Read an Excerpt

Cultural Migrants and Optimal Language Acquisition

Multilingual Matters

Beyond Native-like? The Lexical Profile of a Cultural Migrant in Italy

Anna Gudmundson and Camilla Bardel

Introduction

In this study we analyse lexical variation and lexical sophistication in cross-sectional data from six Swedish L2 users of Italian, who regularly visit Italy or have lived in the country for long periods, but then moved back to Sweden and enrolled in courses in Italian at Stockholm University. The main focus of this chapter, however, is a longitudinal study of a very advanced learner of Italian, a young Swedish woman, Alva, who has chosen to live in Italy for personal reasons. Her level of oral proficiency is very high when it comes to grammar and pronunciation. As regards her vocabulary, there are no formal deviances from the target norm. Aspects of her vocabulary are compared, using lexical profiling data based on the tools Vocd and LOPP f, to that of native speakers of Italian, and in addition to the profiling data, we also carry out a qualitative analysis of her vocabulary, focusing mainly on the low-frequency words and on the words that differ between Alva and the native speakers. Analysing the words that are unique in Alva's production, that is, words that are not used in any of the native speakers' productions, allows us to see if, and in that case how, an advanced learner's vocabulary may differ from that of native speakers. A frequency-based perspective is adopted, according to which high-frequency words are assumed to be learnt earlier than low-frequency words (Cobb & Horst, 2004).

What characterises a cultural migrant from a sociolinguistic and a proficiency point of view? By the term cultural migrant, we mean someone who, for one reason or the other, is personally interested in the target language and culture, and therefore moves to, or spends a lot of time in, a country where the target language is spoken.

According to Forsberg Lundell and Bartning (this volume) cultural migrants are people who decide 'out of their free will, to move to another country permanently and learn another language. The culture of the target language country is highly appealing to them and they make an active choice to live in this new culture'. They are also, according to Forsberg Lundell and Bartning (this volume), often well integrated, with high degrees of education, something that leads to good chances to practice the L2. While the effect of practice in a second language is still an issue of debate (DeKeyser, 2007), it is assumed here that time spent in the TL country under the circumstances characterised by the motivation of someone who has chosen to live in a particular country may promote a high level of L2 attainment.

The informant Alva satisfies all the prerequisites of the above definition of a cultural migrant. In addition to her case, the other informants investigated in this study can be said to fulfil part of the definition. The main difference between them and Alva is that they have moved back to Sweden, their home country. The periods they have spent in Italy are in some cases also shorter than the time spent by Alva, in others longer. Characteristic for the whole group is that they all have high degrees of education, and they all have positive attitudes towards Italy and the Italian language, and during their stays in the country they were all immersed in the target language culture (see Tables 1.1 and 1.2 for detailed information about the informants). In Schumann's (1986) terms, the degree of acculturation can be assumed to be high in all the informants, and thus, as Schumann's acculturation model suggests, their chances of achieving high levels of L2 attainment are high: 'Learners will acquire the target language to the degree they acculturate to the target language group' (Schumann, 1986: 379).

As already mentioned, the particular linguistic aspect of interest in this study is vocabulary use. To our knowledge, the vocabulary of cultural migrants has not been much investigated (but see Forsberg-Lundell & Lindqvist, 2014). It seems reasonable to assume that someone living in the TL country has optimal chances to develop the L2 vocabulary, thanks to a rich input and many opportunities for practice in interaction.

The level of vocabulary richness in oral production can be measured with a number of different measures, ranging from the type-token ratio and more elaborated variants thereof (Daller et al., 2003) to mere judgements made by language teachers (Tidball & Treffers-Daller, 2008). In this study, we will discuss the results obtained from two measures, the D measure, or Vocd (Malvern et al., 2004), which measures lexical variation, and the Lexical Oral Production Profile (LOPPf), a recently developed instrument for measuring lexical sophistication in Italian oral production, by means of proportion of low-frequency words. The production of the learners is compared to frequency bands based on the 3000 most frequent words from target language corpora and the words that fall outside the 3000 most frequent words are the so-called offlist words. In some studies (e.g. Lindqvist et al., 2011) the words that fall outside the 2000 most frequent words, i.e. band 3-words and offlist words, are considered advanced. Hence, LOPPf is an instrument used for measuring lexical sophistication defined as the proportion of low-frequency words in the total of words produced (Laufer & Nation, 1995). Lexical variation, or diversity, can be measured by the simple type/token ratio (TTR), a calculation of the number of types divided by the number of tokens in a text. However, it has been pointed out by many that a problem with TTR is its sensitivity to text length: 'The more words (tokens) a text has, the less likely it is that new words (types) will occur' (McCarthy & Jarvis, 2007: 460). The longer the text, the more often words are repeated, high-frequency words will be repeated more often, in comparison to low-frequency words, and this tendency will increase the longer the text is. A tentative solution to this problem is the D measure (for details see Lindqvist et al., 2013).

In this study, the lexical profiles of the L2 users are compared to those of native speakers. Because native speakers have varied lexical competence, particularly depending on their level of education and reading habits, data were gathered also from a group of L1 speakers who have an educational level that is similar to that of the Swedish informants. Being native-like in this study thus means reaching levels similar to those of native speakers with the same educational background.

In the qualitative analysis of Alva's production, we look into the longitudinal lexical development of this particular L2 user, focusing on the new lemmas that are added at the different times of measurement, starting from the lexical sophistication data. There will be a particular focus on the low-frequency words used by Alva and in a second step we will look at those low-frequency words that are used only by Alva and none of the Italian native speakers. Finally, we will see if there are any words that are used by all the native speakers, but not by Alva.

Data Collection

The data in this study comes from recordings of semi-guided interviews led by a native speaker of Italian contained in the InterIta corpus (see Bardel et al., 2012 for a brief description of the corpus). The topics of conversation are daily activities such as family life, hobbies, studies, job experiences, travelling and stays in Italy. The questions of the different interviews aim at inducing the L2 speakers to use different grammatical structures each time, but there is no systematic intention to elicit different vocabulary from time to time. Some of the recordings include a narration task, which induces the informants to use vocabulary of particular semantic fields. This, as well as other conditions, is the same for all informants, native speakers and non-native speakers.

Each of the six L2 users of this study have been immersed in the target language context for long periods, or they visit the target language country regularly and are then living in contexts where Italian families or friends are highly present (see Tables 1.1 and 1.2). All informants have studied Italian at Stockholm University or do study there at the time of the data collection. One of the learners, Alva, decided to move to Italy and to stay there, and she still lives there at the time of the fourth and last recording with her.

Alva started to learn Italian in a natural context at the age of 27, when she moved to Italy to stay with her Italian boyfriend, as she reports in the following extract from the first recording:

*ALV: sono fidanzata con [/] con un ragazzo di Biella.

I am engaged to a boy from Biella.

*PAO: quindi è per questo motivo che sei in Italia oppure?

so this is the reason for you being in Italy, or ...?

*ALV: esatto sì.

exactly, yes.

*PAO: eh hm quindi mi pare di capire che abiti in Italia soltanto per motivi affettivi oppure ci sono anche altre ragioni?

eh so you live in Italy only for affective reasons, or are there also other motives?

*ALV: ma eh: no mi [/] mi sono trasferita soltanto per il fatto che lui abita lì e poi ovviamente anche sono [/] sono curiosa di [/] di imparare l'italiano + ...

well eh no I moved there just because he lived there and obviously I am curious to learn Italian ...

*ALV: di [/] di avere l' opportunità di [/] di vivere in un altro paese ...

to get the opportunity to live in another country ...

*ALV: ... una cosa che mi interessa molto.

... something that interests me a lot.

*ALV: ma la cosa che mi ha fatto eh trasferirmi lì è stato lui [=! ride].

but what really made me move there was him (laughs).

Alva works as an English teacher and as a designer. After a year in Italy, she enrolled in a web-based course of Italian at the University of Stockholm. She was recorded four times during five years. The other learners of the study are presented in Table 1.2.

As the following extracts show, the learners have positive attitudes towards Italy, its culture and language. Frank, for instance, plans to go to the country and to work there:

*FRA: tz perché [/] eh: perché: eh penso forse di poter abitare e lavorare in Italia perché anche sto studiando economia # eh: a [/] # a me piacerebbe forse lavorare almeno un paio di anni all' estero.

tz because eh because I think that maybe I could live and work in Italy, because I also study economics eh I would like to work maybe at least a couple of years abroad.

*FRA: eh: hm l' Italia sarebbe proprio # eh il paese giusto per [/] per me eh.

eh hm Italy would be exactly the right country for me.

Nora says that Italian is the language she likes the best, because it gives her the liberty to express herself:

*PAO: c'è una lingua del cuore una lingua che ti [>] rispetto alle altre?

is there any language that you prefer over the others?

*NOR: [<].

ah, ok.

*NOR: va beh.

well.

*NOR: l' italiano sì perché mi dà una libertà di [/] di esprimermi molto eh + ...

Italian yes because it gives me the liberty to express myself much eh ...

Ulla, who used to live in Italy in her childhood, misses many aspects of the Italian lifestyle:

*PAO: e quando sei in Svezia che cosa ti manca dell' Italia?

and when you are in Sweden what do you miss about Italy?

*ULL: eh: il modo di vivere.

eh the lifestyle.

*ULL: il modo di pensare.

the way people think.

*ULL: eh ## il clima.

the climate.

*ULL: il mangiare.

the food.

*ULL: sono cose abbastanza importanti.

those are rather important things.

The control group

The control group is a group of six Italian native speakers recorded during their time as Erasmus students at the University of Stockholm. They were all between 21 and 25 years of age at the time of recording.

Research Questions

Previous studies (Bardel et al., 2012; Lindqvist et al., 2011) show that only very advanced L2 learners reach levels that are close to those of native speakers, when it comes to lexical sophistication in oral production. Indeed, all native speakers cannot be expected to have the same degree of lexical richness or sophistication, but our previous studies have shown that the group of Erasmus students examined here all fall within a rather restricted range when it comes to lexical sophistication, as measured by LOPPf. This study investigates whether it is possible for an L2 learner who lives, or has lived in the country, more or less regularly, to achieve a native-like lexical profile as measured both by Vocd and LOPPf.

If so, what characterises the vocabulary of such an advanced L2 speaker, from a qualitative point of view? Are the words used by a cultural migrant the same as those used by native speakers? In order to investigate this, we will look, from a qualitative point of view, into the words used by Alva, focusing on the low-frequency words. We will also look at the words that are used only by Alva and none of the Italian native speakers, trying to categorise them. We think that this could be a way to understand what possibly makes a cultural migrant different from native speakers from a vocabulary perspective. For the same scope, we will see if there are any words that are used by all the native speakers, but not by Alva.

Results

The cross-sectional study

In this section, we aim at answering our first research question, regarding the possibility for a cultural migrant achieving a native-like lexical profile as concerns lexical sophistication and variation.

In Table 1.3, the results for the non-native speakers are shown, in both lexical variation (Vocd) and lexical sophistication (LOPPf).

As Table 1.3 shows, Kristina has the highest lexical variation and Alva has the highest lexical sophistication (i.e. proportion of low-frequency words) in the group.

In Table 1.4, the same measures are reported for the native speakers.

When comparing the two groups, one can see that the lexical variation is clearly higher in the native speakers' group (mean 123.74) than in the non-native speakers (mean 90.92). A Mann-Whitney U test reveals a statistically reliable difference as regards the Vocd value between the native speakers and the non-native speakers, with native speakers receiving higher scores than non-native speakers, z=-2.40, p=0.016. Non-natives had an average rank of 4.00, while natives had an average rank of 9.00.

When it comes to lexical sophistication, the mean is only slightly higher in the native speakers (8.20) as compared with the non-natives (7.88). A Mann-Whitney U test failed to reveal a statistically reliable difference as regards the LOPPf value between the native speakers group and the non-native speaker group, z=-0.962, p=0.336. Non-natives had an average rank of 5.50, while natives had an average rank of 7.50.

This is illustrated in Figure 1.1 and Figure 1.2, where the native and the non-native speakers are represented in the same plots. The horizontal line indicates the point, above which all native speaker values fall:

The difference between native speakers and non-native speakers, when it comes to lexical variation, can be seen above. As can be noted, only the informant Kristina falls within the native-speaker range, while all the others are located below the horizontal line. As for lexical sophistication, however, there is practically no difference, as seen in Figure 1.2.

As can be seen from Figure 1.2, only two informants, Frank and Nina, fall below the native-speaker range, and it can be noted that the difference between Frank and Natale, the native speaker with the lowest LOPPf-value, is very small, 7.31 compared to 7.38.

In summary the results indicate that even though cultural migrants seem to reach a rather advanced level when it comes to lexical sophistication, they don't seem to vary their vocabulary as much as native speakers do. This could be interpreted as if they use many low-frequency words, but that those low-frequency words are repeated more often.

The case of Alva – quantitative results

In this section, a closer look will be taken at Alva and her longitudinal development in the four recordings, comparing her results to the results from three recordings with the native speakers. Table 1.5 and Figure 1.3 show the Vocd values in the sequences of recordings of Alva and the native speakers.

As can be seen in Figure 1.3, while the lexical profiles of the Italian native speakers seem to vary randomly in time 1, 2 and 3, the pattern shown in Alva's recordings could be interpreted as a development. Her profile is slightly lower in the first recording, and then in the lower range of the natives when measured by Vocd. Alva's development could of course also be interpreted as random variation. Since this is a case study, it is in fact impossible to say with certainty that her pattern differs from that of the native speakers (see e.g. Antonio, whose profile could also be interpreted as developmental). But considering the group of native speakers as a whole, it seems as though their pattern differs from that of Alva, which clearly points upwards. Furthermore, it doesn't seem plausible that the native speakers would develop the lexical sophistication of their L1 during their stay in Sweden. The same reasoning is valid for the results concerning lexical sophistication, which is shown in Table 1.6 and Figure 1.4.

---

Excerpted from Cultural Migrants and Optimal Language Acquisition by Fanny Forsberg Lundell, Inge Bartning. Copyright © 2015 Fanny Forsberg Lundell, Inge Bartning and the authors of individual chapters. Excerpted by permission of Multilingual Matters.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

---