The following passage expounds on the current concern about L2 teaching.

Langacker emphasized that meaning is more important than form and that this seems natural from the perspective of students. Nobody would deny that L2 students tend to enjoy understanding the meaning rather than the syntactic structures of utterances. Langacker (2008) stated that grammar has its own meaning. L2 students will wonder how basic grammatical concepts such as nouns, verbs, and subjects can be semantically explained. Langacker said nouns refer to things, verbs refer to processes, and so on. Likewise, cognitive semantics does not have a modular view of language. In other words, it believes that all structures in language are basically meaningful (Lemmens, 2016). Wulff and Nick (2018) also argued that grammar has meaning. For example, they said that the simple morpheme aholic means “being addicted to something,” and the syntactic structure subject-verb-object-object means “something is being transferred” (p. 38). Thus, grammatical meanings sound somewhat strange, but they differ from lexical meanings only in degrees.

This argument benefits L2 learners in several ways. First, the distinction between grammar and lexicon has become less justifiable. Since L2 learners have difficulty in studying grammar, this message helps to reduce their burden of grammar learning. Lewis (1993) emphasized that “Language consists of grammaticalized lexis, not lexicalized grammar” (p. vi). In other words, the lexis plays a primary role in making meaning (Realm, 2009).

Cognitive linguistics argues that language should be explained in only one way. Lemmens (2016) stated that “All structures in language, ranging from morphemes to words to syntactic patterns, are considered […] inherently meaningful” (p. 90). Given that idiomatic and idiosyncratic expressions are neglected in generative grammar (Taylor, 2012), cognitive linguistics can be more reasonable in answering language phenomena.

The term convention means habit. A linguistic convention refers to a collective habit of a speech community. In other words, members of a speech community expect other members to use a language in the same way that they do, such as when they greet each other in the morning with good morning. Technically speaking, they tend to “conform to regularities in their use of utterance types” (Schmid, 2020, p. 87). Kádár and House (2021) defined convention as “shared understandings of meanings and practices” (p. 162). Naturally, members in the same speech community are supposed to encounter conventions every single day. When L2 learners go abroad to learn a second language, they learn these conventional things.

The EFL classroom is concerned with how to teach conventional expressions. Bardovi-Harlig and Vellenga (2012) stated that “Conventional expressions consist of strings such as no problem, nice to meet you, and that’s great, which native speakers use predictably in certain contexts” (p. 77). They said that even advanced students have difficulty in learning conventional expressions. Their experiment showed that instruction in acquiring conventional expressions is helpful because conventional expressions require pragmatic knowledge, which makes learning without instruction very difficult.

Dynel (2017) said that in textbooks for teaching conventional expressions in the classroom, fictional discourse can be used without a problem.

Dynel (2017) clearly stated that film-based materials can be authentic to researchers and L2 students. Even he used a medical drama, House MD (Shore, 2004-2012), in his experiment. He implied that characters’ interactions are more powerful than those of ordinary people. Because conventional utterances have high communicative value, interactions may come from authentic materials such as films.

Finally, conventionalization can be defined as a process that impacts the degree to which people in the same community conform to regularities while they are using utterances in their conversation (Schmid, 2020). If the degree is high, conventionalization becomes convention.

When we humans speak and listen, we need world knowledge as well as linguistic knowledge (Willits, Amato, & MacDonald, 2015). In other words, the integration of linguistic knowledge and world knowledge is necessary to achieve successful communication. Louwerse’s (2008) experiment showed that listeners respond more quickly to word pairs (attic-basement) than to reverse pairs (basement-attic). From the perspective of linguistics, attic-basement or basement-attic are the same descriptions of a building in a vertical spatial arrangement. However, people in the real world prefer attic-basement pairs to reverse pairs just because they are more comfortable with attic-basement pairs. Beare (2019) said that words such as Adam and Eve, bacon and eggs, cause and effect, crime and punishment, dead or alive, fish and chips, and in and out, which have nonreversible order, are like collocations because they always go together. Beare stated that the reason for their being in that order is not clearly linguistic. The answer could be found in world knowledge. Technically speaking, interpersonal pragmatics deals with the importance of world knowledge. It asks why the same cognitive behavior is habitually and systematically presented in different linguistic ways (Locher & Graham, 2021). It is the responsibility of language users to choose appropriate linguistic forms that refer to conventional expressions.

In an EFL film-based classroom, interpersonal pragmatics should be considered truthfully. As is known, films reflect the virtual world, where a lot of interactions occur. First, L2 learners must perform L1 translation in order to understand the pragmatic meaning of words. Despite the blurring of the distinction between language knowledge and world knowledge, many L2 learners are not aware of the fact. When L2 students translate a certain film into L1, they tend to ignore L1 translation. If this happens, they may lose the communicative value of interactional utterances of characters. Therefore, teachers should guide their students to translate with caution.

Finally, the context of world knowledge can be generally defined as “the actual situational context of the linguistic sign(s) or utterance” (Kecskes, 2021, p. 605). The definition reflects that only linguistic meaning cannot be interpreted. Goodwin and Duranti (1992) stated that context is “a frame that surrounds the event and provides resources for its appropriate interpretation” (p. 3). It also implies that context is a frame for interpreting linguistic meaning. Gawron (2019) explained frames in greater detail as follows: “Frames are conceptual structures that provide context for elements of interpretation; their primary role in an account of text understanding is to explain how our text interpretations can leap far beyond what the text literally says” (p. 57). Taken together, L2 students should be aware of the integration of linguistic forms and relevant world knowledge. Considering that films portray a virtual world, they should not lose the world referents.

Long-time language teachers have found that when they used films in the classroom, their students were reluctant to actively participate in activities (Lee, 2020). Film-based materials are expected to make students learn a language with fun, but the reverse often happens. Students seem to be divided between those who merely enjoy watching films and those who watch films to learn a language. Few students are inclined towards the latter. Nevertheless, it has been found that films are the best materials for learning a second language.

In this section, we investigate what kind of expressions L2 learners prefer while they are using TV series materials. They often tend to be disappointed with the results of their work as memorization is fruitless for them. To help the participants overcome such problems, some points were emphasized to them. As mentioned in the literature review, they were instructed on (1) the centrality of meaning, particularly of grammatical meaning; (2) conventional utterances; and (3) world knowledge.

Also, theoretical positions will be represented. They will take the rationale of a usage-based approach in which usage events or actual utterances are keys to improving L2 proficiency. Construction grammar will be particularly considered. As is known, students in film-based classrooms prefer idiomatic expressions. Since it is believed that linguistic structure consists of constructions, including idiomatic expressions, construction grammar belongs to the usage-based approach (Diessel, 2015; Hilpert, 2020). The last position is the meaningful learning theory of Ausubel (1963). When conventional utterances in film scripts are memorized, there should be a meaningful strategy for keeping them in one’s long-term memory.

This study has three participants, and they are college students. According to their TOEIC test results they were all intermediate level. All of them have some experience using films (movies and TV series). In this study, they were asked to watch the American TV series Good Witch, season 1, episode 1 (Pryce, 2015), episode 2 (Lynn, 2015a), and episode 3 (Lynn, 2015b) over more than three months.

Good Witch was chosen because of its story (contextual meaning) and its characters’ utterances (linguistically and ethically good forms). Its main character is Cassie, who has magical powers. However, she does not scare her neighbors away. She regards magic as a gift, not power. She always uses it to help people. Her magic makes her neighbors feel comfortable. She is always thoughtful and sweet, as reflected also in her utterances.

From the perspective of L2 learning, Cassie’s and the other characters’ utterances may give prominence to L2 learners in that her message reflects a sort of touch to L2 learners. Prominence or salience is known to be related to successful L2 learning. This is why Good Witch is considered an appropriate material for L2 learning.

The researcher instructed the participants to read the scripts and then submit expressions (words, phrases, and clauses) that they wanted to learn. He also told them to indicate why they preferred such expressions. In the first activity, after the researcher analyzed the participants’ submissions, he instructed them to study the scripts with a focus on studying the scripts. Vujic (2016) said, “In construction grammar, linguistic expressions are not necessarily syntactic forms” (p. 11). Simply put, every expression can be a construction. A word, complex words, phrases, and clauses are all constructions. Lexical items are constructions, and they are represented partially or fully according to their degree of specificity or schematicity (see Table 1; Fried, 2015, p. 978; cited in Vujic, 2016, p. 11). Syntactic patterns are described in the same way as lexical items.

Table 1 allows for better understanding of schematicity and specificity. Schematicity refers to symbols without concrete words. The second-to-the-last examples in the right column of the table show no word except symbols such as NP and VP. Specificity is the reverse. It consists of content words. In the first examples in the right column of the table, only words are listed. Each construction is represented by degrees of schematicity and specificity. The participants were asked to indicate the differences between the constructions according to their degrees of schematicity and specificity. They were not asked to distinguish lexical items and grammatical patterns. Because meaning is central, it can be a basic unit for describing each construction.

In the second activity, the students were tasked to learn conventional expressions in the script of Good Witch. This is one of the main concerns in this section. The participants had to distinguish between conventional expressions and non-conventional expressions. Conventional expressions are a subset of multiword units. However, the participants did not need knowledge about multiword units. Instead, they had to perform first-language (L1) translation. When they could translate an utterance as one word, it was a conventional expression. For example, good morning can be translated as one word in L1, so it is a conventional expression.

The important thing that the students had to keep focusing on was the differences between linguistic translation and pragmatic translation. Many Korean college students do not know how to distinguish between these two types of translation. Linguistic translation uses the dictionary meaning of an expression, whereas pragmatic translation uses its contextual meaning. More natural translation requires awareness of an expression’s pragmatic translation.

The third activity was about world knowledge. As mentioned earlier, language use needs to combine language knowledge and world knowledge. For better understanding, one example dialogue2 is presented below.

In translating the above dialogue, some background about the two characters is needed. The first background is who Ann and Sarah were. The second is why they mentioned the word lisp. It is common knowledge that L2 learning is not possible with only a dictionary. Therefore, in the third activity, the participants were expected to be context-sensitive. Their reports had to include contextual information.

In the first activity, the participants were supposed to add the degree of schematicity or specificity. Of course, they did not have to use such terminologies. Simply, they had to write the reason why they picked a certain expression, and then the teacher analyzed their reasons. Table 2 shows the results of the first activity.

The participants did not show an ability to analyze schematicity and specificity. They misunderstood meaning as the meaning of the content. Even if they understood the expressions correctly, they probably could not answer them. Despite their mistakes, their data deserve some comment. Participant A intended to learn the meaning of take. He gave one example of this. Since he gave only one example, he did not generalize the meaning of take. Therefore, his expression was forced into the specificity domain. Moreover, he focused on only one word: take. This is not a conventional expression, as conventional expressions should be at least strings of words (Bardovi-Harlig, 2009, 2012, 2014; Edmonds, 2014). Participant B intended to show why he picked It’s no secret. He stated the reason in L1. However, he failed to show whether his chosen expression was generalizable partly or fully (a schematic matter) or specifiable (a specific matter). However, his selection was good in one sense: he selected the conventional expression It’s no secret. It consists of four words but functionally behaves like one word. Participant C’s selection also deserves comment. She tried to generalize left out with examples such as the following.

Certainly, she showed similar situations or uses of the phrasal verb leave out. She tried to generalize the phrasal verb in several situations. This could be in the schematic domain.

In sum, the three participants were not successful in improving the degrees of schematicity and specificity of their chosen expressions. They did not yet understand that language should be interpreted in one way (i.e., with one meaning). However, had they been trained as English major students, they would have had a deeper insight into language.

This activity was the most important for the participants because they wanted to learn as many expressions as possible. Such expressions were primarily idiomatic, and the participants often failed to form sentences with their grammatical knowledge. The results of this activity are shown in Table 3.

In this activity, the participants performed well. Participant A deleted the linguistic meaning of I guess. In the Korean usage of I guess, people do not explicitly express the linguistic meaning. They prefer implicit translations. Participant A followed the Korean usage successfully. It sounded natural. The phrasal verb bring up has several linguistic meanings, but Participant A took up a new meaning, upset, having considered the context. In a sense, he conquered his language. He created a new meaning. This is the right way to improve language proficiency. When an adult asks a child who does not know letters What is this?, pointing to a book, the child will probably simply say Book, not the title of the book. In his view, the context is that he happens to see a book. So, he uses his experience in answering the question. On the other hand, adults often say, I don’t know. They seem to be slaves of a language. In this sense, Participant A did a very good job.

Participant B had to interpret 3:15. What he knew as the context was that the conversation had happened in the clinic. He had to use the context to interpret 3:15. Unfortunately, he failed to try. Though the teacher gave him some cues, he was not successful. In other words, he was not context-sensitive.

Participant C selected Put it back before someone notices. She was good at translating the first part, Put it back. Because the L2 usage and the L1 usage were very similar, she quickly responded to the L1 translation. The problem was the second part: before someone notices. She translated notices with its linguistic meaning. It sounded awkward. She could have used her imagination in such a situation. Still, she could not think of the appropriate Korean usage and she gave up.

In sum, Participant A was excellent at using context. He did very well in extending the linguistic meaning into the pragmatic meaning. Participant B, unlike Participant A, was logical in his thinking, but he could not picture a virtual world. Participant C was in between Participants A and B in this regard.

This activity showed that films, including movies and TV drama series, cannot be interpreted without appropriate context. Social interaction is not exceptional. Therefore, teachers must instruct and train their students better in using films for L2 learning.

Most EFL students tend to think that they must obviously focus on language matters in their L2 learning. When a film is used in the EFL classroom, some students often complain about a lack of language focus. As mentioned earlier, people use language to talk about the real world, not a language.

In the third activity, we investigated how much non-linguistic matters the participants were involved in. When they handed in their reports, the teacher asked them contextual questions based on the information implied in their selected conventional expressions. The results are shown in Table 4.

Participant A selected the conventional expression You’re the last person I would have expected to be here. He explained the last person with the following sample sentence: This is the last thing your father would have wanted. He focused on a language matter, the last person and would have + p.p. This was natural. The teacher asked him, “When you do something wrong in school, who will take care of your punishment?” The teacher and Participant A had a talk about this. After the conversation, Participant A confessed that their conversation gave salience to the form “would have expected.” He thought that such conversation was not relevant to language learning in the first place, but he changed his mind because of its salience.

The teacher asked Participant B the same question that he asked Participant A. He continued to ask him whether “Anything that happens here” has a positive meaning or a negative meaning. Also, he asked again, “If it is negatively expressed, what would the reason be?” Participant B made one comment: he said he was surprised that with nonlinguistic knowledge, it was possible to learn linguistic knowledge. Participant C selected I got George in my office. The teacher asked her two questions: first, why “my office” was used instead of the hospital; and second, why “got” was used. She did not answer the first question, but she answered the second question correctly. Because office and hospital are both places, got was easily inferred.

In sum, world knowledge or contextual knowledge intervened and helped the participants to understand the meanings of other lexical items. Also, such knowledge gave salience to conventional expressions, and consequently, the participants were helped to improve their language learning. In the first activity, the participants performed poorly. Had they handled the first activity well, they could have learned grammatical knowledge. Without separating lexicon and grammar, they would have accessed grammar. In the second activity, they performed relatively well. In the third activity, they needed their teacher’s assistance.