than taking a tree structure as the sole device for representing all information of an utterance, different linguistic perspectives are represented on equal footing (including an utterance’s information structure, functional structure, illocutionary force, and so on).

The implementation shows that a cognitive-functional approach to long-distance dependencies outperforms the filler-gap analysis in several domains: it is more parsimonious, more complete (i.e. it includes a processing model) and it offers a better fit to empirical data on language evolution.

Access the entire article without charge until 31st July 2014

I’m glad that you like it or I’m glad you like it?

LCO-cover-image-for-Cambridge-ExtraBlog post written by  Stefanie Wulff based on an article in the latest issue of Language and Cognition 

When do native speakers say I’m glad that you like it, and when do they drop the complementizer and simply say I’m glad you like it? The factors influencing complementizer realization have intrigued many researchers over the last few decades. The emerging consensus is that various factors jointly determine whether the complementizer is realized, such as how long or complex the parts are that the complementizer connects or how frequent the verb is in actual language use. This study elaborated on previous studies by asking: what about non-native speakers of English? The “rules” for complementizer variation are never taught in the English language classroom, and yet second language learners, at least at an advanced level of proficiency, also drop the complementizer in specific contexts. We wanted to know: are these contexts the same as for native speakers? That is, do the same factors that govern native speakers’ choices also impact learners’ choices?

To that end, we retrieved 3,622 instances of complement constructions from native English corpora and German as well as Spanish learner English corpora. We coded all instances for the factors proposed in previous research and ran a logistic regression model. Our final model suggests that advanced-level learners of English are in fact closely aligned with native speakers in that their choices are, generally speaking, influenced by the same factors. However, a closer look reveals that the relative importance of these factors differs: learners rely more on processing-related factors such as clause complexity, and they are comparatively less sensitive to the statistical associations between a given verb and the complementizer. Overall, learners realize the complementizer more often in contexts in which a native speaker may well omit it. Also, comparing the German and the Spanish learners, we observed that the German speakers make more native-like choices than their Spanish peers. We interpret these findings to reflect (i) the comparatively higher cognitive cost associated with deploying your second language vs. your first, (ii) that second language learning will vary as a function of the first language the learner speaks, and (iii) the crucial role that the input learners receive plays in learning to make native-like choices.

Read the full article ‘That-variation in German and Spanish L2 English’ here