22
Informal English Learning in France

MERYL KUSYK

Introduction

For many French children, exposure to foreign languages begins in the educational realm.1 World languages and cultures are generally introduced in the école maternelle (preschool/kindergarten, ages 3–5), after which instruction of a specific language begins. A second foreign language is added in middle school and the learning of both continues through high school. The number of hours per week of formal classes depends on the grade: in elementary school foreign languages are taught for 1.5 hours per week; this increases to a combined 4 hours per week in middle school and 5.5 hours per week in high school for both languages. Several types of foreign language programs are offered, depending on the school, for example regional languages may be studied or students may enroll in a “European section,” which offers content and language integrated learning. According to the national education policy, all students should reach a B12 level by the end of compulsory education (in general, the end of middle school or 16 years of age), while high school baccalauréat students should have a B2 level upon graduation.3 A recent survey on European language competences conducted by the European Commission (2012) has shown that these goals are a work in progress: of the 14 countries that took part, France ranked second to last for listening, reading, and writing combined and only 14% of students had a B1 level or higher at the end of compulsory education.4

Beyond formal education, one potential means of foreign language exposure is through films and television. Whereas foreign language input via subtitled cinema and television programs is common in certain European countries (Scandinavia, for example), in France, dubbing is the dominant practice: most films are dubbed (53% of European films and 82% of American films) and television programs are almost exclusively dubbed (Directorate‐General 2013).

It is important to note that in France, those who endeavor to learn a second language (L2) experience not only the expected challenges that come with acquiring any new skill but also often grapple with additional feelings of frustration, inadequacy, or anxiety due to a commonly held belief that the French are inherently poor language learners.5 Such beliefs are not innocuous, as issues of affect can have a significant impact on learners' self‐efficacy, motivation, and ultimately, their success in learning an L2. The difficulty that many French learners associate with learning foreign languages presents an interesting point of departure for examining informal language learning, which distinguishes itself from formal learning on several fronts: contact with the L2 is voluntary rather than imposed, frequency of participation can vary drastically (several hours per month, week, or day), the choice of language belongs to the language user (though most research has been conducted on English) and there is no summative evaluation which assesses knowledge of the language as an object of study; rather, the language user mobilizes his or her language skills on a case‐by‐case basis for communicative, problem‐solving, or leisure purposes. In the informal sphere, success is hinged upon effective conveyance or understanding of meaning rather than correct usage of prescriptive grammar rules. These elements contribute to an environment in which an individual may adopt an L2 speaker identity that can allow him or her to experience reduced anxiety and increased self‐confidence in his or her language abilities (Lam 2000). In this way, the informal sphere may represent an opportunity for overcoming some of the frustration and stigma that has traditionally been associated with foreign language learning in France.

Early findings on informal English learning in France

Research on informal language learning in France has grown substantially since 2010, when Toffoli and Sockett published their first article on the online informal learning of English (OILE). They describe the phenomenon as the participation by non‐native speakers in a multitude of leisure activities which involve the use of or the exposure to English. This participation may lead to the acquisition of certain aspects of the L2, of which the non‐native speakers may be aware; however there is no explicit intention to learn the language. In addition, participation in activities takes place outside the educational context and is not imposed by the curriculum. In their 2010 study, the authors reported on a survey which aimed to gather information on French university students’ (n = 222) informal L2 habits online. At this time, it was not known if OILE concerned just a handful of film aficionados and video game players or if its scope extended to a more general student population. The results revealed that more than half the sample (54%) listened to English more than once a week (either as music, films, or series), though other types of L2 usage did not obtain such high percentages: 18% read in English more than once a week and 7% spoke in English more than once a week. The data on listening comprehension prompted further research in this area, beginning with Sockett's (2011) corpus study, which targeted the input from television series to which informal language users are exposed. Sockett studied four‐gram constructions from five popular television series6 and established a frequency list of the 50 most used chunks (for example, what are you doing, I want you to, nothing to do with, I was trying to). A follow‐up study then targeted French students' knowledge of these four‐gram constructions (Kusyk and Sockett 2012) and found that frequent viewers7 had better knowledge of the chunks than non‐frequent viewers. While the entire sample (n = 45) contained participants of various English levels, an additional analysis of only B2 level speakers (n = 12) also generated statistically significant results in favor of the frequent viewers.

Toffoli and Sockett (2013) sought to find out how teachers view informal activity participation in English. They surveyed 30 English teachers at French universities and found that they underestimated the frequencies with which their students participate in informal activities in English; that opinions on the impact of informal activities were generally positive, especially with regard to oral comprehension skills; and that certain teachers integrated informal activities into their courses while others did not due to issues such as legality, intrusion into students' private lives and not being allowed to modify course curriculum. As informal English learning in France has until now primarily focused on university students who, regardless of their major, must take a non‐specialist English class as part of their studies, it is relevant to inquire about the role that formal courses may play and what interface might exist between the formal and informal spheres. Non‐specialist8 English courses may take a variety of forms, from more traditional “focus on form” approaches to more autonomy‐oriented settings such as self‐access or language resource centers. Toffoli and Sockett (2015) addressed the role of formal English learning vis‐à‐vis informal English learning and discussed in particular how language resource centers can stay relevant in an age where non‐native speakers can access the L2 in both authentic and pedagogical forms via their laptops or smartphones. They suggested that the centers must evolve and update their offer so as to propose services that are not easily found in the informal sphere. This could include emphasizing the face‐to‐face nature of the centers, making them a place to meet and practice oral production, encouraging students to come to teachers with questions about aspects of their informal usage with which they need help or helping students navigate pragmatic aspects of the L2 such as context‐dependent appropriateness of register.

While the question of the formal sphere remains central, the foremost research questions in the field of OILE have been investigating the nature and scope of participation in informal activities and the impact of this participation on non‐native speakers' L2 development. Kusyk (2017) sought to expand the scope of Toffoli and Sockett's (2010) initial participation habits survey and to examine the latter question through a longitudinal lens. The following two sections examine these questions in detail.

Participation in online informal activities

One of the first steps in investigating the informal language learning process is understanding the nature of exposure to and use of the L2. More specifically, this involves finding out what informal language learners do, how often they do it, the nature of how they do it, what their reasons are for doing it, and what their perceptions are of doing it. Results from a 2014 questionnaire9 provide some answers to these questions. The target population was French university students majoring in fields other than English and one of the goals of the survey was to take an inventory of how widespread informal L2 use is amongst the general university population. Sample respondents were 538 French university students, mostly between the ages of 18 and 23 and majoring in subjects other than English (biology, medicine, earth sciences, etc.). The sample was split quite evenly between female (50.7%) and male (49.3%) participants. The majority of respondents self‐evaluated their English skills between the A2 and C1 levels,10 with the greatest concentration of abilities attributed to the intermediate B1 and B2 levels.

Participants were asked to fill out the questionnaire during a normal class period. This approach was chosen rather than disseminating the survey on the university listserve as a means of avoiding bias; for example, it is possible that only students who participate in informal activities in English would fill out the questionnaire. Having teachers integrate it into their course material on a given day was therefore part of an effort to increase the likelihood of capturing a sample that was more representative of the overall population, including both those who participate but also those who do not.

With regard to what informal language users in France do and how often they do it, the five most popular activities were comprehension/reception activities: watching series, watching films, watching videos, reading, and listening to music. Participation rates varied greatly depending on the activity: 60% of students never or rarely (one to three hours per month) watch films in English while 26% of students listen to music in English approximately one hour per day. Students were invited to share their favorite television series. In total, 139 series were noted. While 10 accounted for 53.8% of responses, it is interesting to note that 129 series accounted for the other half of the sample. This indicates, on the one hand, a certain homogeneity with regard to input and the more popular series could be examined within the context of a corpus/L2 exposure study. The variety that characterizes the additional 129 series represents, on the other hand, a significant level of content (and potentially linguistic) diversity to which informal language users are exposed.

In addition to reading, listening to music and series, film and video watching, participation data were also collected on multiplayer and single‐player video game use, writing emails, speaking and writing comments or messages on Facebook. The majority of the sample reported that they never participate in these production/interaction activities, though a modest percentage of students did claim to write emails (36%) or interact on Facebook (22%) one to three times per month. Very low percentages were noted for the remainder of the frequency categories.

The above frequency data serve as a reminder that although moderate to high levels of participation have been noted for certain (comprehension) activities, for a substantial portion of the present sample participation in informal activities remains rare or even nonexistent.

Certain questions from the survey aimed to explore how language users participate in informal activities online, for example with regard to viewing modality or efforts to overcome comprehension difficulties. Watching series and films with subtitles, either in French or in English, appears to be commonplace: 53% noted watching series with French subtitles, 20% with English subtitles and 12% without subtitles. (For film watching, the percentages are essentially the same.)11 In some cases, these modalities change over time: 26% of students reported having changed their series viewing mode from watching with subtitles in French to watching with subtitles in English, while 9% went from French subtitles to no subtitles and 6% changed from English subtitles to no subtitles.12

Opting for the activation of modalities such as subtitles can be one way that students seek to scaffold their L2 learning, though it is not entirely clear if their activation is motivated by a desire to understand additional input or rather if they are simply a pre‐established aspect of the encoded viewing content.13 One series of questions sought to examine more directly the extent to which students make efforts to overcome comprehension difficulties. For both music and content viewing (series, films, videos) participants were asked how often they look up unknown words and expressions. Additionally, they were asked how often they make efforts to simply try to remember new words and expressions. Responses to these questions suggest that participants are more amenable to trying to remember novel expressions rather than physically looking them up, with 54% doing this half the time or more, compared to 39% (content viewing) and 33% (listening to music) that actively look up unknown words at this same rate. Efforts to learn or remember new vocabulary could be an indication of wanting to improve one's L2 or it could simply suggest steps being taken to achieve a basic threshold level of content coverage, without which the meaning of a dialogue or scene cannot be understood.

Students also responded to the question of why they participate in informal activities online in English. Interestingly, “improving or maintaining my level of English” came into play for series, film, and video watching but not for the other activities.14 For all activities entertainment or relaxation played a role, though each activity had its own additional reasons. Authenticity, English level, and immediate availability all influenced the decision to view leisure content in English, while having access to a greater range of choice impacted listening to music in English. Several reasons in addition to leisure made significant contributions to reading in English: acquiring general knowledge, needing specific documentation or tutorials, informing oneself, and needing information in relation to a homework assignment or project. This last reason does not signify that teachers assigned homework in English, but rather, that the students could not find the answers that they needed in French and therefore sought out information on the topics in English. As one student commented, “the language is not an obstacle, what's important is the subject.”15 Finally, video game players indicated that the strong presence of games on international servers as well as the presence of more (interesting) games and players on these servers influenced their decision to play video games in English. A substantial number of participant comments in the “other” category for this question also highlighted the importance of playing a game in its original (English) version.

The last section of the questionnaire dealt with aspects of students' perceptions of their informal activity participation, such as the perceived impact of different types of L2 exposure on language skill level as well as general opinions on informal activities in English. When asked to attribute their current level of English to various possible influences, French students indicated that formal English classes and informal activities online played the biggest roles, whereas traveling and friends and family played significantly smaller roles. Sixty‐seven percent of participants thought that English courses contributed either “a lot” or “enormously” to their current English level while 52% felt that this was the case for online informal activities.

A related question asked students to rate the informal activities they believed had an impact on their English skills. Content viewing and reading in English appeared to have the most positive perceived impact, while these two activities and listening to music were the primary candidates for having a “rather positive” impact. Music and social networking were the main activities selected for “not much” of an impact and “not at all” having an impact.

As sample participants were enrolled in a non‐specialist English course during the dissemination of the questionnaire, it was of interest to inquire about possible similarities between the types of English encountered in both the formal and informal spheres. More than half of students (58%) found their course content to be only slightly similar to the informal English they come across online, while 31% found the content to be quite similar.

The results from the questionnaire serve to inform us about how French university students majoring in fields other than English participate in informal activities online in English. They show that informal language usage does not concern all (sample) students and that participation varies widely according to activity. French students use English for entertainment purposes, to relax, to read about current events or to research a homework assignment. In most cases the language plays a secondary role and a premium is placed on the activity itself. Nevertheless, students are aware that informal exposure may impact their L2 skills, though less than half take steps to overcome knowledge gaps when encountering unknown words and expressions. The majority of the sample do not find strong similarities between the English they come across in the formal sphere compared to the informal sphere.

While the questionnaire results provide an overview of French university students' informal participation tendencies, the following section allows a more in‐depth look at the profile of an individual informal language user as well as his language development over time.

A case study on L2 development from informal usage

A 10‐month case study was carried out from May 2015 to March 2016. Its objective was to examine how the L2 develops through participation in informal activities online. Interviews took place every six weeks during which the participant talked about his informal activity usage and wrote a short text about a familiar topic: either a television episode he had watched or a fan fiction episode of a series he liked. Every two weeks the participant filled out an informal habits survey in which he noted how often he had participated in various activities during that time span.

Informal language user profile

When the study commenced, Jérôme (mentioned in footnote 13) was a 21‐year‐old psychology major at a French university. His informal L2 usage profile was representative of the findings of the informal habits survey reported on in the section “Participation in online informal activities,” in that he participated in multiple activities, more of which were comprehension‐based than production‐based. His reasons for participating included leisure, entertainment, communication and problem‐solving and he used subtitles when viewing series and films. His participation in informal activities varied greatly over the 10 months. Reasons for these fluctuations included course workload, exam periods, vacation, fading or increasing interest in a particular activity and availability of desired viewing content (for example a series whose newest season just came out or ended).

Jérôme was an avid series viewer whose favorite programs included House of Cards, Doctor Who, Game of Thrones, Shameless, Vikings, Californication, and Big Bang Theory. He also watched films and videos in English; the former for entertainment purposes and latter for topic‐specific purposes, such as TED Talks or tutorials on cinematography, photography, video game strategy or science. He enjoyed watching series and films in their authentic version and opted for videos in English as he felt that the topics that interested him were more readily and easily available in English than in French. Jérôme was a music enthusiast and preferred English‐speaking artists almost to the point of exclusivity, noting that he did not care for French music. Being able to understand the lyrics was, for him, not compulsory, though if a particular song or artist interested him he would take the time to look up the lyrics and understand them. Often the music was on in the background while he carried out a simultaneous activity such as studying or working on his computer.

Jérôme's informal reading material during the first half of the study consisted primarily of Facebook content, such as Humans of New York, and photography and cinematography blogs or popular science websites. In the fall of 2015 Jérôme's reading rates increased sharply due to demands from his course load. In a select few cases he was asked by his teachers to read a scientific article in English but for the most part his internet searches for specific topics were simply more fruitful when conducted in English rather than in French.

Jérôme devoted some of his leisure time to playing videos games (both single and multiplayer) in English, though less than the above‐mentioned activities. In his single‐player game (Grand Theft Auto) he either heard or read English, while in his multiplayer game (League of Legends) he communicated with teammates in English via written chats. Communication in the multiplayer games was centered on strategy and included abbreviations and jargon; however, it was commonplace for teammates to have daily chitchat with one another during calmer phases of the game. Eventually he lost interest in playing video games, largely due to repeated occurrences of “flaming” – a term that signifies insulting one's teammates or opponents due to their inability to play the game.

With regard to viewing modality, Jérôme regularly used French and English subtitles when watching series, films, and videos in English. He also watched without any subtitles, though this concerned primarily videos. His subtitle usage did not follow any type of linear progression, with all three options being used throughout the entirety of the study. However, as Jérôme explained, there were reasons for the variation in subtitle use: with regard to series, it was usually the case that French subtitles were already integrated into the digital file he received. He often noted that they had been created by an amateur, as he could detect errors in the translations. His need for a visual aid depended on the series: for Big Bang Theory (scientific discourse) and Game of Thrones (fantasy series with British accents) they were necessary, whereas for Californication they were not. While for several series he could understand nearly all of the dialogue (Shameless, Vikings, House of Cards), he still preferred to have the French subtitles “on hand” rather than having to pause the series and look up the word. (Throughout the entirety of the study he did not look up any unknown words from television series.) The occasional translation error did not bother him as he was accustomed to paying attention to the subtitles only when he encountered a gap in his vocabulary knowledge. Toward the end of the study Jérôme noted that he could just as easily watch with English subtitles but the files he received were still encoded with French subtitles. For the few occasions that he did watch with English subtitles, he noticed that he did not have the same comprehension difficulties as he did at the beginning of the study. He attributed this to having become more familiar with his series of choice (from October until the end of the study he only watched Doctor Who) and did not seem to consider that his comprehension skills may have improved:

Yeah sometimes Doctor Who is difficult when they are use uh uh specific words from the Doctor Who universe, but uh uh on this season is more calm about this or maybe because I'm used to it.

(Interview from March 2016: Kusyk 2017, p. 185)

During the final interview (March 2016) Jérôme noted that he might watch the upcoming season of House of Cards with English rather than French subtitles. With regard to film viewing, Jérôme was able to actively choose in which language the subtitles appear, and for the most part he chose French. As film watching was often a group activity carried out with his roommates or his girlfriend, this choice was a reflection of the group's modality preference. For videos, Jerôme explicitly sought out English subtitles, as the majority of the videos he watched were scientific or subject‐specific and it was important for him to learn the original terminology.

Jérôme was asked at each interview to self‐assess his English language abilities. He was given a CEFR language assessment grid from which the levels (A1, B2, etc.) had been removed and featured only descriptions of ability for listening, reading, spoken interaction, spoken production, and writing (for example: I can understand extended speech and lectures and follow even complex lines of argument provided the topic is reasonably familiar. I can understand most TV news and current affairs programmes. I can understand the majority of films in standard dialect).

During the first half of the study Jérôme found it difficult to self‐assess his L2 skills as his conception of assessment was tightly associated with notions of (scholastic) testing and he had not been graded on certain aspects of his L2 for quite some time. Therefore, he would try to recall, for example, his previous grades in composition writing rather than his ability to communicate with his video game teammates as the determiner of success for writing. It was toward the middle of the study that he began to use his actions in the informal sphere to justify his abilities in the L2. This delayed recognition could be due to several reasons, one of which is the tendency to overlook the informal sphere as a legitimate source of learning or skill acquisition (Schugurensky 2007).

Throughout most of the study Jérôme did not attend any formal English courses, though for the last five weeks of the study he was required, as part of his psychology major, to be present at a language resource center for two hours per week to work on his English. During his time at the center he did not attend a language course in the traditional sense, but rather was supposed to complete activities of his choice (prepare a presentation, watch a movie, speak with peers, etc.). Jérôme had a clear disdain for this obligation, which sometimes included writing a summary of something he did in English outside of the center (for example watching a television episode):

I do not like this way of working. I like to to read a lot of things things in English I like to watch th things in En English. That's pleasure to do it because they are very very interesting stuff. But I do not have to uh to have to uh sum up things uh to prove that I did it.

(Interview from January 2016: Kusyk 2017, p. 188)

For Jérôme, the language center was not conducive to getting actual school work done as it was too noisy and he was unable to concentrate. As a result, he often watched TED Talks or other videos, though this was frustrating and was perceived to be an invasion of his privacy as these are things that he preferred to do of his own volition, in his own time, and not in association with a course assignment. Interestingly, this particular center appeared to be open to recognizing and valuing informal practices, in that participation in activities in the informal sphere could count toward a student's grade. As a verification measure, students were required to provide some type of proof of participation in these activities, for example writing a summary. Jérôme's experience also brings up the question of autonomy. Developing and fostering autonomy is a commonly stated goal for university language resource centers; however, Jérôme felt that his ability to successfully complete tasks on his own was somewhat undermined:

Interviewer:

 
and how how is [going to the language center] for you?

Participant:

That's annoying. Yeah I do not like this [language resource center].
I do not like to to be forced to do something in a specific way and place and uh where I cannot really do it properly. And I hate that uh people ask me where I am in my work like we have a deadline just let me manage my own time.

(Interview from March 2016: Kusyk 2017, p. 188)

Whereas this particular center recognized and validated language activities that students did in the informal sphere (albeit with obligatory verification measures), this did not appear to be the case for certain aspects of autonomy development, such as managing one's time and completing tasks. The development of autonomy is, certainly, a complex notion that is difficult both to define and to assess, though it is closely linked to the ability to successfully complete tasks in an L2. Jérôme's frustrations with being micromanaged and having to write summaries of his leisure activities speak to the larger question of what an effective formal–informal interface could or should look like. If his opinions are any indication, the methods of recognizing and validating skills acquired in the informal sphere will be a fundamental piece of the puzzle.

Language development

The present section presents Jérôme's English development over the course of 10 months. During this time, informal activities were almost exclusively his only contact with English. On rare occasions, he spoke with acquaintances at social gatherings in English and at the very end of the study he participated in some activities in English at a language resource center (for a maximum of two hours per week, see the section “Informal language user profile”). Written production data were collected at interviews that took place every six weeks and were then analyzed according to 20 different complexity, accuracy, fluency, and chunk measures of linguistic development. Written data were taken from a writing activity, which took 15 minutes and was about a leisure topic familiar to him.

Complexity signifies the ability to produce sophisticated or difficult constructions and represents the upper limits of a non‐native speaker's L2 abilities. Accuracy is denoted by an absence of errors and an ability to produce the language in accordance with its traditional grammatical rules. Fluency refers to the automaticity of access to L2 knowledge, demonstrated by the ability to produce the L2 in real time without undue pauses or hesitations (Wolfe‐Quintero et al. 1998; Skehan 1998; Ellis and Barkhuizen 2005; Skehan 2001; Foster and Skehan 1996). Due to limitations in working memory capacity, it is difficult to simultaneously process or produce both the meaning (fluency) and form (complexity, accuracy) of linguistic content and inevitably one is prioritized over another (either consciously or unconsciously). Chunks (multiword expressions) were also chosen as a development measure as their use has been shown to be a robust measure in determining L2 skill level (Verspoor et al. 2012) and they are considered to be characteristic of native discourse (Sinclair 1991).

This case study was carried out within a dynamic usage‐based framework.16 The theoretical and methodological foundations of this approach consider variability and nonlinearity to be inherent elements of language development and emphasize that language is learned through meaningful experiences situated in social contexts. Notions such as instability and stability within L2 subsystems (for example, the subsystems of lexical complexity or accuracy) as well as the interconnectivity between individual variables (for example, the individual linguistic measures) are used to describe development. Accordingly, the following presentation of results17 centers on two main perspectives: first, discovering the relationships that exist between the informal activities and the linguistic development measures helps us understand how informal language use and exposure impact the evolving L2 system. Secondly, examining the different relationships that exist between the development measures themselves sheds light on how the L2 system develops internally over time. As this internal analysis will show, there is a constant push and pull between the variables (in this case, the 20 complexity, accuracy, fluency, and chunk measures); there is both progression and regression, or in other words, nonlinear change.

The following subsections examine the different ways in which Jérôme's L2 developed over time through his participation in informal activities online. The different types of relationships (supportive, competitive, conditional) found between the informal activities and the development measures will first be presented, followed by an analysis of the types of associations that exist between the measures themselves.18

Written development: relationships with informal activities

The informal activities whose frequencies correlated with various written linguistic development measures were videos watching, film watching, and “relaxation activities,” a combination of film watching, series watching, and listening to music. Reading may potentially have been in a conditional association with two measures. The following presents a detailed look at the different types of associations between the variables.

Videos

Videos correlated with lexical sophistication, from the complexity category, and with incorrect verb tense, from the accuracy category. The lexical sophistication measure comprises different levels of vocabulary frequencies: K1 represents the 1000 most frequent words, K2 signifies the second most frequent set of 1000, and so on such that the higher levels represent words that are used less frequently and are therefore considered to be more sophisticated. The correlated levels of lexical sophistication were:

  • K1: rs = −0.747, p = 0.033
  • K3–4: rs = 0.782, p = 0.022
  • K5–6: rs = 0.747, p = 0.033
  • K7–15: rs = −0.790, p = 0.020

Videos were therefore positively associated with K3–4 and K5–6 and negatively associated with K1 and K7–15. These relationships suggest that watching videos influenced Jérôme's lexical sophistication in that he used more moderately sophisticated vocabulary but avoided low (K1) and high (K7–15) sophistication levels. Videos were also found to be in a competitive relationship with the accuracy measure incorrect verb tense (rs = −0.713, p = 0.047). This signifies that the more he watched videos, the fewer errors of this type he produced.

Films

Film viewing correlated with two development measures. Firstly, there was a positive association with lexical sophistication, a complexity measure (rs = 0.788, p = 0.020), signifying that both variables “grew” together. Secondly, film viewing correlated negatively with singular/plural errors, an accuracy measure (rs = −0.765, p = 0.027), indicating that as film viewing increased, errors of this type decreased.

Relaxation activities

Jérôme's informal activities were grouped into two sets based on participation characteristics: films, series, and music were grouped into relaxation activities as these were activities that he did for pleasure or to wind down and during which he made no efforts to learn the language. Videos, reading, and video games were grouped into specific interest activities as they were activities in which Jérôme expressed a desire to learn the content (thereby requiring him to look up terminology) or in which he had to communicate with others in order to complete a communal task. Relaxation activities were found to be in a negative correlation with singular/plural errors, an accuracy measure (rs = −0.830, p = 0.011), indicating that increased participation in these activities was associated with a reduction of this type of error.

Reading

While reading was not found to be directly associated (correlated) with any individual measure, an analysis of the variable trajectories suggests that it might have been in a conditional19 relationship with clauses per T‐unit, a complexity measure, and incorrect verb tense, an accuracy measure. Informal reading in English increased sharply beginning in October. Shortly after this time, the number of clauses per T‐unit increased dramatically (indicating higher complexity) and the number of incorrect verb tense errors decreased suddenly, indicating better accuracy. In January, when reading rates decreased sharply, complexity then also decreased while error production increased. Though no statistically significant relationship was detected between the variables, these communal changes in trajectories could constitute a point of departure for potentially fruitful further study.

It is interesting to note that, with regard to participation frequency, Jérôme's top three activities – music, reading, and series viewing – did not individually correlate with any of the development measures.20 (Nor did any of the activities correlate internally amongst themselves.) While frequency of input is a central element within the dynamic usage‐based perspective, it is also important to consider the essential roles that variables such as level of attention, motivation, salience, or meaningfulness all play. For example, videos correlate with several measures, despite having a low overall frequency (just 29 hours over the course of the study divided between 0.5 and 2.5 hours per week). Jérôme was, however, intrinsically motivated to learn the content of the videos and he looked up unknown words and activated English subtitles, which can help direct attention to otherwise non‐salient elements.

Written development: associations between measures

The following subsection on written complexity, chunks, accuracy, and fluency will examine how these L2 subsystems developed over the 10 months during which Jérôme participated in informal activities in English. Their changes – both progressions and regressions – highlight the intricacies of L2 development: the dynamic and complex relationships that exist between variables (both intermeasure and intrameasure) as well as the different levels of in/stability that characterize the subsystems and their components.

Written complexity

The two grammatical complexity measures, T‐unit length and clauses per T‐unit were found to be in four correlations with other development measures. Firstly, T‐unit length correlated with clause length, a fluency measure (rs = 0.738, p = 0.037). This relationship can be considered within the greater methodological debate surrounding the individual measure T‐unit length, and whether it should belong to fluency or accuracy.21 An analysis of the Jérôme's T‐units suggested that increased length was the result of adding more words to his sentences as opposed to increasing length by adding subordinate clauses. This would tend to support the argument that T‐unit length is a measure of fluency; however, it may also be a measure that adapts to the public in question: more beginner or intermediate L2 speakers may increase length with words while more advanced speakers may increase length with subordinate clauses. T‐unit length also correlated with errors per T‐unit (rs = 0.790, p = 0.020) and incorrect word choice based on the L1 (rs = 0.717, p = 0.046), both accuracy measures. In the former case, complexity increased as errors increased (thus, poorer accuracy), which indicates that there is a “push and pull” between complexity and accuracy – they do not improve together.22 The latter correlation serves to inform us about which type of error is likely to occur when T‐unit length increases.

Finally, the grammatical complexity measure clauses per T‐unit also correlated with the accuracy measure incorrect word choice based on the L1 (rs = 0.736, p = 0.038), supporting the notion that the more complex Jérôme's written productions in English were, the more likely it was that this specific error would occur.

The final correlated measure of complexity was that of lexical complexity,23 specifically, lexical sophistication. The majority of vocabulary used by Jérôme (72–88%) was from the K1 category – in other words the most frequent 1000 words.24 Words from the K2 category also made up Jérôme's lexicon, however at a lesser rate (3–16%). Jérôme used more sophisticated words from time to time, such as winked, grumbled, intern, slaughter (K5–6) or stealth, tuxedo, earthling, stethoscope (K7–15).25 Correlation analyses showed that there was a competitive relationship between K1 and K2 (rs = −0.905, p = 0.002) and between K1 and K3–4 (rs = −0.874, p = 0.005), indicating that when Jérôme used more sophisticated language, it came from the K2 and K3–4 categories. A positive correlation between K2 and K3–4 (rs = 0.802, p = 0.017) supports this notion and suggests that the two variables grow together. Finally, a negative association between K3–4 and words per text, a fluency measure (rs = −0.731 p = 0.040) suggests that Jérôme sacrificed fluency when writing more sophisticated texts, perhaps needing more time to choose his words. Likewise, this relationship implies that he was able to write more words when using more simple language. This would suggest that Jérôme's use of more sophisticated vocabulary was not yet automatized.

Written chunks

The three types of chunks measured in this study were lexical collocations, verb particle constructions and fixed phrases. Jérôme used between three and five chunks per interview indicating that, as a whole, chunk usage was a somewhat stable aspect of his L2; however, on an individual level each measure showed a considerable amount of variation. Interestingly, only on two occasions (September and October) were all three chunks types used in the same text. Examples of chunk usage include: to run (someone) over, to cheer (someone) up (verb particle constructions); bare hands, strong accent (lexical collocations); nice to meet you, you have the right to remain silent (fixed phrases).

Written accuracy

The global measure of accuracy, errors per T‐unit, provided an overview of general accuracy development over the course of the study. After an initial decrease in errors at the beginning of the study, the number of errors per T‐unit remained between 0.86 and 0.65 for the remainder of the study, though dropped in October and December to 0.41. Certain errors were committed only rarely, such as use of word in the L1 or incorrect syntax based on the L1, while the rest fluctuated quite considerably. This fluctuation indicates that these were aspects of Jérôme's L2 that were not yet fully stable. A closer look at some of the individual measures allows an in‐depth look at the developing L2 system.

The aggregate measure incorrect word choice (total) combines incorrect word choice and incorrect word choice based on the L1. An overall downward trend (decrease in errors) was punctuated by periodic regressions (increase in errors), indicating that the variable may have been slowly shifting toward a new state that was qualitatively different from its previous state. In other words, Jérôme's lexical accuracy appeared to be slowly improving. Taking into consideration the nature of his informal practices (focus on meaning and not on linguistic form), it is perhaps not surprising to see an improvement in his lexical skills.

An analysis of the measure incorrect conjugation revealed an overall total of 19, of which 11 concerned the absence of the third person singular “s” (the scene take place) and 7 concerned a “doubly conjugated” simple past for constructions containing the auxiliary to do (did you saw, they did not heard). Although the trajectory of this variable showed considerable instability, the internal nature of the error appeared to be rather stable as it only concerned two types of conjugation error. The omission of the third person singular “s” is notoriously difficult for French learners of English, likely due to a combination of cue redundancy, cue unreliability, and the tendency to not pronounce the final letter(s) of a word in French. Likewise, L1 French speakers also do not use to do as an auxiliary in past tense conjugations. A closer analysis of the specific errors committed for the measures missing word and incorrect word choice revealed additional cases of L1 influence. For example, (missing words shown in parentheses) a young doctor in (an) internship based on un jeune médecin en stage; and cut (off) her answer based on et lui a coupé la parole.

With regard to significant associations amongst the accuracy measures themselves, two in particular are of note. Firstly, errors per T‐unit and incorrect word choice (rs = 0.880, p = 0.004) were found to be in a supportive relationship, meaning that the trajectories of these global and specific accuracy measures (respectively) were closely related. Secondly, incorrect verb tense and spelling errors correlated negatively (rs = −0.738, p = 0.037), indicating intrameasure competition, or a push and pull effect between the two variables.

Written fluency

The measure clause length was used to measure written fluency. A significant association was found between this fluency measure and an accuracy measure, missing word (rs = 0.806, p = 0.016). The positive correlation means that as fluency increases, so do errors, and vice versa. Considering Jérôme’s general downward trend of written fluency and of global error production (indicating better accuracy) it would appear as though he prioritized written accuracy over written fluency in the long term. The “choice” to prioritize one measure over another, either consciously or not, is a common theme in the literature on the complexity, accuracy, and fluency development measures and is illustrated here in the improvement of accuracy to the detriment of fluency.

Implications from the present research

Several key takeaways from this study can help us make sense of how an L2 develops through participation in informal activities. Firstly, development is complex. There is no one activity that dominates the informal landscape and not all activities correlate with the development measures. Interestingly, Jérôme’s most frequent activities (music and series viewing) did not correlate individually with any measures, pointing to the importance of factors other than frequency, such as salience in the input, personal meaning attributed to the activity, and levels of attention and motivation. Additionally, in the present study the majority of the associations between variables were found amongst the development measures themselves, rather than between the activities and the measures. This observation underscores the dynamic nature of development in that the many parts of the L2 system are constantly evolving, interacting, and adapting to each other over time. As such, any view of L2 change that does not consider how the system internally develops potentially misses out on relevant and fruitful areas of exploration.

Jérôme's L2 development can be characterized by both stability and instability. In some cases, strong fluctuations in the rates of a measure belied a quite stable internal nature, for example with the accuracy measure incorrect conjugation. Analyzing the in/stability of trajectories can be helpful in identifying different types of relationships, such as the potential conditional relationship between reading, clauses per T‐unit and incorrect verb tense errors in Jérôme's written productions. Additionally, this can help identify possible qualitative state changes, as in the case of incorrect word choice (total). As far as other global trends are concerned, trend lines helped identify five lexically associated areas of progression (lexical diversity, lexical sophistication K3–4, incorrect word choice and missing word). These measures, coupled with the possible qualitative state change (progression) for incorrect word choice (total) suggest that qualitative change in Jérôme's lexical knowledge and usage took place through his participation in informal activities. In other words, Jérôme's lexical accuracy and complexity showed signs of gradual improvement. An improvement in lexical skills does not necessarily come as a surprise when taking into consideration the nature of Jérôme's informal practices, which emphasize a focus on meaning and not on linguistic form. Finally, the L1 appears to play a significant role in Jérôme's L2 development, as several in‐depth analyses revealed L1 influence in error production.

The role of the formal sphere remains a pressing issue. As is the case with Jérôme, the majority of university students in France attend non‐specialist English courses and as such a great many students find themselves confronted with two different L2 usage contexts – formal and informal. Questions regarding possible bridges that could (or even should) be built between the two spheres as well as potential ways of recognizing and validating authentic, informal usage should be a focus of future informal language learning research.

Finally, as illustrated in both the questionnaire and the case study, informal language users participate in a variety of activities and for a variety of reasons. In an effort to better understand the informal language usage phenomenon in France, it may be helpful to consider informal language user profiles, distinguished by reasons for participating (leisure, entertainment, communication, learning), nature of activity (comprehension/reception or production/interaction), and level of effort made to learn or remember new words and structures. As Jérôme has shown, different activities represent different types of usage, goals, and levels of attention, all of which can impact potential learning. For future studies it would be useful to examine passive versus active styles of informal L2 usage.

Conclusion

Informal language learning can take many forms. This chapter has shown how one particular form, the online informal learning of English (OILE), currently takes place amongst university students in France. It is a phenomenon that cannot be characterized by one set of practices or activity and is motivated for the most part by reasons other than an explicit intention to learn the language. As research moves forward, it will be important to consider specific contextual issues, such as French pupils' results in European foreign language assessments or the stigma associated with foreign language learning in France, when examining what role informal L2 exposure and use might play in the equation. The case study results indicate that L2 development in the informal sphere is complex and dynamic. For now, only the tip of the informal iceberg has been explored and a great number of questions and directions for future research remain. It will be the task of informal L2 researchers to identify robust and reliable methods of measuring informal L2 change that can be reproduced in future studies.

REFERENCES

  1. Brudermann, C. and Poteaux, N. (2013). Enseignement des langues à l'université et TIC: enjeux actuels et exemples de dispositifs. Apprendre avec le numérique dans l'enseignement supérieur au 21ème siècle [University language teaching and ICT: Current issues and examples of digital tools: Learning with technology in higher education in the 21st century]. https://www.canal‐u.tv/video/universite_paris_1_pantheon_sorbonne/enseignement_des_langues_a_l_universite_et_tic_enjeux_actuels_et_exemples_de_dispositifs.11778.
  2. de Bot, K. and Larsen‐Freeman, D. (2011). Researching second language development from a dynamic systems theory perspective. In: A Dynamic Approach to Second Language Development: Methods and Techniques (eds. M. Verspoor, K.D. Bot and W. Lowie). Amsterdam/Philadelphia: John Benjamins Publishing.
  3. Directorate‐General Education and Culture (European Commission). (2013). Study on the Use of Subtitling. https://publications.europa.eu/en/publication‐detail/‐/publication/e4d5cbf4‐a839‐4a8a‐81d0‐7b19a22cc5ce/.
  4. Ellis, R. and Barkhuizen, G. (2005). Analyzing Learner Language. Oxford: Oxford University Press.
  5. European Commission. (2012). First European survey on language competences: Final report. https://crell.jrc.ec.europa.eu/sites/default/files/files/eslc/ESLC_Final%20Report_210612.pdf.
  6. Fleurot, G. (2013). Les Français sont vraiment nuls en anglais. [The French are no good at English]. Slate. http://www.slate.fr/story/68577/francais‐nuls‐anglais.
  7. Foster, P. and Skehan, P. (1996). The influence of planning and task type on second language performance. Studies in Second Language Acquisition 18 (3): 299–323.
  8. Halimi, S. (2012). Apprendre les langues Apprendre le monde [Learn languages Learn the world]. Rapport présenté au ministre de l'éducation, de la jeunesse et de la vie associative.
  9. Kusyk, M. (2017). Les dynamiques du développement de l'anglais au travers d'activités informelles en ligne: une étude exploratoire auprès d'étudiants français et allemands [The dynamics of English development through informal online activities: an exploratory study of French and German students]. University of Strasbourg/Karlsruhe University of Education. https://phka.bsz‐bw.de/frontdoor/index/index/docId/116.
  10. Kusyk, M. and Sockett, G. (2012). From informal resource usage to incidental language acquisition: language uptake from online television viewing in English. ASp, la revue du GERAS 62: 45–65.
  11. Lam, W.S.E. (2000). L2 literacy and the design of the self: a case study of a teenager writing on the internet. TESOL Quarterly 34: 457–482.
  12. MacWhinney, B. (2000). The CHILDES Project: Tools for Analyzing Talk, 3e. Mahwah, NJ: Lawrence Erlbaum Associates https://talkbank.org/manuals/CLAN.pdf.
  13. Norris, J.M. and Ortega, L. (2009). Towards an organic approach to investigating CAF in instructed SLA: the case of complexity. Applied Linguistics 30 (4): 555–578.
  14. Ortega, L. (2015). Usage‐based SLA: a research habitus whose time has come. In: Usage‐based Perspectives on Second Language Learning (eds. T. Cadierno and S. Eskildsen), 353–373. Berlin: De Gruyter Mouton.
  15. Rojas, D. (2016). Les Français et les langues étrangères [The French and foreign languages]. La Revue des Ressources.https://www.larevuedesressources.org/les‐francais‐et‐les‐langues‐etrangeres,2920.html.
  16. Schugurensky, D. (2007). « Vingt mille lieues sous les mers » : les quatre défis de l'apprentissage informel” [Twenty thousand leagues under the sea: the four challenges of informal learning]. Revue française de pédagogie 160: 13–27.
  17. Sinclair, J. (1991). Corpus, Concordance, Collocation. Oxford: Oxford University Press.
  18. Skehan, P. (1998). Task‐based instruction. Annual Review of Applied Linguistics 18: 268–286.
  19. Skehan, P. (2001). Tasks and language performance. In: Researching Pedagogic Tasks: Second Language Learning, Teaching and Testing (eds. M. Bygate, P. Skehan and M. Swain). Harlow: Longman.
  20. Sockett, G. (2011). From the cultural hegemony of English to online informal learning: cluster frequency as an indicator of relevance in authentic documents. ASp, la revue du GERAS 60: 5–20.
  21. Toffoli, D. and Sockett, G. (2010). How non‐specialist students of English practice informal learning using Web 2.0 tools. ASp, la Revue du GERAS 58: 125–144.
  22. Toffoli, D. and Sockett, G. (2013). University teachers' perceptions of Online Informal Learning of English (OILE). Computer Assisted Language Learning 28 (1): 7–21.
  23. Toffoli, D. and Sockett, G. (2015). L'apprentissage informel de l'anglais en ligne (AIAL): quelles conséquences pour les centres de ressources en langues?” [Online informal learning of English (OILE): What are the consequences for language resource centers?]. Recherche et pratiques pédagogiques en langues de spécialité. Cahiers de l APLIUT XXXIV (1): 147–165.
  24. Tomasello, M. (2009). The usage‐based theory of language acquisition. In: The Cambridge Handbook of Child Language, 69–88. Cambridge: Cambridge University Press.
  25. van Geert, P. (2009). Systèmes dynamiques et modélisation dans l'étude de l'acquisition du langage: tendances, transitions et fluctuations. {Dynamic systems and modeling in the study of language acquisition: trends, transitions and fluctuations]. In: Apprentissage des langues (eds. M. Kail, M. Fayol and M. Hickman). Paris: CNRS Éditions.
  26. Verspoor, M., Schmid, M.S., and Xu, X. (2012). A dynamic usage based perspective on L2 writing. Journal of Second Language Writing 21 (3): 239–263.
  27. Verspoor, M. and van Dijk, M. (2011). Visualizing interactions between variables. In: A Dynamic Approach to Second Language Development (eds. M. Verspoor, K. de Bot and W. Lowie). Amsterdam/Philadelphia: John Benjamins.
  28. Wolfe‐Quintero, K., Inagaki, S., and Kim, H.‐Y. (1998). Second Language Development in Writing: Measures of Fluency, Accuracy and Complexity. Honolulu: University of Hawai'i Press.

Notes

  1. 1 This refers to exposure on a systematic level; certainly many French children also come into contact with additional languages and cultures through familial settings.
  2. 2 The B1 and B2 levels are “independent user” levels and refer to the Council of Europe's Common European Framework of Reference for Languages (CEFR): https://www.coe.int/en/web/common-european-framework-reference-languages/home.
  3. 3 An outline of the French government's foreign language programs in public schools is available here: http://www.education.gouv.fr/cid206/leslangues-vivantes-etrangeres.html#Au_lyc%C3%A9e%20g%C3%A9n%C3%A9ral,%20technologique%20et%20professionnel.
  4. 4 Participants were students in the last grade of middle school (14–15 years old), which corresponds to France's targeted B1 level. French students were rated at the following levels: 31% pre-A1, 40% A1, 15% A2, 9% B1, 5% B2. Source: http://www.surveylang.org/media/ExecutivesummaryoftheESLC_210612.pdf.
  5. 5 See Rojas (2016) and Fleurot (2013) for discussions in popular culture media; Brudermann and Poteaux (2013) discuss the cliché in an academic setting; Halimi (2012) reassures in an official report that “the French are indeed capable of learning multiple languages” (p. 21).
  6. 6 The names of popular television series were noted by students in Toffoli and Sockett's 2010 questionnaire study. The five selected for the corpus were House, How I Met Your Mother, One Tree Hill, Lost, and Desperate Housewives. One year's worth of episodes (70 hours or 500 000 words of dialogue) were analyzed.
  7. 7 Participants in the study were asked to indicate one of four rates of viewing: never, rarely (less than once a month), sometimes (between once a month and once a week), and regularly (more than once a week). If they answered “regularly,” a follow-up question asked them to indicate how many hours per week; on average this amounted to four hours per week.
  8. 8 This term refers to students who major in a field other than English.
  9. 9 The questionnaire was disseminated in 2014 and contained 61 questions – full results can be found (in French) in Kusyk 2017.
  10. 10 A1 – A2 are “basic users,” B1 – B2 are “independent users” (as mentioned in footnote 2), and C1 – C2 are “proficient users.” See the CEFR link in footnote 2 for a more detailed description of the language levels.
  11. 11 Two additional response categories were offered with these questions: “not applicable” and “other,” which were chosen by 11% and 3% of the sample, respectively. The questionnaire did not include a question on the modality of video watching, as at the time of dissemination it was not common to be able to activate subtitles for general video viewing.
  12. 12 In addition: 40% noted no change, 17% were not concerned and 3% selected “other”. The changes in viewing mode for films were essentially identical.
  13. 13 The case study participant (see following section), Jérôme (a pseudonym), clarified during interviews that the presence of subtitles in series or film viewing was not always the result of an active decision on his part; sometimes they were already integrated into the streaming file and neither activation nor deactivation were possible. As a result, both of the following interpretations of the data from the questionnaire may be applicable: the students actively seek out files with integrated subtitles with the intent of using them as a visual aid or the students do not necessarily need the subtitles (or do not want them) but are unable to find streaming services offering non-subtitled files.
  14. 14 The categories available for selection for the questions “why do you participate in … in English?” were based on responses from previous and pre-test questionnaires in which all categories were open. For the current questionnaire, an open option was nevertheless provided in case students did not find any of the proposed reasons applicable. Participants were allowed to select as many categories as they wished.
  15. 15 “La langue n'est pas un frein, l'important est le sujet” (Kusyk 2017, p. 162).
  16. 16 See for example Verspoor et al. 2012; Tomasello, 2009; de Bot and Larsen-Freeman 2011; Ortega 2015.
  17. 17  A selection of the results is presented in this chapter. Full results can be found in Kusyk 2017.
  18. 18 In the dynamic usage-based perspective, “supportive growers” (variables that grow together) are expressed through a positive correlation while “competitive growers” are expressed through a negative correlation. Here, informal activity frequency (number of hours) was used to calculate the correlations. Spearman's correlation (rather than Pearson's correlation) was used as it is better adapted for data sets that are nonlinear, have outliers, and do not have normal distributions. Because of a low n (8, number of interviews), the results from the correlation tests are to be interpreted prudently, and can be seen as potentially fruitful points of departure from which further relational analyses can be conducted.
  19. 19 Conditional growers is a term used in the dynamic usage-based literature (see van Geert 2009; Verspoor and van Dijk 2011) which signifies that a minimum level of growth in one variable is necessary before another variable can achieve substantial growth.
  20. 20 Overall Jérôme dedicated 163 hours to music listening, 66.5 hours to reading. and 50 hours to series watching.
  21. 21 See Norris and Ortega 2009; Wolfe-Quintero et al. 1998. In short, some researchers believe that length of T-units is a complexity measure as it is assumed that the longer the T-unit is, the more complex it is because increased length would, in this case, signify added subordinate clauses. Other researchers believe that it belongs to fluency as increased length signifies more words, which is one of the definitions of written fluency.
  22. 22 The prioritization of one measure (complexity, accuracy, or fluency; CAF) over another, whether conscious or not, is a fundamental aspect of CAF research field (see Wolfe-Quintero et al. 1998; Skehan 1998; R. Ellis and Barkhuizen 2005; Skehan 2001; Foster and Skehan 1996).
  23. 23 Lexical complexity was made up of two submeasures: lexical diversity, represented by VOCD, a calculation generated by the language analysis software CLAN (MacWhinney 2000, p. 135) and lexical sophistication, represented by the Vocabulary Profiler created by Tom Cobb (http://www.lextutor.ca/vp).
  24. 24 The corpus used for these frequency bands was the BNC-COCA 1-25 K: the 25 000 most frequent word families from both the British National Corpus (100 000 000 oral and written words) and the Corpus of Contemporary American English (version 2012, 450 000 000 oral and written words).
  25. 25 Frequency bands were combined (such as K5 and K6) so as to render the dissertation graphics more legible (see Kusyk 2017). The usage of words in the K7 – K15 frequencies was rare and it made sense to combine them rather than have eight additional lines on a graph.
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
3.138.113.188