Teacher Directives in Children’s Musical Instrument Instruction: Activity Context, Student Cooperation, and Institutional Priority

ABSTRACT

Identifying precisely what teachers do to elicit desired changes in their students’ knowledge and skill is a long-lasting challenge of educational research. Here, we use conversation analysis to contribute to a deeper understanding of this matter by considering how Finnish-speaking musical instrument teachers use directives to guide their students. Our data consist of 10 video-recorded instrument lessons (violin, piano, guitar, and ukulele). In our findings, we provide an account for the variance in the musical instrument teachers’ use of six second-person directive forms in Finnish. We argue that the teachers’ choices between these directive forms are warranted by three dimensions of the participants’ conduct: (1) location of the directive within the participants’ wider activity structure, (2) degree of the student’s cooperation at the given moment, and (3) the institutional priority of action that is being called for.

KEYWORDS:

• Directives
• music education
• music instrument instruction
• conversation analysis
• Finnish

This paper examines teacher directives in music instrument instruction. More precisely, we are looking into the teachers’ choices of directives, suggesting these choices to be warranted by three dimensions of the participants’ conduct: (1) activity context: how the directive is positioned in the on-going activity, (2) student cooperation: to what extent the student co-operates with the teacher, and (3) institutional priority: to what degree the directive relates to the institutionally central content matter that is currently being taught. The theoretical framework for the examination comes from conversation analysis and the educational examination of music pedagogy, as embedded in the institutional setting of instrument instruction.

Conveying Musical Knowledge and Skill

Even thorough knowledge of a specific subject matter or skill does not ensure that a person is successful in communicating that knowledge or skill to others. Besides content knowledge – as expertise in the subject matter – effective teachers have a distinct relationship with the taught content; they know how to combine their students’ previous understanding and skills to build their learning, and to convey the subject matter in a comprehensible way (Millican, 2013). This pedagogical content knowledge, according to Shulman (2013), “goes beyond knowledge of subject matter per se to the dimension of subject matter knowledge for teaching  …  the ways of representing and formulating the subject that make it comprehensible to others” (pp. 6–7). Although research on teachers’ pedagogical content knowledge in relation to music teaching is rather scarce, the topic has been studied in relation to the teaching of other subjects.

The studies that do examine pedagogical content knowledge in relation to music instruction have often highlighted the importance of modeling – that is, the teachers’ ways of conveying the expressive elements of music through singing, gesturing, or playing. In line with that, the role of the body’s corporeal significance in learning in educational settings has also been increasingly acknowledged. Thus, Evans, Davies, and Rich (2009) have analyzed the relationship between the body and education by linking the agency of the former in cultural reproduction and discussing how the corporeal realities of children influence their values, self, and sense of position. The notions of modeling and embodied learning are particularly relevant in the context of music instrument learning, as it has much to do with physical actions – for example, how to hold and move the bow to get a clear sound from a cello.

How is musical knowledge and skill then best communicated? In a pre- and post-instruction study, Dickey (1991) compared the effectiveness of modeling and verbal instruction in music classrooms, finding support for the hypothesis that modeling is more effective than verbal instruction in conveying musical knowledge and skill. For example, he found that the use of teacher-demonstration–student-imitation cycles to teach melodic patterns significantly helped the development of the students’ ear-to-hand coordination abilities. Duke and Simmons (2006), who observed the video-recorded teaching of three expert college music pedagogues, described exquisite modeling as an important component of successful music instruction. For instance, they saw the teachers to juxtapose “a remarkably faithful imitation of the student’s performance with their intended model of the performance goal, evincing a definitive level of technical command and fluency” (p. 159).

Yet music education research has shown that mere modeling is seldom enough to count as a successful act of instruction. For example, drawing on ethnomethodology, Weeks (1996) described the “illustrative expressions” (singing, humming, and chanting) through which conductors in orchestra rehearsals demonstrate their views on the music to be played, noting that the specific sense of these illustrations needs to be articulated verbally. Likewise, in a systematic observation study, Duke (1999) investigated teachers’ and students’ behavior in Suzuki string lessons, finding high proportions of teacher talk (65% of instructional time) as a crucial element of instruction. Hence, although musical instruction relies also on means other than verbal conduct, the central role of what the teacher says during the instruction cannot be undermined, and, arguably, the linguistic details of verbal instruction – as embedded within their immediate embodied action environment – can have a notable influence on student learning and behavior.

Verbal Instruction

In music instrument instruction, as in other educational contexts, verbal instructional utterances have been shown to fill various functions, ranging from feedback, advice, and correction to displays of enthusiasm, empathy, and authority (e.g., Duke, 1999; Duke & Henninger, 2002). While one important aspect of the teachers’ pedagogical skill is to tailor their instructional utterances so as to take into consideration the differences between individual learners (see Li, 2015), in this paper,we will describe a range of other aspects that skillful teachers orient to when instructing their students. Since such pedagogical choices may not always reach the teachers’ level of reflexive consciousness (e.g., Clark, 1980; Kuusisto & Lamminmäki-Vartia, 2012), it is important to examine their corollaries in the teachers’ and students’ actual conduct in order to get a thorough understanding of their pedagogical value.

Verbal instruction has been particularly extensively studied in conversation analysis (CA). Instructional activities have their characteristic “interactional architecture” (Seedhouse, 2004) and many CA researchers have a long-standing interest in examining the details of that organization. Conversation analysis studies have, for example, addressed the ways in which the fundamental conversational structures, such as turn-taking and sequence organization, are applied to instructional purposes (De Stefani & Gazin, 2014; Lerner, 1995; Macbeth, 1991; McHoul, 1978; Weeks, 1985). The initiation-response-evaluation sequence has been identified as one of the most essential interactional structures between teachers and learners, and a large body of studies exists on how that structure is oriented to and used in the context of various instructional activities (e.g., Mehan, 1979, 1985; Waring, 2009; Zemel & Koschmann, 2011). There is also much literature on the practices of correction (e.g., Deppermann, 2015; Lindwall, Lymer, & Greiffenhagen, 2015; Lindwall & Ekström, 2012), which provide “a valuable heuristic device for elucidating aspects of the distinctive orderliness of activities to which they are adapted or addressed” (Weeks, 1985, p. 232).

In CA, instructional activities have been examined in a great variety of settings, such as classroom (e.g., St. John & Cromdal, 2016), dance lessons (Keevallik, 2010, 2013), surgical training (Zemel & Koschmann, 2014), pre-clinical dental training (Hindmarsh, Hyland, & Banerjee, 2014), and driving instruction (Broth, Cromdal, & Levin, 2017; De Stefani & Gazin, 2014; Deppermann, 2015; Rauniomaa, 2017). Several CA studies have also been conducted in musical settings, such as orchestra rehearsals (Parton, 2014; Weeks, 1996), choir rehearsals (Merlino, 2014), vocal master classes (Reed & Szczepek Reed, 2014; Szczepek Reed, Reed, & Haddon, 2013), composition workshops (Veronesi, 2014), and instrument lessons (Nishizaka, 2006). As for the details of verbal instructional utterances, CA studies have pointed to the instructors’ systematic orientations to the learners’ level of performance (see, e.g., Vehviläinen, 2009) and described the instructors’ detailed ways of tailoring their instructions so as to pre-empt the learners’ resistance (see, e.g., Vehviläinen, 2012) or to invoke the learners’ momentary rights and responsibilities (Rauniomaa, 2017).

There are also educational researchers who have used CA to address the issue of learning per se – something that is ultimately the aim of all instructional activities. From the point of view of CA, which targets the micro level of the turn-by-turn unfolding of interaction, such focus is a challenge – given that, typically, or at least from the traditional cognitive-mentalist perspective (see, e.g., Doughty & Long, 2003), learning is understood as changes occurring on a macro level (Rusk, Pörn, Sahlström, & Slotte-Lüttge, 2015, p. 42). One way of dealing with this challenge has been to identify systematic changes in the participants’ use of sequential structures over a longitudinal period of time and consider these changes as learning (e.g., Brouwer & Wagner, 2004; Martin & Sahlström, 2010). Another possibility suggested in the literature is to consider learning as social action accomplished in situ (Sahlström, 2011). This insight has encouraged, for example, analysis of how participants link what is said or done to past and future occurrences, topicalise understanding, orient to knowledge asymmetries (Sahlström, 2011), or ratify each other as knowing participants within a group (Melander, 2012).

Conversation analysis has also been regarded as a powerful approach to study socialization within a sociocultural perspective (Carlgren, 2009; Sahlström, 2009). In this view, learning may be captured through the concept of participation (Lave & Wegner, 1991) and “conceived as a process of becoming member of a certain community” (Sfard, 1998, p. 6). Also here, the analysis of instructional sequences can be fruitful (see, e.g., Lee, 2010). While instruction is implemented in the learner’s (instructed) responsive action, its success relies on his or her competent grasping of what the instructor is up to (Mondada, 2014b, p. 134). The situated sense of an instruction is thus elementarily tied to its surrounding activities and context, and the learner’s growing ability to perceive instructions as intelligible is an important aspect of socialization. Furthermore, as demonstrated by Broth et al. (2017), instruction is not only about informing the learner what needs to be done but also about conveying how to do it and why. Learners are thereby socialized into specific forms of reasoning, perceiving, and acting as competent members of a group (about “professional vision,” see Goodwin, 1994) and – as we will argue in this study – much of this can happen implicitly through the subtle nuances in the teachers’ verbal instructional utterances.

Verbal instruction can take a multitude of linguistic forms. In this paper, we will consider verbal instruction in the form of directives. The term directive is inherited from speech act theory and refers to an act prompting the hearer to take a particular action (e.g., Searle, 1976). Still, directives can also be implemented in a multitude of ways. Directives can take the form of different types of declaratives (e.g., now you’ll take the bow), imperatives (e.g., take the bow), or interrogatives (e.g., will you take the bow). Recently, within CA, there has been a growing interest in how different directive forms configure in the management of practical action. The selection of the imperative form, for example, has been linked to the material and temporal context of local “here and now” cooperative activities (Goodwin & Cekaite, 2013; Keisanen & Rauniomaa, 2012; Mondada, 2014c), where the question about the fit of the requested action with the trajectory of what the recipient is currently doing (Rossi, 2012; Wootton, 1997; Zinken & Ogiermann, 2011, 2013) plays a significant role. The speaker’s selection between different directive forms has also been shown to be influenced by contextual considerations, such as the authoritative status of the speaker in relation to the recipient (Stevanovic & Peräkylä, 2012, 2014). In our analysis of verbal instruction during music lessons, we will thus also draw on this body of CA literature

In this paper, we will examine how music teachers design their directives to students in relation to various contingencies in the participants’ interactional conduct during instrument lessons. More specifically, our research question is: What warrants the use of different forms of directives for Finnish-speaking instrument music teachers? Our results will be subsequently considered with reference to what they tell us about music instrument instruction as a pedagogical institution and about music learning more generally.

Second-Person Directive Forms in Finnish

In this study, we seek to provide an account for the variance in the Finnish music instrument teachers’ use of six different directive¹ forms. Linguistically, all these forms are in the second person – that is, they are used by the speaker to tell the recipient what to do. On the basis of both earlier research and common intuitions, they come across as relatively interchangeable. To be able to appreciate the apparent initial similarity of these six forms, we will first explain their linguistic features.

In Finnish, similarly to many other languages, the most stereotypical and routine way to tell another person what to do is through turns that contain the finite verb in the imperative mood (e.g., laita se sinne “put it there”). In Finnish, the analysis of such turns also needs to take into consideration the different clitic particles (e.g., -kin, -hAn) that a Finnish speaker may attach to the imperative finite verb (the main verb of the sentence). One such particle is the clitic -pA (realized as -pa or -pä, subject to vowel harmony). According to earlier literature, -pA is used by speakers high in social hierarchy to mark their directives as unproblematic (Hakulinen et al., 2004, p. 800, 1580–1581). Finnish imperatives may also occur with the clitic particle -s, which has been suggested to lend the utterance a flavor of plea and immediacy (“Odotas nyt vähän kun katson” [Wait a minute and let me look] [Hakulinen et al., 2004, p. 803]). Yet another crucial particle in this regard is the compound particle -pAs, which is a combination of both -pA and -s (Hakulinen et al., 2004, p. 801). As we will demonstrate in this paper, in the context of Finnish music instrument instruction, -pA and -pAs are used in distinct ways, serving different instructional goals.

In addition, the directive uses of second-person declaratives (e.g., laitat sen sinne “you put it there”) and interrogatives (e.g., laitatko sen sinne “do/will you put it there”) are highly conventionalized in Finnish.² In contrast to “genuine” declaratives and interrogatives, the directive use of these forms has been associated with the absence of any overt subject pronoun referring to the actor (Yli-Vakkuri, 1986, pp. 155–157). In this way, they are reminiscent of imperatives and, indeed, previous studies have shown that, similarly to imperatives, both second-person declaratives and interrogatives are often used in unproblematic and routine-like directives (on declarative directives, see, e.g., Sorjonen [2001]; on interrogative directives, see, e.g., Lappalainen [2008]; Rouhikoski [2015]).³ However, as we will show below, from the point of view of instruction, these forms have distinct functions that distinguish them not only from each other but also from the imperatives.

Data and Methods

The data for our study come from 10 video-recorded 30–40-min-long instrument lessons (violin, piano, guitar, and ukulele) from different parts of Southern Finland. The interactions are dyadic between the teacher and student, with four different teachers and seven different students from 5 to 14 years of age. All the participants took part in the study on a voluntary basis and parental consent was gained for the children’s participation.⁴ The findings are reported anonymously as regards to names and geographical locations. However, when the use of selected pictures is regarded important for the exemplification of the results, as is the case in the present analysis in connection to Extracts 1–3, a permission for the publication was separately sought from the participants and, in the case of children, their guardians.

The interactional data are analyzed with CA (Clift, 2016; Heritage, 1984; Psathas, 1995; Schegloff, 2007; Sidnell, 2010; Sidnell & Stivers, 2013). In CA, turns at talk are not primarily treated as expressions of thoughts and ideas, but as resources of social action. Conversation analysis, then, is about studying the ways in which social action is organized into sequences. In institutional contexts, CA has been used to unravel the ways in which talk is specialized to accomplish the institutional tasks at hand and thus to describe how social institutions, such as musical instrument instruction examined here, are “talked into being” (Heritage, 1984, p. 290; see also Drew & Heritage, 1992). This paper provides details on how this very process proceeds in the educational context of music instrument instruction.

Although our research question is focused on the linguistic design of spoken utterances, our analytic approach is wider. In line with recent developments in the study of embodied interaction and multimodality (see, e.g., Mondada, 2016; Streeck, Goodwin, & LeBaron, 2011), our point of departure is that participants in interaction use spoken utterances, embodied behavior and material artefacts in concert with each other in order to coordinate their joint activities and to reach mutual understanding of what they are up to at each moment of interaction. In the context of musical instrument instruction, this means that the teachers’ talk, bodily conduct, and the material environment (including the musical instruments, note stands, etc.) contextualize one another, providing public resources for the participants to design their actions so as to be mutually recognizable and intelligible.

The video-recorded interactional data were transcribed according to the CA transcription conventions (see Appendix A). The three dimensions of conduct discussed in this paper emerged in our inductive, data-driven analysis of the interactional data. To report our findings we selected data extracts that illustrate the typical patterns found across our data. As typical for the qualitative CA studies, instead of developing quantitative evidence for the paper’s claims, the paper provides a detailed description of each data extract that demonstrates some systematicity in the participants’ orientations across cases.

The transcripts of the data extracts have three lines. The first line provides the original Finnish interaction, transcribed according to the CA conventions (Schegloff, 2007, pp. 265–270; see Appendix A). The second line is a word-for-word English translation of the original Finnish speech, which also includes grammatical glosses (for the meaning of the glossing abbreviations, see Appendix B). The third, boldfaced line presents an idiomatic English translation. The word-for-word gloss line has been left out in those cases where it does not provide any information additional to the idiomatic translation line.

Results

In this section, we will present our results. We will first describe our sample and then seek to account for the variance in the teachers’ use of six different second-person directive forms.

Sample Description

The distribution of the different student-directed second-person directive forms used by the musical instrument teachers in our data can be seen in Table 1. The six second-person directive forms that this study focuses on are indicated in black. In other words, the bracketed directives in gray are not taken into account in the analysis.

Table 1. The distribution of the teachers' second-person directive forms by in our data set.

	Form	Example	N
Imperatives	Bare	Laita [put]	178
	With cliticised -pA	Laitapa [put]	107
	With cliticised -pAs	Laitapas [put]	78
	With cliticised -s	Laitas [put]	68
Declaratives	Non-modal	Laitat [you put]	109
	(Modal)	Voit laittaa [you may put]	(119)
Interrogatives	Non-modal	Laitatko [do/will you put]	64
	(Modal)	Voisitko laittaa [could you put]	(31)
Total			604 (754)

Accounting for the Variance in the Directive Form Selection

Next, we will account for the variance in the use of the six above-described second-person directive forms. We suggest that the teachers’ choices in this regard are warranted by three dimensions of the participants’ conduct. The first dimension has to do with the location of the directive within the participants’ wider activity structure. The second concerns the degree of the student’s cooperation at the moment of the directive. The third dimension is about the type of priority of the action that is being called for. In the following, we will discuss these dimensions one by one.

Location of the Directive Within the Participants’ Current Activity

As many other educational settings, musical instrument instruction also has a more-or-less predictable format. A typical music instrument lesson consists of a series of actions, such as tuning the instrument, warming up, playing the “homework” pieces, working on specific segments of these pieces, and introducing new material to practice at home. As CA research has stressed, the structural organization of social action allows people to anticipate its unfolding in order to coordinate their contributions to it, and the linguistic forms of utterances often systematically reflect that organization (see, e.g., Raevaara, 2017).

Social action is reflexive, which means that actions are constitutive parts of those activities in which they are embedded (Heritage, 1984). Thus, an utterance with a particular linguistic form not only reflects its location within the activity, but also constructs and shapes that very activity. Such activity management has been described, for example, with reference to participants designing their utterances as a continuation of the current activity or as a proposal for a new one (e.g., Heritage & Sorjonen, 1994; Stivers & Sidnell, 2016). In our data, we also found the teachers’ use of second-person directive forms to be sensitive to such considerations.

Here, the teachers’ use of imperatives with -pAs are a case in point. These turns are most frequently used during transitions from one activity to the next. While such transitions may often involve great changes in the participants’ behaviors, such as moving from preparatory stretching exercises to playing with the instrument, such transitions can also be subtler. Extract 1, drawn from a ukulele lesson of a 14-year-old student, is an example of a relatively subtle shift within the wider activity framework of the lesson. Previously, the student has played a piece that he has practiced at home during the past week. Thereafter, the teacher has provided feedback to the student regarding his playing. The extract starts at the point at which the teacher identifies a specific problem in the student’s playing: his failure to put the second finger on the right place on the third string (lines 1–3).

Extract 1 (KT1 04:51)

01 T: se on vähän hakusessa (.) että löydät sen kakkosen

it be a.bit lost   PRT find-2 it-GEN two-GEN

it is a bit lost (.) so that you’ll find the second (finger)

02niin ↑tota? (.) °niin katsotaan se kolmannella kielellä

PRT PRT   PRT see-PASS it third-ADE string-ADE

so erm (.) so let’s check where it is on the third string

03se olis   tolla Frame 1 kielellä kakkonen°

it be-COND that-ADE   string-ADE two

it would be on that string the second [finger]

04soitapas   se muutaman Frame 2 kerran  niin sormi

play-IMP-CLI it a.few-GEN   time-GEN PRT finger

play it a few times so that [your] finger

05muistaa missä se °ääni° on

remember where it tone be

remembers where that tone is

During his explanation of the student’s problem (lines 1–3), the teacher is leaning towards the student, using a pen to point to a specific place on the neck of the ukulele (Figure 1). After having brought his explanation to a close, the teacher issues an imperative with the clitic particle -pAs (“soitapas” [play] line 4). In so doing, he initiates a shift in the participants’ lesson activity: one from the teacher feedback to the student playing. The verbally launched transition also has a correlate in the teacher’s embodied conduct: he performs prominent hand gestures and leans back, away from the student (Figure 2).

Figure 1. Extract 1: Frame 1.

Figure 2. Extract 1: Frame 2.

Extract 1 thus illustrates two important features that the teacher imperatives with the clitic -pAs have in our data: (1) they occur during transitions from one activity (or activity segment) to a next and (2) they are accompanied by visible changes in the bodily orientation of the teacher toward the student – something that underlines the verbally launched activity shift.

In our data, the teachers’ use of imperatives with -s appear to have an interactional function pretty similar to that of the -pAs-imperatives. However, we observed a tendency according to which the -s-imperatives are used to initiate activity segments that are relatively small – segments that could be seen to as “sub-activities” (e.g., “Try [it] first with the left hand”). Also the visible changes in the teacher-student bodily orientation that characterize the use of -pAs-imperatives are largely absent from the -s-imperatives.

Another directive form in our data that has a specific relation to the temporal structure of the participants’ joint activities is the second-person declarative. In other contexts, such as in workplace meetings, second-person declaratives can sometimes be regarded as interactionally problematic in that they assume the recipient’s compliance without considering the recipient’s willingness or capability to comply; they simply state what is going to happen, even if it is the recipient, not the speaker, who is supposed to carry out the action in question (see e.g., Stevanovic, 2011). However, in the instructional setting of our data, declarative directives do not seem interactionally problematic, as they target particular actions that are directly related to the learning aims. Typically, they occur in the context of already ongoing activities, where the directive is not only about the teacher telling the student what to do, but about him or her instructing the student how to best accomplish the task. This is the case, for example, in Extract 2, where a 5-year-old violin student is engaged in a finger exercise.

Extract 2 (VT2 12:38)

01 T: yks kaks (.) koitat pitää  muut pyöreinä  alhaalla (.) ja

one two   try-2  keep-INF other round-PL-ESS beneath   and

one two (.) you try to keep the other fingers round beneath (.) and

02no{stat vaan t- yhtä  sormee  ja sitäki   ↑pyöreenä

raise-2 PRT  one-PAR finger-PAR and it-PAR-CLI round-ESS

you raise just one finger and also that as round

{Frame 1

03ei suorana  vaan

NEG straight-ESS PRT

not straight but

04 (0.6)

05 T: näin.

like.this

like this.

The teacher issues a second-person declarative directive (“you try to keep the other fingers round beneath and you raise just one finger,” lines 1–2) in the middle of an ongoing finger exercise – at a moment at which the student is looking keenly at her fingers, demonstrably trying to carry out the exercise correctly. The directive is accompanied by the teacher moving the student’s finger several times back and forth (see Figure 3).

Figure 3. Extract 2: Frame 1.

Extract 2 exemplifies the common features found in the use of second-person declarative directives (statements of what the student does) in our instructional data. First, their use is warranted by the student being already engaged in the activity at hand – a local activity environment where how-to instructions are particularly frequent and well received. Yet, as soon as a student’s engagement is wanting (e.g., lack of gaze toward the teacher or a relevant object), the teachers in our data are likely to use other directives to tell the student what to do. Accordingly, a declarative directive is never used as the first directive within a larger activity segment but, usually, the student’s engagement had been previously established, for example, by using a pAs-imperative (see Extract 1). In addition, the declarative directives are typical in the context of repetitive cyclical actions (e.g., raising and lowering fingers). In such contexts, the second-person declarative directives are often temporarily related with the participants’ embodied conduct so that it is left open whether they are about backward-looking correction or forward-looking instruction. Instead, even over a very short amount of time, these utterances may move flexibly between describing and prescribing behavior. This has important corollaries for the nature of the actions typically targeted by declarative directives – something we will return to further below.

Finally, our data includes bare imperatives – that is, imperatives without clitic particles. Previous studies (e.g., Keisanen & Rauniomaa, 2012; Lappalainen, 2008; Mondada, 2014c; Rossi, 2012; Sorjonen, 2001; Wootton, 2005) have shown that directives of this type are frequently used in interactional environments where the amount of effort that is being called for may be very little and the compliance may thus be regarded as self-evident (“Look at this”) or where the nominated action may offer a solution to a problem that the recipient has made publicly available through his previous actions (“I’m so hungry” – “Take a cookie”). In many instances, similarly to the second-person declarative directives described above, the nominated action has been shown to be part of a larger action or activity that the recipient has already committed to. This is also clearly systematically the case in our data, where the bare imperatives – unlike their -pAs-counterparts – typically occur in the middle of ongoing action. In this sense, they are similar to the declarative directives discussed above. However, the use of the bare imperatives differs from that of the declarative directives in that they are typically used in an anticipatory way, to pre-empt a possible problem in what the recipient is just about to do.

Extract 3 is from a piano lesson of a 6-year-old boy. Previously, he has played the piece Playmates from the Suzuki piano school, which he has practiced for the day’s lesson. Thereafter, the teacher has given feedback on his body posture and advised him to correct the position of his hands. Just before the beginning of the extract, the teacher has urged the student to try to play the piece once more – ”first with the right hand,” as shown in line 1. As indicated by the musical notes included in the transcript, the student starts to play in the middle of the teacher’s word oikeella “with the right” (line 1).

Extract 3 (PT1 0:11)

01 T: ensin oikeella just putoo kärki edellä hyvä, (0.8)

first right-ADE PRT fall tip ahead good

first with the right (hand) falls with the tip ahead good, (0.8)

02ja toi on rento ja

and that be relaxed and

and that is relaxed and ((touches the student’s right shoulder))

03suorana   siellä on puunrunko, (0.5) hyvä

straight-ESS there be tree-stem    good

straight there is a tree stem, (0.5) good

04sit ku Frame 1 tulee pomppu (.) niin jousta Frame 2

PRT PRT  come  jump   PRT be.elastic-IMP

then when a jump comes (.) so be elastic

The student’s playing of the first two bars of the piece is accompanied by the teacher commenting on the ways the student’s fingers touch the piano keys (line 1), advising him to relieve a tension in his right shoulder (line 2), and guiding him to correct his overall posture (line 3). In her subsequent utterance (“then when a jump comes,” line 4), however, she makes a shift from backward-looking correction to forward-looking instruction: she implies that the student should next change his way of playing without yet verbally specifying what that change entails. Nonetheless, the position of her right hand (Figure 4) – something that the student may perhaps see through his peripheral vision – allows one to anticipate that the matter may have to do with the elasticity of the wrist during the playing of the upcoming staccato notes (notes of shortened duration). And indeed, the teacher later does complete her sentence by saying niin jousta “so be elastic.” This happens only after the student has first played the staccato note with an observably stiff wrist. However, given that the imperative occurs just before the student plays the second staccato note, the student has the possibility to respond immediately: even if his second staccato note does not exhibit any wrist movement analogous to that of the teacher, he nonetheless lets his hand “jump” somewhat higher than before (Figure 5).

Figure 4. Extract 3: Frame 1.

Figure 5. Extract 3: Frame 2.

Was the teacher’s imperative directive responsive to a problem in the student’s playing? Yes, in the sense that the teacher produced the imperative verb of her sentence only after an observable problem in the student’s conduct (playing a staccato with a stiff wrist). Notably, however, the matter that the teacher’s imperative directive was produced in continuation of a sentence that was already launched before the student played his first staccato note, it was framed as being part of an anticipatory instructional action by the teacher (“then when a jump comes so be elastic,” line 4). It is this pre-emptive function in the use of the imperative directives that constitutes a general pattern in our data. The imperative directives treat the student’s previous conduct as relatively unproblematic. Even if there would de facto be a problem in the student’s prior conduct, the imperative directives make “no big deal” out of the problem. Furthermore, to highlight their “neutrally instructive” nature, the imperative directives are usually delivered exactly at the moment when the recipient’s compliance becomes critical.

In sum, Extracts 1–3 demonstrated the common patterns in the Finnish music teachers’ use of second-person directive forms with reference to the location of the directive within the participants’ wider activity. We argue that these choices of directive form – particularly when used consistently over time – not only contribute to the students’ understanding of what their teachers at a given moment are up to, but also help the students to get a better overall grasp of the series of activities that constitute an instrument lesson.

Degree of Student Cooperation

Achieving order and maintaining control of lessons is an important element of pedagogical practice and teachers’ directives often serve that purpose (Macbeth, 1991). Still, situations where the student does not do what is expected of him or her are not uncommon; the institutional role of the teacher does not guarantee his or her right to determine future action, but the institutional hierarchies and their local entitlements must be managed at the level of turn-by-turn unfolding of interaction (Kent, 2012; Stevanovic & Peräkylä, 2012). It is here that the teacher’s choices between different directive forms become crucial. The second dimension of conduct, which, we argue, informs the teacher’s directive form selection, is the degree of student cooperation at the moment of the directive. As suggested in the previous section, the warrant for the teachers’ use of second-person declarative directives (see Extract 2) and bare imperatives is rooted in the students being already committed to the ongoing joint action or activity. But what happens when the student’s cooperation is lacking?

Our data suggest that one resource for the teacher to point to a problem in the student’s momentary conduct is the use of interrogative directives – that is, directives in the form of a question. Unlike the declarative directives, the interrogative directives highlight the contingency of the student’s compliance and problematize it. The use of an interrogative directive is a way for the teacher to display that the student’s compliance is due – immediately.

Extract 4 is from a violin lesson with the same 5-year-old student as in Extract 2. At the beginning of the extract, the teacher seems to assume the student’s readiness for a new activity and launches one (lauletaanpas tästä “let’s sing from here,” line 1).⁵ The student starts to sing along with the teacher (line 3), but just before the extract she has run away from the camera scope to jump on a nearby sofa. It is pretty obvious that she is not concentrating on the task: she starts to sing too late and sings during a rest (see lines 2 and 3). The teacher thus interrupts the singing task and asks the student to come and look at the notes (line 4) and to sit (line 6).

Extract 4 (VT1 5:38)

01 T: lauletaanpas   tästä.   se   menee  näin. (.)

sing-IMP-PL1-CLI DEM-SG3-ELA DEM-SG3 go-SG3 like.this

let’s sing from here it goes like this

04 T:  tuuksä   kattoo   nuo°tista°

come-SG2-Q + SG2 look-INF-ILL note-ELA

do/will you come to look at the notes

05 (.)

06 sä voit vaikka ↑istua tässä samalla, hh

SG2 may-SG2 PRT   sit-INF here same-ADE

you could for example sit here at the same time

07 N: hmm,

hmm,

The teacher’s directive “tuuksä kattoo nuotista” [do/will you come to look at the notes] (line 4) is a question. More specifically, it has the form of a second-person interrogative directive. As usual in the context of such directives in our data, it is apparent that the student is not quite doing what she is supposed to do at the moment at which the directive is issued. Thus, to the extent that the interrogative directives, as suggested above, problematize something in the recipient’s conduct, in this case, there are obvious grounds for the teacher to do that. This is a systematic pattern in our data; the interrogative directives by the teacher are regularly preceded by sudden and unanticipated failures by the student, which hinder the smooth unfolding of what the participants are supposed to do.

Even if, in our data, the interrogative directives by the teacher are most often followed by the student’s immediate compliance, this is not always the case – and this is not the case in this particular data extract, where the teacher faces a problem of the student only providing a minimal response (line 6) and then, during a silence (line 8), carrying out her situation-inappropriate activities outside of the camera frame. At this point the teacher resorts to another directive form: the imperative with -pA.

Extract 4 (continues)

06sä voit   vaikka ↑istua tässä samalla, hh

SG2 may-SG2 PRT   sit-INF here same-ADE

you could for example sit here at the same time

07 N: hmm,

hmm,

08 (1.5)

09 T:  tuleppa , ((snaps her fingers))

come-IMP-SG2-CLI

come on

10 (1.0) ((The student comes to sit.))

11 T: joo-o?

PRT

okay

The teacher’s imperative “tuleppa” [come on] (line 9) is accompanied by her snapping the fingers in a way that hints that the student’s compliance is expected now. And indeed, what happens next is that the student complies (see lines 10–11). While, in this particular case, it is difficult to assess the extent to which the student’s reaction is responsive to the linguistic form of the -pA-directive, on one hand, and to the finger snapping, on the other, the extract nevertheless informs us about the context in which the -pA-imperatives systematically occur in our data. The -pA-imperatives occur in contexts where the student’s overall behavior has been wanting, given all the previous attempts by the teacher to get the student to actively engage with the task at hand. From this perspective, -pA-imperatives acquire an element of moral reproach: they mark the student’s compliance as a matter that has been relevant for already a while, indicating that the teacher may not be willing to wait any longer.

If the -pA-imperatives are vehicles of moral reproach, it is only to be expected that they are usually not the first directives to be resorted to, but that they are regularly preceded by other directives. One resource for the first attempts to regain control over the lesson’s agenda is offered by the interrogative directives, which, albeit problematizing the recipient’s conduct, do this without conveying moral superiority of the speaker in relation to the recipient. Hence, by displaying sensitivity to the degree of student cooperation in the design of their directives, the teachers give constant feedback to their students with respect to their behavior during the instrument lessons.

Type of Action Priority

Finally, the teachers’ selection between different directive forms is also informed by the type of action that is being called for – that is, the “content” of the directive. More specifically, we will argue that here the question is about two different types of priority of action: while some actions are central to musical instrument instruction as an institution, other actions are essential in that they provide the preconditions for the realization of the institutionally central actions.

To demonstrate the point about the two different types of action priorities, let us reconsider Extract 2, where the teacher issued a declarative directive (“you try to keep the other fingers round beneath and you raise just one finger,” lines 1–2) in the middle of an ongoing finger exercise. Previously, we pointed out declarative directives to be frequent in the context of repetitive cyclical actions, where they move flexibly between describing and prescribing behavior. In line with this, the declarative directives occur systematically in connection with actions that are central to learning to play the given instrument (e.g., how to position fingers, how to hold the bow). In other words, the declarative directives not only instruct what the recipient is expected to do in the “here and now” of the participants’ encounter, but also in the future – indeed, whenever the student will play the instrument he or she is about to learn. In the declarative directives, the immediate and distant futures are bound together.

As pointed out above, the interrogative directives, then again, highlight the contingency of the student’s compliance, while yet conveying that the student’s compliance is due immediately. In Extract 6, the student is just about to start to play a violin piece with a piano accompaniment to be provided by her grandmother. The extract starts by the teacher explaining that the student should start playing at the same time as her grandmother does (line 1), which is followed by sitting down (line 2) and producing a compliance token “okei” [okay] (line 3).

Extract 5 (VT2 6:34)

01 T: lähtee ↑mummin    kanssa yh#tä  ai#°kaa°

start grandma-GEN PRT one-PAR time-PAR

starts at the same time with grandma

02 (.) ((N sits down on a chair.))

03 N: oke:i?

PRT

okay

04 T: seisoksä?

stand-2-Q + SG2

do/will you stand (up)?

05 (2.3) ((N stands up.))

The teacher treats the student’s previous sitting down as problematic: she issues an interrogative directive seisoksä “do/will you stand (up)” to remedy the problem (line 4). In response to the teacher’s directive, the student stands up (line 5) and a bit later starts to play according to the teacher’s previous instruction (not shown in the transcript).

Although we have already discussed the teachers’ use of the second-person interrogative directives in our data (see our discussion on Extract 4), Extract 5 demonstrates additional common features in their usage. First, unlike in the instances of the declarative directives, an orderly feature of the interrogative directives is that they do not target the very actions at the core of the institutional activity of instrument instruction but rather the preconditions for those core activities. Thus, for example, at the beginning of new activities, the teachers’ interrogative directives typically involve attempts to get their students’ attention or even physical presence (e.g., “do/will you come from below the grand piano”). Second, in line with the notion of the second-person interrogative directives securing the preconditions of the participants’ main activates, such directives regularly target actions that may be accomplished only once – something with reference to which the declarative directives were quite different. While the fingers and the bow must be moved in a particular way as long as one keeps on playing the violin, it is enough to come from below the grand piano only once. Therefore, by choosing to issue a directive in the form of the interrogative, the teachers may even reflexively orient to a wish that certain actions (e.g., hiding under the grand piano) will not become a routine part of every instrument lesson.

In sum, even if a young music student may have no idea of what instrument instruction is about and he or she may sometimes challenge the whole activity framework through his or her behavior, the teachers have their pedagogically adequate ways to continuously talk the institution of music instrument instruction “into being” (Heritage, 1984, p. 290). As suggested in this section, the selection between the declarative and interrogative directives constitute one way of doing this.

Discussion

Identifying precisely what music instrument teachers may do to elicit desired changes in their students’ musical knowledge and skill is a long-lasting challenge of music education research. In this paper, we have contributed to a deeper understanding of this issue by considering Finnish-speaking instrument music teachers’ directives in relation to three dimensions of conduct that emerged in our inductive, data-driven analysis: (1) location of the directive within the participants’ wider activity structure, (2) degree of the student’s cooperation at the given moment, and (3) the institutional priority of action that is being called for. So what do these dimensions tell us about instrument instruction as an institution and about music learning as such?

The significance of the first dimension – location of the directive within the participants’ wider activity structure – is evident. It is certainly a different thing to initiate something entirely new than to exert modifications to an activity that is already going on. Systematic and regular differences in how different types of directives are positioned in the ongoing activity presents itself as a mechanism by which students gradually learn to grasp how instructional activities normatively proceed. Arguably, a thorough understanding of how instruction is sequentially organized, segmented, and parsed is elementary to how the teacher’s directives and other instructional utterances become followable in the first place. Such knowledge is also a crucial aspect of socialization into the community of music learners and makers (cf. Gabor, 2009) in that it allows one to participate competently in music instructional activities (cf. Lave & Wegner, 1991).

The second dimension – the degree of the student’s cooperation at the moment of the directive – is essentially based on the first one. When the students gradually gain knowledge about the normative structure of music instruction, they also learn to understand their own activities in the framework of what is expected and will be rewarded and what kinds of mistakes may cause sanctions. Duke and Simmons (2006) have suggested that one aspect of successful music teaching is that it proceeds “at an intense, rapid pace” and that teachers state their feedback and directives “succinctly and straightforwardly” (p. 13). To enable this, the student needs to cooperate with the teacher uninterruptedly, which necessitates detailed knowledge on what such cooperation consist of. Our analysis suggests that this happens by the teachers at each moment tailoring their directives with respect to the degree of student cooperation, thus continually giving their students feedback on how they are doing with respect to the degree of cooperation and, when needed, encouraging this.

The third dimension of conduct that warrants the usage of different directive forms has to do with the type of action that at each moment is being called for. Our data suggest that during music instrument instruction, the design of teacher directives is also sensitive to whether the nominated actions are essential for instrument learning as such or whether they target the preconditions of instructional activity. Previous studies have shown how music teachers “select lesson targets” (Duke & Simmons, 2006, p. 12), or establish “learnables” (Reed & Szczepek Reed, 2014) or “what to learn” (Nishizaka, 2006) – something essential for carrying out a successful lesson where the student concretely learns something. Our analysis showed that the teachers’ choices between different directive forms define the essentials in getting onwards in the learning process and are thus intertwined with the students’ learning trajectories. Where a beginner may try to learn everything at once without being able to separate the essential from the unessential, with increasing competence, the students learn to target their learning into those areas that actually take their skills into more advanced levels. As our study suggests, the teachers’ choices between different directive forms provide the students with an implicit online analysis of such priority hierarchies.

Previous literature has highlighted the role of corporeal presence in influencing what, how, and why children learn (Evans et al., 2009). Our data examples illustrated how this may happen in the context of musical instrument instruction. For example, there were situations where the teacher leant back, increasing his distance from the student, or where the teacher snapped her fingers to prompt immediate student compliance. As we have shown in our analysis, such embodied behaviors were interconnected with the teachers’ use of directives in quite specific ways (e.g., the use of -pAs-imperatives being characterized by visible changes in the teacher-student bodily orientation). These types of connections should be examined further to increase understanding of the role of teachers’ corporeal actions during instructional activities.

Our findings also illustrate that experienced teachers can be very skillful in fine-tuning their directives into respectful, yet effective, expressions. In pedagogically challenging situations, the teachers may seemingly not pay attention to the students’ previous pitfalls, but yet utilize the exact knowledge about these shortcomings in the design of their directives to the students. For instance, when our young learner was hiding under the grand piano, instead of telling her off, the teacher kindly asked her whether she would come and see the notes together with the teacher, while the “out-of-ordinariness” of the child’s behavior was conveyed only implicitly. Arguably, the teachers’ use of notably constructive expressions is more likely to encourage than discourage the child towards progress in his or her instrument learning.

All in all, our study has shed light on a multitude of ways in which the subtle verbal nuances in the teacher’s acts of telling the student to do something may play a role in socializing the student into specific forms of reasoning, perceiving, and acting as a competent member of a group. Given that these nuances are seldom a focus of the teacher’s conscious attention, but something that he or she has learnt to use instinctively and implicitly, CA can add a whole new level of precision to the understanding of how such socialization works in practice.

Disclosure Statement

No potential conflict of interest was reported by the authors.

ORCID

Melisa Stevanovic http://orcid.org/0000-0002-0429-1672

Additional information

Funding

This research was financially supported by the Academy of Finland [grant numbers 274735, 284595, and 298052].

Notes

1 Many studies analysing the teachers’ interactional conduct in different educational environments prefer the term instruction (see, e.g., De Stefani & Gazin, 2014; Lindwall et al., 2015; Mondada, 2014a). Also in our data, most of the teachers’ educational interventions are turns through which they tell the students to do some action and also offer information on how to carry out the action and could therefore be characterized as instructions. This does, however, not apply to all of them, which is why we will use the general term directive.

2 In this paper, we will consider only those declarative and interrogative directive forms that have the finite verb in the indicative mood and lack modal auxiliaries. While both modal declarative directive forms (e.g., “voit laittaa sen sinne” [you may put it there]) and modal interrogative directive forms (e.g., “voisitko laittaa sen sinne” [could you put it there]) are also frequent in the context of music instrument instruction, the dynamics in their use differs from that of the indicative forms to the extent that we have excluded them from this study.

3 Unlike imperatives, the declarative and interrogative forms have a second-person inflectional ending in the finite verb (e.g., laita-t “put-SG2” “you put”), which, in the case of an interrogative, is also followed by the question clitic -kO (e.g., laita-t-ko “put-SG2-Q” “do/will you put”).

4 The study has been committed into following the strictest ethical guidelines, including those set in national Responsible conduct of research and procedures for handling allegations of misconduct in Finland (Finnish Advisory Board on Research Integrity, 2012) and international The European Code of Conduct for Research Integrity (All European Academies ALLEA, 2017) contexts. The participation of young children has meant additional sensitivity for the planning and carrying out of the study.

5 Notably, the directive is in the form of a hortative with a clitic particle -pAs attached to it. Similarly to the -pAs-imperatives, the hortative directives with -pAs convey that the targeted action is first within a series of actions, while nevertheless implying that it will be carried out by the teacher and the student together—unlike in the cases of -pAs-imperatives, where the nominated actor is the student only (see Stevanovic, 2017).

Appendices

Appendix A. Conversation analytic transcription conventions.

Table

.	Pitch fall
?	Pitch rise
,	Level pitch
↑↓	Marked pitch movement
underlining	Emphasis
-	Truncation
[ ]	Overlap
=	Latching of turns
(0.5)	Pause (length in tenths of a second)
(.)	Micropause
:	Lengthening of a sound
#	Creaky voice quality
°	Whisper
<word>	Slow speech rate
>word<	Fast speech rate
↑_____↑	Beginning, duration, and end of playing activity (arrows pointing upwards to the precise point in talk or silence where playing begins or ends)

Appendix B. Glossing abbreviations.

Table

PL	Plural
2	Second person
SG2	Second- person singular pronoun
GEN	Genetive
PAR	Partitive
ESS	Essive
ADE	Adessive
INF	Infinitive
COND	Conditional
CLI	Particle clitic
Q	Question clitic
PASS	Passive
PST	Past tense

Singular, third person, nominative, active, and present tense are forms that have been considered unmarked. These grammatical properties of words have not been indicated separately.