Monday, 28 January 2013

Coming to the end

Speaker A:            “Let’s discuss the use of final particles in conversational English”
Speaker B:            “Ok … What does that mean though?”

Alexander Haselow was interested in the fact that, during the past few decades, the words actually, even, though, then, and anyway have become more and more present towards the end of an utterance in spoken English.  So, for example, speakers will often say things like I’ll have to do it quickly though or I never liked it anyway. In this position these words are called final particles’ and Haselow wanted to investigate further what their function in conversations might be. He analysed data taken from the British component of the International Corpus of English, focusing only on speech in private, more intimate contexts, as this is unplanned and exactly where final particles occur most frequently.

Although when used in other parts of a sentence (or ‘utterance’ in speech) these words can be labelled grammatically as various things, including ‘adverbs’ or ‘connectives’, Haselow argues that in a final position it is hard to determine what their grammatical function is. It seems that when they are used in this position, they take on a different meaning to that which they have mid-utterance. They seem to provide a type of subjective ‘comment’ on something that has been said earlier in the conversation. For example, then in You’re not coming then suggests that the speaker has understood or ‘inferred’ this from something that has been said earlier or, in the case of I’m not coming actually, the actually suggests that the speaker is going against expectations that may have arisen from what has previously been said. In both of the above examples the main clause itself makes sense without the final particle, so it is not dependent on it. However, this is not the case for the final particle as it cannot stand alone and needs some speech to latch onto, even if it is just one word as in Alright then.

Haselow felt that the main function of final particles is to link its utterance to a previous part of the conversation whilst, at the same time, showing how the speaker wants their utterance to be understood.  It is this second function that determines the final particle that the speaker will use, as can be seen below:

Speaker A:            Bill’s going to the party
Speaker B:            I wouldn’t go then/though/anyway/actually/even

The choice of final particle here determines how Speaker B’s utterance relates to Speaker A’s. Final particles therefore appear to link their utterance to a previous one, either from a different or the same speaker. Because of this, conversations cannot be started with an utterance ending with a final particle. For example, I can’t collect you though isn’t just stating a fact (‘I can’t collect you’) but is also somehow contradicting something that has previously been said. 

So final particles allow speakers to manipulate conversations, as they signal to other participants how they want them to understand what they are saying. They are vital elements of interactive spoken language where utterances need to be linked in order to create a meaningful conversation. It is because speech is unplanned and unrehearsed that this linking often has to be done in retrospect during the conversation itself and this is where final particles play such a vital role. For example, having these words at your disposal during a conversation means that you can use them to correct a preceding utterance, as in I didn’t do it though or maybe ask for the validity of something, as in You didn’t see it then? The end of an utterance is the easiest and most convenient place to slot these words in and the reason that they are most frequently found in this final position.

Who would have thought that having a conversation required such constant negotiation of meaning from its speakers? I never said it was simple though!
Haselow, Alexander  (2012) Subjectivity, intersubjectivity and the negotiation of common ground in spoken discourse: Final particles in English. Journal of Pragmatics 32: 182-204.

This summary was written by Gemma Stoyle

Monday, 21 January 2013

Growing up bilingually

one language or two: does it make a difference?

Do bilingual children develop language more slowly than children acquiring only one language? This is the question that a team of researchers from Florida Atlantic University set out to answer.

Erika Hoff and her colleagues measured the language skills of 103 children, both boys and girls, when they were aged 1 year and 10 months, 2 years and 1 month, and finally at the age of 2 years and 6 months. 47 children were bilingual, acquiring both Spanish and English, and 56 were monolingual, acquiring only English. All the children were from families with a high socioeconomic status.

Overall, the English language skills of the monolingual English-learning children were more advanced and improved more rapidly during this period than the English language skills of the bilingual children, though in all cases the scores were within the normal range of variation for monolingual children. The bilingual children lagged behind the monolingual children by about three months.

Importantly, though, this was only the case when the children’s skills in English were measured. When both English and Spanish words were included, the total vocabulary size for the bilingual children was no different from that of the monolingual children.
In other words, the bilingually developing children were learning words at the same rate as monolingual children, but their word learning was, like their language exposure, divided between two languages. Previous research had also shown that the size of a child’s vocabulary depended on the amount and type of language to which they were exposed, but what was new about this study was that similar results were obtained for grammar. Measurements of the grammatical complexity of the children’s speech, such as whether they produced combinations of words rather than single word utterances, also showed that the bilingual children lagged behind the monolingual children, but only when their skills in English were tested – not when their abilities in both English and Spanish were taken into account. As the researchers point out, this contradicts the idea that children’s vocabulary depends on what they hear, whereas children’s grammar develops as their cognitive processes become more mature. The findings suggest, instead, a link between vocabulary and grammar that could be either direct or indirect.

An important further finding was that the pace of language development reflected the amount of exposure to the two languages. The researchers divided the bilingual children into three groups: those who heard Spanish more often than English, those who heard English more often, and those whose exposure to the two languages was roughly equal. The size of the difference in the language skills of the monolingual and the bilingual children depended on the extent of the bilingual children’s exposure to each of their two languages. Thus, across all measures of English language skills, the bilingual children who heard English more often than Spanish scored most like the monolingual English-speaking children, followed by the balanced bilingual children (those who heard English and Spanish more or less equally).

The researchers point out that their findings have implications for education, as bilingual children may be more cognitively able than their scores indicate if they are tested in only one of their languages. Skill level in a single language, in other words, is not the same indicator of ability for bilingual children as it is for monolingual children.

Hoff, Erika, Core, Cynthia, Place, Silvia, Rumiche, Rosario, SeƱor, Melissa and Parra, Marisol (2012). Dual language exposure and earlybilingual development. Journal of Child Language 39: 1-27.


This summary was written by Jenny Cheshire

Tuesday, 15 January 2013

All the lads and lasses!

There are still lots of lads and lasses in the northeast of England

In data collected in the 1950s for the Survey of English Dialects (SED) it was found that there was a ‘north-south’ divide in the use of the terms lad and lass, with lad being the main term used in the north to mean BOY whereas boy itself was the main variant in the south. Lass competed with other terms (such as girl, wench, maiden, missy) but it was still mainly confined to the north. When words for SON or DAUGHTER were elicited, the term lad competed with son in the north and, to a lesser degree, the term lass competed with daughter, but also only in the north. Research in the 1980s predicted that because lass was more variable than lad then it was likely that the term lass would be lost before lad.

Researchers Joan Beal and Lourdes Burbano-Elizondo set out to investigate patterns of variation and change in the use of these terms by comparing the 1950s data to data collected between 1994 and 2005 for two sociolinguistic studies in the northeast of England, the first in Newcastle and the second in Sunderland.

The Newcastle data revealed that both lad and lass is still used by all age groups and among all social classes, although lad is more widespread than lass. The results also showed (with the exception of middle class males) that males use both lad and lass more than females, so there appears to be variation according to gender. Like the SED data, the Newcastle data revealed that lad could be used to mean BOY or SON (though the latter was rare) but, unlike the SED data, it could also be used to mean SEXUAL PARTNER (in this case more often than boyfriend but less often than husband). Overall, though, lad was used in the vast majority of cases to mean YOUNG MAN. In the case of lass, girl was a much stronger rival than boy was for lad and the most frequent meaning for lass was YOUNG WOMAN. The use of lass to mean DAUGHTER was as rare as lad was to mean SON. However, lass could also be used to mean SEXUAL PARTNER, both in the sense of GIRLFRIEND and WIFE (whereas lad to mean HUSBAND was rare). Although there was a tendency for lower social groups to use more lad and lass, there was no firm correlation between social class and variable use of the terms; the researchers suggest that style or stance may be a greater predictor of their use but further research would be needed to shed light on this.

In the Sunderland data the results revealed that, in line with Newcastle, there was variation according to gender, with males using both lad and lass a lot more than females, especially in the sense of CHILD (male/female). Lad and lass could also be used to mean BROTHER/SISTER (although these were rare variants). When the meaning was CHILD or BROTHER/SISTER, lad and lass appeared at equal levels but lass was used much more frequently than lad (both by males and females) when used to mean SEXUAL PARTNER.   

Taking the results from both studies, the researchers state that both words are still used in the northeast of England but that their use may be declining. Generally speaking, both terms are more likely to be used by males than females but lass is used less often than lad. The exception is that lass is used more often than lad in the sense of SEXUAL PARTNER, with lass being the most commonly-used term for FEMALE SEXUAL PARTNER, especially by males. This leads the researchers to conclude that lass may not be dying out (as predicted in the 1980s research) but that it is undergoing semantic shift (a change in meaning) in northeast dialects whereby its primary meaning is SEXUAL PARTNER.
Beal, J. and Burbano-Elizondo, L. (2012). ‘All the Lads and Lasses’: lexical variation in Tyne and Wear. English Today 28 (4): 10-22.

doi: 10.1017/S0266078412000351

This summary was written by Sue Fox

Tuesday, 8 January 2013

Throwing light on language variation

Which sounds more natural to you?
(1) she turned on the light or (2) she turned the light on;
(3) he cut open the melon or (4) he cut the melon open

If you find (2) and (4) more natural, the chances are you’re from the UK or Ireland; if (1) and (3) sound better, you’re more likely to be from North America. This was the unexpected finding of Bill Haddican and Daniel Ezra Johnson’s recent research. They tested the judgments of 297 native speakers of English. Almost half of these people were from the UK or Ireland; the other half were from the US and Canada. They were asked to rate sentences like the ones above on an 11-point scale, where 0 was ‘bad’ and 10 was ‘good’. People from the UK and Ireland judged ‘discontinuous’ sentences as ‘good’ more often than those from Canada and the US (discontinuous sentences are those where the verb and its particle are separated, like cut and open in he cut the melon open).

Haddican and Johnson then looked at how people in the UK and the US produce sentences containing a verb and a particle: are they more likely to use continuous forms (he cut open the melon) or discontinuous forms (he cut the melon open)? They collected a Twitter corpus from two areas in the USA – within a 150 mile radius of Pittsburgh, Pennsylvania and of Concord, New Hampshire – and two areas in the UK – within a 150 mile radius of Oxford, England and of Glasgow, Scotland. They searched for tweets in this corpus that contained either turn on the light or turn the light on, finding more than 2000. Although there were no significant differences between the use of turn on the light and turn the light on between the two regions of the USA or the two regions of the UK, the results from the US and the UK overall mirrored the results of the judgment tests: the discontinuous form (turn the light on) was used more often in tweets from the UK, and the continuous form (turn on the light) occurred more often in tweets from the US.  

Why should these differences exist between UK English and US English? Haddican and Johnson could see three possible explanations. First, English in the UK could be changing, with speakers of British English moving towards the use of the discontinuous order, but only after the period of North American colonization. Second, it could be English in the US that is changing, with American speakers of English moving towards the use of the continuous order (turn on the light). A third possibility is that both varieties of English are changing, but at different rates; for example, varieties could be moving toward the discontinuous order, but UK dialects have moved further and/or faster.

Haddican and Johnson point out that we would need to analyse verb-particle combinations in earlier periods of English to find out which of these possibilities was most likely to be right. They make a start by analyzing 5 different verb particle combinations in the Corpus of Historical American English (COHA). They analysed the frequency of put out a hand/put a hand out; put out the fire/put the fire out; put out the light/put the light out; bring up the subject/bring the subject up; and the past form brought up the subject/brought the subject up. They found a steady increase since 1850 in the frequency of the discontinuous forms suggesting, they claim, that the third explanation is the most plausible one: both UK and US English are moving towards the use of the discontinuous structure, but the change has progressed more quickly in the British Isles.

What is needed now, they point out, is historical research on the use of these two structures in UK English, to confirm that here too speakers have been moving over time towards an increased use of discontinuous forms.
Haddican, Bill and Johnson, Daniel Ezra Johnson (2012) Effects on the Particle Verb Alternation across English Dialects. University of Pennsylvania Working Papers in Linguistics 18 (2): 31-40

This summary was written by Jenny Cheshire