Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Is ‘seeing’ believing? What comprises the focus of seeing, how is it seen and who decides what is to be privileged in doing so? Such is the dilemma facing all observational investigations since what can be ‘seen’ is always impaired or enhanced by what each person brings to their gaze—be it frameworks or ideologies that limit or create potential. How much more challenging is such seeing when the subject of our gaze is an infant or toddler who speaks a distinct corporeal language that has long been forgotten by the adult, and who draws from a sociocultural domain that is only partially glimpsed by the early childhood teacher or researcher? In this chapter I expand on the idea of ‘seeing’ as a dialogic endeavour—thus calling for an exploration of voice that goes beyond singular monologic parameters, into the polyphonic terrain of speculation, uncertainty and reflexivity. Taking this approach, I argue that there is potential to re-vision the very young child as a competent yet vulnerable communicator of and with many voices, one who is capable of conveying complex meaning through genres that strategically orient them towards or away from intersubjective harmony.

The genesis for this chapter draws from my doctoral study, which sought to interrogate how adults noticed, recognised and responded to very young children using dialogic methodology (White, 2009a). In this study the central tenets of dialogism based on the works of Russian philosopher Mikhail Bakhtin (Bakhtin, 1968, 1981, 1984, 1986, 1990; Medvedev & Bakhtin, 1978; Voloshinov, 1973), when operationalised, offered a legitimate means of ‘seeing’ the ‘voices’ of infants and toddlers. Drawing on Goethe’s notion of visibility—‘the seeing eye’ (Bakhtin, 1986, p. 27)—Bakhtin develops the principle that the only ethical role of an author in evaluative activity is to ‘give way to the work of the eye that contemplates the need for performance and creativity in a particular place and at a particular time’ (p. 38). This concept is further developed in his earlier works which highlight the essential surplus of seeing that an author offers her subject and which compels them towards an ethical evaluation of an act, as well as the surplus the subject retains in any analysis (Bakhtin, 1990). In dialogic research, therefore, this visual surplus is central to investigation because it enables a focus on the interpreted experience of ‘other’ rather than consummated truth claims by the lone researcher:

For one cannot ever really see one’s own exterior and comprehend it as a whole, and no mirrors or photographs can help; our real exterior can be seen and understood only by other people, because they are located outside of us in space and because they are others. (Bakhtin, 1986, p. 7)

‘Seeing’, and its authorial surplus, then, refers to interpretative activity that considers the visual act that can be seen and its bestowed or imbued meaning by others as well as the toddler’s perceived intention. Both, it is argued, are necessary if adults are to enter into such interpretative domains with very young children. Such an approach does not dismiss the significance of listening but embraces the eye and its orientation as central to meaning making. Seen in this light, apprehension is only possible through participative thought and action—both central tenets of dialogism.

In a dialogic approach ‘voice’ is located within the broader domain of utterance, since attention is paid to the interpretative nature of language for participants and its potential for meaning in a particular context. The efforts involved in interpreting such acts, as well as the acts themselves, do not escape the attention of the dialogic researcher. Indeed, they invite the researcher to explore ‘voice’ as an answerable act that is interanimated by the voices of others who may or may not be present at the time. As utterance, ‘voice’ can include any sound, gesture, movement or word that has the potential to be recognized by other in social exchange. In this chapter, ‘voice’ is therefore presented as a plural concept that denotes both a consideration of the detailed forms of language, their strategic orientation on the part of the child and the ideologies brought to the experience by the adult.

Such an expanded notion of ‘voice’—beyond speech to visual acts of social engagement with others—creates opportunities for the researcher to glimpse the complex personality of an individual within the world. For Bakhtin, personality is not an isolated construct and refers to both the spirit (bestowed) and soul (performed acts) of a person (Sullivan, 2007). I suggest that in its current pedagogical location, interpretations of ‘voice’ by teachers or researchers that focus exclusively on only spirit (e.g., dispositions, theories of mind) or soul (e.g., activity, skill, verbal speech) in an exclusive manner can promote, ignore or even silence the very young child—rendering them voiceless objects in both pedagogy and research activity. In this chapter I suggest that dialogism provides an antidote to such a dilemma since both are central to investigation and research takes place between subjects and the multiple voices each bring to interpretation.

Dialogism is now a recognised research methodology in fields of psychology, linguistics and education. While dialogism is now also considered to be a legitimate pedagogical orientation (Matusov, 2009; White & Peters, in press), its application in early childhood education is still tentative. As Bakhtin saw the world in terms of socially situated semantic orientations rather than an ordered system of monologic thought, dialogic methodology has typically been employed to explore language as a social entity. Dialogism begins and ends with the everyday exchange and is embedded in reality, that is, what can be literally ‘seen’ with the eye as artistic or aesthetic contemplation. As a result, dialogism rejects abstract scientific approaches—instead highlighting the detail of language interaction and its evaluative meaning in social exchange (Hicks, 2002).

Dialogism therefore assumes that all language (verbal and non-verbal, written or spoken) is social. Indeed, for Bakhtin, language is meaning. Rather than delivering random linguistic messages, Bakhtin argues that one is always deliberately provoking a response from another out of a genuine desire to communicate even when they are not heard by other since, for Bakhtin, there is always an assumption of an invisible ‘thou’. Seen in this light, the purpose of communication is not to reach an end point, or necessarily to reach a point of agreement, but ultimately to negotiate or even reject what is being offered in a dialogic interplay of form-shaping ideology.Footnote 1 Hence the act is constantly being shaped by the receiver as well as the speaker in a dialogic ‘dance’ of meaning making because, as Holquist (1998) points out, there can never be a single unitary plane that is interpreted identically. Language is always situated in a specific time and space.

Based on the central tenets of dialogism, I devised an approach that allowed me to ‘see’ dialogically by employing video and interview methods to approach everyday acts. Examining the interpretations of a teacher, parents and myself in scrutinising language acts of toddlers, while ensuring that the toddlers themselves contributed authentically to the research process, provided a means of developing insights into both the toddlers and those who sought to understand them. Through the employment of utterance as a central unit of analysis and genre as a framework for analysis, I was able to ‘see’ more through the eyes of others in dialogue since each of us added different insights that were not previously accessible to us as individuals. Of particular significance, and in response to the challenges in the field for research with toddlers, dialogic methodology enabled toddler voices, in their many and varied genres, to actively contribute to the investigation. As such I make the claim that dialogic methodology enables those who are potentially voiceless to have their complex and multiplicitous voice(s) ‘seen’ in research activity.

In this chapter I describe a study of an early childhood education teacher, 18- to 20-month-old toddler, her parents and myself as researcher using dialogic methodology as a means of interpretation. I provide an overview of the purpose, research questions, innovative methods and their alignment to the philosophy underpinning the study. Presenting brief examples from the point of view of four dialogic partners—toddler, teacher, parent and myself—I explain how we exploited genre as a potential portal to intersubjectivity. The chapter concludes with a consideration of the relevance of this methodology to early childhood education research in dialogic relationship with (instead of for, on or about) toddlers. In doing so, I present an alternative means of entering into the semiotic sphere of the toddler. I argue that the dialogic approach employed offered a means of accessing multiple voices as genres—some that could be seen and/or heard and others that, due to the dialogic intent of either the adult and their ideologies or the toddler and their orientations, could not. ‘Seeing’ is therefore recast as a dialogic quest that offers significant insights into the personality of the toddler that are seldom afforded through dominant research and assessment methodologies in the field.

The Investigative Context

The study was conducted in Aotearoa, New Zealand, where a focus on ‘noticing, recognising and responding’ to young children (Education Review Office, 2007; Ministry of Education, 2004, 2007, 2009a, 2009b) is a particularly significant phenomenon in the contemporary context. In this location, early childhood education teachers are required to engage in assessment for, not of, the learning of children in formal educational settings—birth to five years of age. Such assessment practice is framed around dispositional theory that seeks to promote and profile important habits of mind as determined by the interpretive teacher in consultation with the child’s family (Carr, 2009).

Assessment endeavour had proven to be elusive for some teachers of infants and toddlers (Stuart, 2008) since in order to notice significant dispositions, teachers must be able to ‘see’ the very young child. This task is especially challenging when the semiotic modes of communication are not shared and the child moves between the diverse semiotic spaces of home and early childhood education institution. Since infants and toddlers are the fastest growing population in early childhood education settings within wealthy countries (UNICEF, 2008) and are spending increasingly long periods of time with teachers as opposed to family members, interpreting children’s acts through assessment activity is now viewed as central to their learning and development (Education Review Office, 2007). A primary means of gaining insight currently resides in the notion of ‘voice’ which, in contemporary early childhood education ideology, is considered to be graspable through a listening pedagogy comprising observed (or photographed) activity, dialogue with parents and written narrative (see, for example, Perkins, 2009; Sands & Weston, 2010). As such, ‘voice’ is unproblematically positioned as a deliverable, observable, authored statement about the child’s learning progress, which claims to speak for and with the child (for a further discussion of this concept, see White & Peters, 2011).

The broader phenomenon of ‘seeing’ important aspects of learning or development is also a task for researchers who seek to situate their work within the field of early childhood education. Researchers across multiple domains have employed a variety of observational techniques and methodologies to try to understand ‘other’ for a range of purposes. Dalli et al. (2011) suggest that a majority of studies of under-two-year-olds, internationally, take place between mothers and infants in the home or between infants and toddlers as peers. More recently, observational approaches in research have sought to capture ‘voice’ of very young children by generating data through what has been conceptualized as a mosaic approach (Clark, 2004), and in Reggio Emilia early childhood education pedagogy has been reframed as research activity on this basis (Millikan, 2003). In these locations claims are made about children based on multi-modal forms of communication ranging from photographs, to artwork, to verbal language records—all of which claim to constitute ‘voice’.

While the notion of seeing ‘voice’, in its expanded interpretation, is a legitimate means of accessing and observing the language of children within early childhood education, its application in both teaching and research domains is elusive. Løkken (1999) suggests that a rationale for the prolonged dearth of research about toddlers is associated with the unique challenge they present for adult interpretation. It is highly likely that this challenge is an important reason for the small number of empirical research studies that focus specifically on this age group, in particular under-one-year-olds. The few claims that have typically been made about infants and toddlers in education are based on developmental or psychological epistemologies that emphasise their teleologic journey into adult domains or take a scientific view (Berthelsen, 2010). Such a stance ignores the social, corporeal, and potentially discursive nature of very young children’s interpreted communication enabled by dialogic methodology. As Broström and Hansen (2010) point out, research with very young children in educational contexts should address four key points:

  1. (1)

    the dynamic of the pedagogue-child relation (care, empathy, acknowledgement, etc.);

  2. (2)

    the pedagogue-content relation: how the pedagogue presents the content to the child;

  3. (3)

    the child’s relation to other children; and

  4. (4)

    the pedagogue’s relation to a group of children (p. 97)

Paying attention to an expanded view of voice(s) through dialogic means goes some way towards addressing these points by exploring the point of view of each participant, not least of whom is the teacher, rather than an isolated analysis on the part of the researcher. As such, important insights can be offered to adults in their quest to interpret meaning and gain a deep appreciation of the toddler.

Investigation took place in a high-quality mixed-age Education and Care centre, located in Wellington, New Zealand. In keeping with the suggestions of Markova and Linell (1986), I chose to work with the same participants over different activity contexts with the aim of affording dialogic depth, rather than breadth, of inquiry. Like other dialogic educational researchers, my focus was on an individual within the social context of the group so that I could investigate acts from multiple points of view, rather than an established standpoint. To this end I worked closely with one toddler, Zoe, a European New Zealand female, aged 18 months, at the beginning of the project, who spent 40 hours a week in the centre context and had attended the centre from age 6 months. Her teacher, Alicia, was a fully qualified early childhood education teacher, of Māori descent, who had several years’ experience in the field. Alicia spent in excess of 17 hours in video recorded dialogue with me about what she had ‘seen’ in Zoe’s acts—some during her paid non-contact time and some in the early evening. Zoe’s parents were actively involved in the study and brought their own interpretations, sharing over seven hours of recorded dialogue about their daughter. Together, Zoe, Alicia and her parents contributed much to the interpretative arena in which this study was located. Coupled with my attendance at staff meetings and less formal events at the centre during this four-month period, these dialogues enabled me to enter into what Bakhtin refers to as the heteroglossic domain of inquiry that comprised Zoe’s world.Footnote 2

The researcher’s participation in the study should not be underestimated in dialogic methodology. My role expanded well beyond the hours described above. Over the four-month period of fieldwork, I entered the site almost daily to attend meetings, share transcripts, talk with staff and/or parents about the footage, share findings and generally become part of the early childhood education context. Since my interpretations were to be juxtaposed with those of the other participants, I maintained a constant vigil on my authoritative position as ‘researcher’ and ‘expert’, as opposed to a visitor (or outsider) to the centre and in the life of the toddler. It was at these ‘sites of struggle’ (Frank, 2005, p. 971) that I lingered in my interpretations. I contend that these challenges keenly represent Bakhtin’s entreaty that authorship is a moral activity as well as an evaluative one. My constant attention to this domain was reflected in the research journal notes I took, which explored my relationships with participants (including the other children and adults in the early childhood education setting), my ethical role as researcher, my ultimate research quest and what all of this meant in a dialogic investigation of this nature. Several dilemmas were faced in this regard, not least of which was the constant challenge to avoid speaking on behalf of others—a polyphonic entreaty endorsed by Bakhtin (1984) sometimes described as ‘ventriloquising’, but previously unexplored in research with the very young child. I do not claim to have avoided this inevitable trap altogether, but the moral tenets of dialogism constantly reminded me of the need to allow participants to speak for themselves whenever possible.

Conceptual Tools for Dialogic Investigation

Dialogic methodology, then, allowed me to enter into the interpretative spaces that reveal or conceal what can and cannot be seen by adults when very young children are engaged in language acts. Bakhtin (1986) describes this emphasis as ‘utterance’, a concept I adopted as the unit of analysis for the investigation. Utterance can be described as a reflection of the conditions and goals of communication, that is, its orientations and ideologies, as well as the communication itself. As Bakhtin explains, an utterance is only achieved when it is answerable. I interpreted this to mean that language should be interrogated inside and outside of direct social relationships, by individuals and in dialogue with others. I selected language acts that were therefore noticed by some or all participants (including myself) and which constituted some sense of meaning accordingly. Recognition of meanings expressed as surprise or wonderment were considered especially relevant to interpretation because the act and its perceived meaning could be said to have ‘str[uc]k the same key in each other’s spiritual instrument’ (Cassirer, 1953, p. 160)—a feat that Bakhtin (see also, Voloshinov, 1973) suggests is necessary in order to engage in meaning making.

The point of view of each participant on the language acts of the toddler was therefore central to inquiry. According to Holquist (in Bakhtin, 1981), ‘point of view is always situated. It must first of all be situated in a physical body that occupies time and space, but time and space as embodied in a particular human at a particular time and in a particular place’ (p. xxviii). I accessed each participant’s point of view in a variety of ways since each employed different styles of communication. For example, the teacher preferred to record her ideas prior to discussion, the parents favoured informal dialogue, and the toddlers employed a style of communication closely aligned to eavesdropping—providing ‘language crumbs’ for adults to pick up on a trail of discovery. These contributions, as form, contributed to the overall point of view each participant brought to the interpretative experience.

Utterance provided an important way of accessing such points of view. Central to my quest was to draw on the interpretive presence of the teacher, tasked with noticing, recognising and responding to the very young child. However, I also wanted to find ways of examining the point of view of the young child herself, as central protagonist in her own learning experience, as well as the views of her parents, who know her intimately, and my own researcher interpretations, as an outsider with keen observation skills. As such, I was deeply concerned with ‘voice’ and its manifestation in this investigation.

Genre is described by Bakhtin (1986) as the means by which an individual can orient meaning to another through the selection of relevant form and content in utterance. It was through genre that I was able to access the language crumbs offered by toddlers in their everyday acts and recognise them as genres of social exchange. As ‘the drive belts from the history of society to the history of language’ (Bakhtin, 1986, p. 65), genres represent the strategic orientation of a particular communicative form to convey (or perhaps even frustrate) meaning. Bakhtin (1986) describes genres as a combination of both form (that is the language form in which communication is presented) and content (that is, its interpreted meaning). By approaching language in this way in my research, a finely tuned analysis of the nature of its delivery was promoted—and its composition excavated—for possible clues to meaning. Together, these comprised an analysis of genre, which was achieved through the combination of form (what could be seen) and content (interpretation). For a fuller discussion of these concepts, see White (2009a).

While significant clues are provided in his later work (Bakhtin, 1984, 1986), I struggled to reconcile the interpretive aspect of my quest with Bakhtin’s dual emphasis on utterance as the answerable language act and genre as the strategy through which the act could be answered. A pilot study proved to be invaluable in resolving the issue (White, 2009b), particularly as there were few research precedents for such an approach to the study of toddlers. While I agreed with Matusov (2007), who suggests that the unit(s) of analysis should be selected in relation to its purpose within the wider study and therefore develop out of the context of data generation, I was also aware of the fact that I had brought Bakhtinian views strongly to bear on this research. In doing so, I argue that the Janus-like features of utterance (that is, in focussing on both the language act and its answerability) embody the very essence of dialogism, while genre is the means by which language may be recognised (or not). While holding important clues to meaning, genre was a means to an analytical end, but not an end in itself. Indeed, given the contestable nature of toddler communication and its interpretation, I concluded that recognition would never provide the complete picture, only partial glimpses that, taken together, provide an enhanced appreciation rather than collective ‘truth’.

As already explained, I was keen to interrogate the nature of interpretation as well as any claims made by each participant through observation and dialogue. My ultimate quest was to find a way to understand the toddler as a dynamic personality in her own right with much to contribute rather than merely as learner with much to receive from more knowledgeable others. In doing so I concur with Sullivan (2007), who suggests learning is a received act whereas personality represents both soul and spirit and is accessible only through interpretations of observed language acts. Bakhtin (1986) suggests that, despite being an essential feature of authorship, attempts to understand personality are fraught. Though never isolated, neither is personality always collective, since while the individual is part of a culture, they should never be fully consumed by it. His research imperative is therefore to access both the whole and parts of a personality as enacted and interpreted by other.

In accordance with this argument, I determined ways of examining language from both the individual view point, through genre, as well as its interpretations in dialogue with others. Bakhtin’s attention to the polyphonic novel provided inspiration in this regard. As an artistic device, polyphony structures manipulate events in order to portray their complexity. Considered alongside visual surplus, which prioritises ‘the work of the eye that contemplates the need for performance and creativity in a particular place and at a particular time’ (Bakhtin, 1986, p. 38) polyphony offered scope for exploring the visual field of research participants while suspending any monologic claims of certainty about what could or could not be seen or interpreted about the toddler. Taken together, attention to genre, polyphony and visual surplus provided an entry point to the dialogic study of an expanded version of voice and its interpretations. They provided not only a means of ‘seeing’ language acts and their orientations through multiple points of view but also the potential of accessing voice without privileging one interpretation over another. I operationalised these concepts by developing a polyphonic visual ethnographic approach to investigation, as a means of affording visual surplus and responding to Bakhtin’s seeing entreaty, which is described in the section that follows.

Operationalisation of Dialogism

Out of this conceptual understanding and in accordance with my research priorities, I developed two key approaches that supported a dialogic research approach in the early childhood education context with very young children. These I call polyphonic video and re-probing interview. The first, polyphonic video, involved the collation and timed synchronisation of video footage focusing on the toddler in the everyday centre context and taken from the multiple vantage points of a teacher, a toddler, and myself, which were not dissected in any way prior to participant interpretation.Footnote 3 I did this in order to facilitate opportunities for participants (including myself) to engage in detailed, unabridged encounters with toddler language acts in communication with others in the local early childhood education setting context. I wanted to understand what language genres were more likely to be noticed, recognised and responded to, by whom and for what reason and the impact of such recognition or non-recognition on the way the toddler could be seen. Located within the New Zealand assessment context described earlier, this focus was of great significance to the teacher in this study as well as my own research agenda. As such, the emphasis of the study responds to dialogic forms of validity that draw inspiration from the local community in which the investigation is set, and the extent to which the study creates ‘capacity for continual response’ within this space (Sullivan & McCarthy, 2005, p. 633). Generating dialogue between participants was central to the approaches I employed.

Participant interpretations were documented by visual and audio means through a series of re-probing interviews, which drew on the polyphonic footage to try to make sense of the toddler’s language acts as genre. Through this second approach to data generation I was able to exploit Bakhtin’s notion of visual surplus as a potential means of investigating the insights each participant brought to their interpretation (both in terms of what is literally seen on film and what could be noticed as worthy of discussing in interpretation). In other words, I sought to offer a ‘view’ of the toddler’s experience, literally and metaphorically, from the visual field of each participant. In doing so, I considered the additional perspectives each person could bring to the assessment experience in trying to generate meaning. What I did not anticipate, and what featured highly in the data, was the visual surplus the toddler also brought to the authorship experience, and the potential of such authoring on the part of the toddler to contribute to, and challenge, adult perspectives, including my own. This insight was made possible because interviews were videoed, and in doing so, granted access to non-verbal as well as verbal forms of dialogue.

A visual means of capturing Zoe’s language acts was achieved by attaching a small hat-cam to the teacher’s head (on a hat that was attached to a backpack with a video camera inside), another small hat-cam on Zoe’s head (on a head band that was remotely connected to a video camera elsewhere) and a third pan-camera held by the researcher. Placing a camera in the eye line of participants highlighted visual subjectivities in relation to where the head was looking as opposed to where the camera was strategically shot, as is typically the case with visual ethnography in education (see Fig. 4.1). I argue that, coupled with participant interpretation of what could be seen, polyphonic footage is an authentic representation of Bakhtinian authorial surplus. Emphasis is placed on the additional insights that can be gained through visual interpretation from different people since Bakhtin (1990) argues that each person ‘knows and sees more in the direction in which the hero is looking and seeing, but also in the different direction, in a direction which is in principle inaccessible to the hero himself’ (p. 13).

Fig. 4.1
figure 4_1_210735_1_En

Split-screen polyphonic footage shot showing the same view from the child, teacher and researcher camera lens

Full size image

Individual videos were then combined in a time-synchronised split-screen representation that was offered to each participant for analysis. Participants were invited to identify ‘noticed and recognised’ language forms, as genre, based on a set of criteria developed in the pilot and their interpreted metaphoricFootnote 4 meanings from video-provided data that was entered into an analytic software programme available at the time (Webbsoft, 2007). I struggled to involve the toddler in this kind of interpretation (White, 2010), a point I return to in the sections that follow.

Dialogic Analysis

In order to fully exploit utterance as the central unit of analysis, a four-staged approach was taken to data analysis, as follows:

  1. 1.

    Independent participant coding of genres based on one-hour excerpts of polyphonic video footage.

  2. 2.

    Participant dialogue with researcher during videoed re-probing interviews.

  3. 3.

    Researcher enters generated codes into the software programme ‘Snapper’ (Webbsoft, 2007) alongside coded analysis of participant discussion styles.

  4. 4.

    Analysis of genres, discussion styles and associated footage that represents some form of noticing, recognising and/or responding for one or more participant. Initial selections made on this basis.

Of significance was not only what was identified (that is, its form), and by whom, but how it was identified in dialogue (that is, its interpretation), since dialogic methodology foregrounds the nature of dialogue as much as its content. Coding was considered as an appropriate means of analysing dialogic interactions—both in the toddler acts and during their analysis by participants—in order to consider both the acts and their interpretations as a dialogic unit. The various recognised language forms and their descriptions were employed as a means of coding. These included biting, clapping, facial gesture, use of fist, laughter, hitting, hugging, pointing, outstretched hand, sound and a range of other forms that involved the employment of various resources. Of particular note was the high proportion of communication forms that were not verbal and that employed body parts in their delivery. Although pointing has now been recognised as a strategic means of communication for infants and toddlers (Tomasello, Carpenter, & Liszkowski, 2007), many other non-verbal language forms are often overlooked in research activity, considered as mere body functioning and therefore interpreted literally for this age group (McNeill, 2005; Roth, 2001). When seen within a dialogic framework, however, these language forms, as genres, can be viewed as symbolic strategic orientations worthy of communicative consideration.

The interpretations given to each of these acts, as a second set of codes, varied greatly between participants and ranged from demonstrating, explaining and naming, to resisting, mimicking, pleasing and tricking. There were several acts that escaped adult interpretation—remaining curious and ungraspable acts in the interpretive process. These were typically those that took place in moments that were not shared visually by the teacher on camera, and which were characterised by body movements such as running across spaces with arms extended outwards and squealing sounds. I described these acts as free-form genres (White, 2009a) that eluded adult interpretation. Those that the teacher could not explain were described as ‘routine’ because they appeared to be commonplace acts that bore no significance. Still others that both teacher and parent clearly saw but, without my prompting, did not choose to interpret were associated with what I have described as intimacy genres (White, 2009a). These involved the toddler touching or probing the teacher, such as pointing to her breast and saying ‘milk’ or touching her facial mole repeatedly over months. I speculate that my ability to bring such acts to the interpretive domain were due to my outsider’s perspective, a point Bakhtin (1990) highlights is necessary for evaluative activity.

Others, which were repeated over time, could be viewed as important cues to meaning. For example, upturned palms in the presence of an adult came to be seen as a quest for meaning, like saying, ‘I don’t know, do you? ’ On another occasion, we noticed the toddler closely scrutinising a peer crying and receiving cuddles from the teacher, then repeating the same noise and facial expression in order to receive similar attention. Acts such as these were only graspable to me with the benefit of a researcher gaze over an extended period of time and the insights of other adults who knew the toddler well, an intimacy that Bakhtin (1990) suggests is also necessary. His stance is that both insider and outsider perspectives are important, and that one should never subsume the other, lest the subject becomes lost as a personality in his or her own right.

A third set of codes that provided the framework for understanding genres, their orientations and their interpretations was based on the styles of discussion that adults used when engaged in interpretive dialogue with one another. Together with the previous two coding criteria, I was able to examine footage in relation to what was noticed, by whom as well as what was not noticed, and some insights in relation to why this was the case. Codes ranged from acquiescence, agreement, inquiry and embellishment to blocking, disagreement and uncertainty. Drawing from the work of Matusov and Smith (2005, p. 706) I added a further set of criteria as finalise, objectivise and subjectivise—three concepts that were featured in much of the dialogue.

The software provided a means of bringing to bear three different coding sets on the split-screen polyphonic footage. I was able to code what was seen, by whom and the nature of dialogue during the interpretations against specific pieces of footage. Not only did this approach enable detailed and complex analysis, but it also provided a means of visiting and revisiting the footage from multiple standpoints.

What Could Be Seen and by Whom

The interpretive moments generated out of dialogic investigation represent what Bakhtin describes as ‘the fire of aesthetic value’ (1990, p. 91). Analysis revealed that such fire was profoundly kindled by the authorial surplus each participant brought to the interpretive domain. The resulting differences in what could be ‘seen’ drew from the ideologies each brought to the experience. For instance, the parent frequently saw the acts of her child as a reflection of herself or simply did not see the significance of the acts at all; the teacher viewed those same acts in light of her public accountability to the state and family (a stance I argue is influenced by the authoritative discourses at play within the education setting) or chose not to foreground them in dialogue because they revealed aspects of herself; whilst I straddled an outside–inside stance based on my academic responsibilities and professional background (a dual position Bakhtin, 1990, describes as essential to dialogic authorship). My ability to see anything at all was interdependent on the dialogue I shared with the parent, teacher and toddler herself, since I relied on their insights to influence my own.

The toddler, who seemed to become increasingly and astutely aware of her central location in our gaze, provided remarkable insights through the research process. Much to my surprise, these insights occurred as much as in the interpretive aftermath of the videoed acts as in initial viewing of the acts themselves. As I had invited Zoe to participate in interviews that took place between her parents and myself, her presence during these dialogues enabled her to contribute in the interpretative process in subtle ways. I did not always appreciate these contributions at the time of the interview, only afterwards in watching the footage beyond verbal dialogue exchanges and drawing together the threads of each research encounter as a greater whole. It was here I noticed Zoe employing strategies such as eavesdropping, gesturing and offering artefacts as clues to the adult participants during interviews, and re-enacting aspects of our dialogue in subsequent filming sessions. These contributions provided important insights into both the genres employed by the toddler to generate meaning and her strategic orientations towards (or away from) authorship by others. Such strategies were therefore by no means the exclusive domain of the adults and, in my view, represent ‘aesthetic fire’ and its ignition through an expanded appreciation of voice in work with toddlers.

An example of such appreciation is illustrated in a language act that I have called ‘Baby Rock’ (White, 2009b). The genesis of this act lay in an early piece of film entitled ‘Rosters Rule’ in which the toddler’s mother (Lynette) noticed the toddler (Zoe) rocking a doll (White, 2009b). During this earlier sequence, it was my interpretation that the toddler was attempting to attract the attention of her teacher (Alicia) through a variety of mediating strategies. Alicia, at the time, was engaged with setting up the environment for children transitioning from group time, and then in settling the smallest babies on the floor. After several attempts to draw the teacher’s eye towards her, using a range of strategies such as offering toys, moving to different parts of the room or calling out, Zoe selects a doll from the shelf and holds it to her chest in the same rocking motion as her teacher. At this, the teacher responds both visually and verbally, and Zoe engages in a sustained period of nurturing activity with the doll as Alicia cares for the ‘real’ babies alongside her. Their mutual gaze is evident when seen polyphonically through split-screens.

During re-probing interviews Alicia expressed her absolute horror at this episode, describing it as a negative reflection of her teaching practice. This positioning made it difficult for Alicia to ‘see’ anything about Zoe at all since her view was shrouded by disappointment in her own practice, which, in her view, was characterised by chaos. When I suggested to her that perhaps Zoe could be seeking her attention through strategic acts, Alicia changed the subject, turning the dialogue back to herself, saying things like, ‘I wasn’t doing my job’ or ‘I kept thinking “Oh what am I missing, what am I missing?” … I know I should have been out there’ [Teacher Interview 5]. Alicia’s anxiety revealed not only her disappointment in herself, but also the enormous pressure she felt to meet what she perceived as authorial accountabilities.

In an interview based on the same episode noticed by Lynette, however, the dialogue took a different turn. Lynette focussed on links between Zoe’s interest in rocking the baby and her recent experience at home with a friend’s baby. Zoe’s keen interest was described, and Lynette demonstrated the way the baby was fed and rocked in the car seat, including the positioning of a blanket over the car seat when the baby went to sleep. Lynette explained how fascinated Zoe had been with these events. Zoe was present during this interview. When dialogue focussed on the baby and doll, Zoe retrieved two very different dolls from the shelves and passed them to her mother and me. Lynette was given the new doll with blonde hair (Lynette is blonde) and I was given a weathered doll (presumably due to my status as the ‘senior’ member of the group). Together we sang the nursery rhyme ‘rock-a-bye-baby’ and talked about the dolls and their sleeping habits.

Hours after this re-probing interview I re-entered the Education and Care setting to commence a filming episode. Zoe brought the same weathered doll over to me immediately, retrieving it from another child to do so. As I was preparing the cameras Zoe ran around the room calling ‘Yay yay’ and eagerly embraced the cam-hat that she was invited to wear. From my point of view it seemed that as soon as filming began, Zoe adopted a performative stance. Immediately she retrieved another doll from the shelves—this time a dark-skinned doll—and took it outside to Alicia. With the addition of a basket (similar in shape to a baby car seat), she proceeded to rock the baby, sing ‘Rock rock’ and place a blanket over the basket in the exact same way as her mother had described in the earlier interview. In the face of adversity, that is, two other toddlers attempting to take the doll from her, Zoe persisted in her possession of this doll. She consistently returned to her teacher with verbal language prompts about its ‘eyes’, ‘nose’ and ‘bottom’ as she demonstrated cleaning the baby in the same way as she has seen Alicia do with the babies in the centre. During this entire film sequence, both Alicia and Zoe shared the same visual fields, suggesting a high degree of intersubjectivity, which was borne out in the subsequent interview with Alicia.

Re-probing interview dialogue with Alicia following this act was lively. Alicia was keenly attuned to the features of the act that she perceived were based on her own practice, laughing and declaring ‘That’s what we do in the sleep room—exactly that … when she [Zoe] is in the sleep room she is the baby … she is the doll …. And the teachers stroke her back for her to go to sleep’ [Teacher Interview 6]. Through this dialogue, Alicia began to recognise the use of the doll as a means of engaging her teacher, which she described as grabbing her attention—‘hook, line and sinker’Footnote 5 [Teacher Interview 6]. Her interpretations were reinforced by an analysis of the shared visual gaze on the doll, captured across all three camera views. Here it seemed that, if Zoe’s goal was to achieve intersubjectivity, she had been successful through this intentional act, which both flattered and amazed her teacher.

The act also struck a chord with Lynette, who exclaimed, ‘She does it with dolls but she’s also starting to try to do it with adults [laughs]…. It’s quite funny really’ [Parent Interview 5], and she proceeded to describe Zoe’s recent attempts to nurture adults at home by pretending to change their nappy. I suggested to Lynette that Zoe may also be acting on things she had seen in the centre setting. Lynette’s response was resolute: ‘She remembers it herself—you know—with us changing her’ [Parent Interview 5]. Again, there was a keen sense that this act could be attributed to the adult, and to remembered experiences—this time very explicitly claimed from the home and not the centre.

Two weeks later, in a focus group interview between Alicia, me, Lynette, Zoe and Zoe’s father Mark, dolls featured heavily again. Again, it seemed as if Zoe called on the dolls as a means of strategic orientation. As adults were seated around a table, Zoe carefully selected the same set of dark-skinned, blond and weathered dolls for Alicia, Lynette and me. During this interview session, Zoe focussed a great deal on Alicia—touching her, sitting on her knee and repeatedly returning to the dark-skinned doll with blankets and spoons for feeding. In the teacher interview that followed, Alicia speculated: ‘I’m thinking, “Are you making that connection between the doll and me?”. I mean, it’s got black hair, it’s got brown skin like me and I’m thinking … maybe …. It crossed my mind and then I thought “naaaaah”’ [Teacher Interview 6]. Alicia confessed her discomfort with the levels of intimacy offered by Zoe during the interview and her association with the ‘little brown baby’—an act that rekindled negative memories of a similar doll Alicia had been identified with by others when she was a child.Footnote 6 Alicia’s dialogue reveals a subjectivity that Zoe may not have had access to but which appeared to have a strong impact on the way Alicia responded to Zoe’s repeated suggestion that Alicia claim this doll in what I interpret as an intersubjective attempt. Of significance to Alicia was her developing consideration that this 18-month-old toddler might be capable of such strategic intent, and the provocations that Zoe’s acts held for her own subjectivity. This, despite her preconceived professional knowledge base of ‘developmentally appropriate practice’, based on a three-year programme of study preparing her for teaching, that suggested to her that a toddler simply does not have such capabilities.

This example of the doll represents one of many language acts and their interpretation by participants based on subjectivities that each brought to the arena. Through dialogic investigation, Zoe is represented as dynamic, sophisticated and deeply attuned to the subjectivities of the adults around her when adults opened themselves up to ‘seeing’ more. The centre supervisor declared that, as a result of Alicia’s in-depth authorship experience, all staff came to appreciate ‘the Zoe that’s deeper’ too since Alicia was able to share her interpretations with others as a result of these insights. ‘Seen’ through dialogic eyes none of the acts that were filmed could be interpreted as (or by) singular voice but rather the complex interplay of multiple subjectivities, performed for dynamic purposes, and constantly in a state of flux rather than as a finalised feat. While the interpretations are, in the final analysis, the researcher’s, in dialogic research they must be valid for the community in which they are located. As Bakhtin (1990) explains, what can be seen is always influenced by the ‘other’s possible emotional-volitional reaction to my outward manifestation—his possible enthusiasm, love, astonishment, or compassion for me’ (p. 31). In this location, authentic interpretations that generate personality (as opposed to narrow assessments of ‘learning’ or ‘development’) hold great potential for discovery and rediscovery, but they do not make claims about all toddlers in all settings or even for this toddler in a finalised manner. Zoe retains the right to exercise her dialogic loopholeFootnote 7 in this regard.

Revealing Alternatives

Seen through dialogic eyes, Zoe’s manipulative engagement(s) with the dolls provides useful insights into the nature of her acts as genres that are employed for strategic purposes, in the same way as adults use genres to convey meaning. Although a small number of researchers have reported on early childhood education activities or experiences as specific genre that very young children must come to understand if they are to be successful in that environment (see, for example, Cohen & Uhry, 2007; Gillen, 2000), little emphasis has been placed on those genres the toddler brings to bear on social experience as dialogic exchange. Through an exploration of genres there is potential to enter into an experience of toddler ‘voice(s)’ as an intentional and sustained act of meaning making that takes place within, between and beyond the early childhood education and other settings in which the young child resides. In the case of ‘Baby Rock’, for example, the genre of dramatic play through mimicry was employed with great strategic skill.

Entry into the complex arena of dialogic voice, however, means that the researcher has to maintain a dual focus on the acts of the toddler and the interpretations of those around her. Utterance, as the unit of analysis, provided a way of accessing both. Of equal significance to a deeper understanding of very young children are the important insights generated about adult subjectivities and their location in the wider ideologic spaces that surround and profoundly author the toddler’s life. These insights not only consider the toddler as a personality, a peer and member of a culture but also pay attention to the discourses that construct the way the toddler can (or cannot) be seen by others and the various cultures in which she finds herself. Since this study has suggested that the toddler is altered by and simultaneously altering those views, there is ample scope in dialogic research methodology to consider the toddler as authored and authoring—in other words, as both a research subject and collaborator in her own right.

Over recent years there has been a shift to seeking out ways of using video authentically in work with very young children and the subjectivities that inevitably exist when film is shot, or interpreted by another (see, for example, Tobin & Hsueh, 2008). Polyphonic video provides a means of engaging with these subjectivities. Indeed, I came to appreciate Zoe’s joy in participating in her own authorship experience as a claim she was able to make on the research through her interactions. Accessing the subjectivities of children under two years of age requires visual modes of investigation and invites the researcher to work sensitively alongside others in engaging with very young children and their voice(s). Taking a dialogic stance, then, not to support authentic use of footage, is unethical where under-two-year-olds are concerned.

Regardless of any argument I can make to the contrary, there will be those with concerns about the practice of putting a camera on the head of a very young child. Some will argue against the singling out of one small child in a group, or the potential power relationships held by adults in the Education and Care context. Others will focus on the technical aspects, suggesting that there may be some danger in using electronic equipment so close to the body. These and other ethical concerns were part of an organic and collective process I employed in working with all participants to ensure that toddler preferences were upheld over my research priorities. Aside from the obvious issues like safety—an aspect I took very seriously—such consideration required a great deal of risk on my part since I was committed to responding to the cues offered by the toddler. If Zoe (or the other children who acted as pilot participants or back-up) did not wish to wear the hat-cam, filming would cease. If her parents, teacher or I felt that her engagement with the project was negatively impacting on Zoe in any way, the project would have ended. However, this was not the case, from our point of view at least, and may, in part, be an outcome of our careful introduction of the camera to the early childhood centre context. Through such means, Zoe was acutely aware of the impact of the cameras on her experience and her central role in our interpretations. Her responses suggest that it is naïve to think that research can ever be done without altering the event and/or children involved. Indeed, the interpretation that such realisation affords provides further scope for involving very young children more fully in research activity, recognising the performative nature of their lives and providing avenues through which they can contribute to greater insights on the part of adults whom they rely on to understand them better.

Dialogic methodology has only begun to reveal its potential for understanding more about the way we are able to see infants and toddlers. Polyphonic video holds great promise for entering into the interpretive spaces that locate, frame and, at times, limit the way young children can be seen. Taken together, these approaches provide a means of celebrating the unique personality of the individual within and between groups. Furthermore, dialogic methodology supports the view that all language holds meaning and, in doing so, invites researchers and teachers to engage with multiple modes of communication—many of which are visual. As technology improves, we can only imagine the possibilities for future research with this age group when visual research methods are further exploited and shared with the research community as ‘boundary objects allowing us to become an adaptive, distributed, collaborative, expert community’ (Derry et al., 2010).

Concluding Remarks

Dialogic methodology when invoked to ‘see’ the very young child demonstrates that she can never be understood in isolation from others and their interpretive stance. Moreover, any ‘seeing’ is located within a moment of time. I have argued that this is an ethical, pedagogical and empirical imperative for early childhood education practice and research. This finding is supported in ethnographic research, which suggests no image or imaging can be viewed as truth in itself. The subjectivities of those who take the photographs, shoot the film or employ other means of observing the child not only influence but actually shape what can or cannot be seen. With this understanding it becomes of central importance for researchers to recognise the ways in which interpretations of voice are limited or expanded through the methodologies employed. I have argued that dialogic methodology, through detailed scrutiny and dialogue with others presents a means of expanding and appreciating voice as a multiple construct within, between and beyond the ‘self’.

These results represent only a small sample of the findings generated out of this study. Others, associated with assessment ideologies and an emphasis on toddler language, are presented elsewhere (White, 2009a, 2009b, 2010; White & Peters, 2011). Combined, they suggest that dialogic methodology has a legitimate place in infant and toddler research since it can yield insights that are not so easily generated through traditional approaches. In doing so, toddlers, and perhaps other potentially marginalised voices, can be seen and heard in research activity and pedagogical assessment activity alike. As such, their unending potential can be revealed as multiple voices authentically emerge and challenge theoretical strongholds that limit possibility. The alternative is voicelessness, or as Bakhtin (1984) more dramatically suggests, ‘absolute death (non-being)’ (p. 284) when one is utterly ‘known’ by another. To be ‘seen’ is therefore not only a research imperative, but it is also a moral obligation for those who seek to understand more about the very youngest members of society and, if they are open to it, about themselves.