Collaborative cognition: Co-creating children’s artwork in an educational context

The study of child art within developmental psychology

Child art has been studied within developmental psychology since the 19th century and is an established area of research (Coates & Coates, 2006). Broadly speaking, interest in children’s drawings has been informed by two different approaches: one centring on mapping out developmental patterns and the other on psychological assessment. An experimental approach has largely informed research that aims to identify key milestones in children’s drawing development. Within this body of research drawings completed by children of various ages were used to propose stage theories relating to specific areas of children’s artistic development, such as the representation of the human figure (M. Cox, 1993). It is argued that the invariant stages of development identified in this research revealed what normal children of a certain age would include in their drawings. This gold standard of normal development along with psychoanalytic theory has informed the use of children’s drawings as assessment tools. In a clinical context children’s drawings are construed as “an expression of their (children’s) unconscious mind, something which isn’t easily accessible” (Wilson, 1993, p. 37). Drawing tasks have been used to access the child’s personality (Machover, 1949), current emotional state (Koppitz, 1968, 1984), and attitudes towards significant people or events in their life (Fox & Thomas, 1990; Sechrest & Wallace, 1964; Thomas, Chaigne, & Fox, 1989). Outside of a clinical context drawings have been used to access cognitive functioning and measure IQ (Harris, 1963; Silver, 1978, 1988, 1993). Within an assessment context it is assumed that the level of detail in drawings completed by children can be analysed by a trained adult to gain access to the child’s inner world or mental functioning.

This brief overview highlights that within developmental psychology children’s drawings have been largely conceptualised as a direct reflection of the child’s cognitive functioning, their inner emotional world, and their developmental maturity. As such the majority of research investigating children’s drawings conforms to what Potter (2000) has termed a cognitivist agenda within psychology. The cognitivist agenda is characterised by an individualist approach that privileges the internal workings of individual people and considers these to be the key source of psychological explanation. Consequently, children are conceptualised as “autonomous individuals” who are reducible to measurable mental phenomena (Stetsenko & Arievitch, 2004, p. 475). The wider context that shaped the creation of the artwork is not examined and experimental research within developmental psychology has focused on what Potter (2000, p. 35) has termed the “output” (drawings) of the “cognitive system” (the child). Drawings completed by children have been considered to be of value because they give insight into the underlying cognitive and emotional processes located within the child.

An alternative to cognitivism

Researchers who use Discursive Psychological approaches have developed an alternative framework to cognitivism, which suggests that cognitive phenomena are situated in discursive practices (Wiggins & Potter, 2003). Early research conducted by Edwards and Middleton (1986) demonstrated that in some contexts memory is constructed through social interaction and that collective remembering is a well-practised activity in which people use a number of linguistic devices to prompt, correct each other, and add in extra pieces of information to create a sequential narrative. Following the argument that memory is not simply an internal cognitive process, Middleton and Edwards (1990) called for research to explore “how people construct versions of events and their own mental processes within the practices of everyday conversation” (p. 110). A discursive approach to cognition has been applied to a number of processes such as shared knowledge, emotion, and child cognition (Edwards, 1997), attitudes (Wiggins & Potter, 2003), and the perception of noise (Stokoe & Hepburn, 2005). This body of discursive research suggests that a limitation of cognitivism is that it does not attend to the action orientation and co-construction of cognitive activities (Potter, 2000).

In response to the call for psychology to attend to the “the collective aspects of human existence” (Martin & Sugarman, 1999, p. 11) this paper brings a discursive approach to the area of child art. Rather than focusing on drawings completed by children in an experimental context, this article analyses interactions that occurred between children in primary school art classes as they created their own artwork. Previous classroom-based research that has investigated other forms of creative activity highlights the benefits of adopting this approach. Baker-Sennett, Matusov, and Rogoff’s (1992) analysis of classroom interaction between children as they wrote a script for a play demonstrated that children worked together to critique, embellish, and develop ideas. The presentation of creative planning as a flexible and socially negotiated process highlights the need for an exploration of what Coates and Coates (2006) have termed an “essential ingredient in each drawing’s production” (p. 221)—the social interaction which led to the creation of the artwork itself.

Process-centred research

A growing number of researchers studying children’s art have started to explore the utterances and interactions which lead to the creation of children’s artwork. Matthews (1999, 2003) has conducted considerable research centred on the actual artistic process rather than the finished product. Matthews utilised ethnographic methods to conduct longitudinal studies that investigated the artistic development of his three children, two grandchildren, and classes of children he taught in London nurseries. A specific concern was to examine the ways in which children go about creating artwork with an emphasis on the meaning and purpose children assign to the marks they make on the paper.

Matthews’ (1999, 2003) research offered a valuable insight into the drawing experience but a focus on investigating skill development and the meaning artwork has for children meant that the wider social interactions which shaped the creation of the artwork were not explored. Recent research conducted by Coates and Coates (2006) has moved into this area by exploring the relationship between young children’s (3 to 5 year-olds) drawings and their accompanying narrative. The ethnographic methods adopted by these researchers revealed that talk and interaction played an important role in the creation of artwork. Children jointly created narratives surrounding the images depicted, scaffolded the drawing processes, and drawing activities also gave children the opportunity to engage in social talk. This supported Thompson’s (2000) anecdotal evidence that drawing in a kindergarten class is a collaborative activity.

An advantage of process-centred research is that it allows an examination of psychological phenomena (in this case creating artwork) from the positions of the participants themselves (Potter, 2000). This article further builds upon a process-centred approach and draws on aspects of Discursive Psychological analysis to explore the ways in which the artistic process is socially mediated. Discursive Psychology is an analytic approach that lends itself to the focus of the current article as it works with naturalistic data to explore how “cognition” is co-constructed in social interaction (Edwards & Potter, 1992). Discursive Psychology is characterised by a radical relativist ontological position in which language is not seen as a window to the mind or a direct reflection of cognition. The analyst works at the micro level of talk to examine the action orientation of language—how language is used to achieve certain functions such as blaming (Wooffitt, 2001). As such the analyst does not move their attention beyond the immediate interaction and the wider context the interaction takes place in is seen as irrelevant unless the speakers themselves make direct reference to it (Schegloff, 1997).

An exclusive focus on the micro aspects of talk has faced criticism from researchers such as Wetherell (2001) who have stressed the importance of attending to the wider contexts which shape social interaction. Wetherell (1998) argued that by “rarely raising their eyes from the next turn in conversation” (p. 400) and only attending to the micro context Discursive Psychologists address just a “tiny fragment” of social life. Drawing on the work of Laclau (1993) and the suggestion that “society … can be understood as a vast argumentative texture through which people construct their reality” (p. 341), Wetherell (2001) likened wider social, cultural, and historical contexts as threads that weave through the micro context of social interaction. Following this analogy Discursive Psychologists isolate a small piece of fabric from the texture of society and analyse it in detail. However, they fail to take into account the wider threads outside of the boundary, which make up the rest of the fabric and indeed weave their way inside the boundary. This position has recently received support from Kaposi (2013), who demonstrated the importance of a rhetorical psychological analysis which incorporates local discursive action, moral and political context, and psychosocial context when researching identity construction. In response, this paper seeks to develop a multi-level analysis, which in part draws on Discursive Psychology to explore how the creation of child art is socially mediated, but widens analytic attention beyond the micro context.

The importance of materiality

The critical realist approach adopted by Sims-Schouten, Riley, and Willig (2007) stressed the need for a multi-level analysis which attends to the material context. Sims-Schouten et al. (2007) argued that in line with a critical realist framework “material practices are not reducible to discourse, or without meaning unless interpreted discursively; rather, material practices are given ontological status that is independent of, but in relation with discursive practices” (p. 102). Thus, critical realism involves a realist ontology, which assumes there is a real world produced by causal mechanisms (Bhaskar, 1979). However, unlike positivist accounts, causal mechanisms are situated in multi-layered social systems, which are context dependent and reliant on the actions of human activity (Howard & Davies, 2013; Pawson, 1989). In light of this, critical realists argue that it is not possible to directly and objectively access causal mechanisms. This is because people make sense of and interpret the material contexts that shape their experiences (Willig, 1999). Experiences are consequently mediated by intransitive knowledge which encompasses the material world and transitive knowledge which encompasses the ways in which we conceptualise the material world using language (Carter & New, 2004). A strength of a critical realist approach is that it acknowledges a relationship between materiality, social context, and human agency. This approach enables an analysis of talk, text, and the material aspects of human experience.

Nightingale and Cromby (1999) define materiality as “the elemental, physical nature of the world in which we are embedded, its ‘thing-ness’ and solidarity” (p. 11). The material world encompasses the physical things that surround us such as trees, sand, rocks, and buildings. It also includes the properties of these things such as the way they smell and what it feels like to touch them. The material world can appear in language, an eloquent piece of prose about mountains for example, but it is not reducible to it (Nightingale & Cromby, 1999). As such it is important to explore and acknowledge how the physical surroundings shape the way that phenomena are constructed (Stoppard, 1998; Yardley, 1996). This could mean examining embodiment (lived bodily experience and issues such as disability), the physical nature of the world (the physical objects that surround us as discussed above), and violence and power (from interpersonal experience to weapons of mass destruction). In line with this Parker (1992) argued that materiality is important because it places constraints upon discourse relating to physical coercion, the organisation of space, and the physical orientation of people to specific discourses and subject positions.

Cromby and Harper’s (2009) discussion of paranoia neatly illustrates the ways in which materiality and discourse work together to construct a psychological concept. Cromby and Harper (2009) demonstrated how psychology has traditionally conceptualised paranoia as a product of faulty cognition. This reductionist position fails to acknowledge the neurological basis of paranoia and the wider social and material contexts which can contribute to a paranoid state. Cromby and Harper’s (2009) multi-level analysis of paranoia demonstrated that paranoia is linked to brain features and therefore it is an embodied state of being. However, the feelings and physiological responses to paranoia such as increased heart rate need to be interpreted. This interpretation is mediated by both wider discourses available and embodiment as our brain shapes the ways in which we perceive the outside world. Finally, material conditions such as the areas people live in and social inequality are important to consider as they have been linked to paranoia. As such, paranoia is presented as a concept that is “co-constituted in the dialectical associations between subjectivity and relational, social and material influences” (Cromby & Harper, 2009, p. 335).

This article draws upon critical realism in order to explore the relationship between discursive and material contexts. A specific concern is to explore how cognitive activities associated with creating artwork such as planning and making marks on the paper are mediated by both social interaction and the material context they are created in. In line with the arguments above the following analysis focuses on the material constraints placed on discourse.

Ethnography as a method that moves beyond talk and text

Sims-Schouten et al. (2007) argued that one of the major issues faced by qualitative researchers who aim to explore the material world is the lack of a systematic method which enables analysis to explore discursive and material contexts. When working to develop a critical realist methodology Sims-Schouten et al. (2007) stressed the importance of a multi-level analysis which draws upon “discursive practice (e.g. Edwards, 1997), Foucauldian discourse analysis (e.g. Willig, 2001) and an examination of embodied, material and institutional practices which may be considered to have extra-discursive ontology” (p. 107). They argue that a successful analysis attends to the micro and macro levels of talk and the material practices which shape what is said. In practice this involved first conducting a literature search which included academic sources, government policy documents, and non-academic sources to identify the material practices and the dominant discourses that shape people’s experiences. The next step involved assessing the material contexts relevant to the population participating in the research through the use of questionnaires and completing fact sheets. The final stage involved analysing interviews. Sims-Schouten et al. (2007) argued that collection of these different types of data and use of different analytic techniques enabled links to be made between the analysis of the interview, the discourses available, and the material contexts which shaped experience.

The analysis presented in this paper is a multi-level analysis. However, unlike Sims-Schouten et al. (2007), the use of an ethnographic methodology in the reported research enabled a range of data to be collected for analysis together in a naturalistic context (the classroom). This allows direct links to be made between interaction amongst the children and the wider contexts these interactions took place in. Ethnography encompasses a number of methods such as interviews and observation to study people in naturalistic settings. A key aim of an ethnographic approach is for the researcher to be immersed in the “symbolic world in which people live” with a view to understanding “the meanings people apply to their own experiences” (Fielding, 1993, p. 157). In line with an ethnographic methodology, the first author worked as a voluntary classroom assistant for approximately 6 weeks on art projects held in the following classes in two Staffordshire primary schools: Reception (4–5 year-olds), Year 1 (5–6 year-olds), Year 4 (8–9 year-olds), and Year 6 (10–11 year-olds). During the art project each class received one art lesson per week which ran in the afternoon session. To summarise, a total of eight teachers (with an average class size of 28 children) participated in this research, 18 hours were spent observing each age group, and a total of 72 hours were spent in the classroom during this project.

When working as a classroom assistant the first author helped teachers set up and tidy art materials and sometimes played an active role in the art lesson by helping children who had queries and talking to children about their artwork. This enabled the first author to immerse herself in the classroom context and write a reflexive field diary based on her experiences and observations. In addition to this, video and audio equipment were used to record the last art lesson of the project and photographs were taken of all the artwork created during the recorded lesson. Collection of visual data was particularly important as it allowed the analysis to move beyond talk and text and explore the material context which shaped the children’s interaction and the child’s artwork.

Developing a multi-level analysis

The following analysis uses a case study approach which brings together a Discursive Psychology analysis and ethnographic commentary. Each case study first presents data collected during the ethnographic phase of the research to provide background information about the art activity and a photograph of the finished piece of artwork. Discursive Psychology is then used to analyse interaction that occurred between children which shaped the creation of the artwork. The case study then ends with a summary of notes taken from the ethnographic phase of the research and, where relevant, a photograph of the children working.

The combination of ethnographic commentary and Discursive Psychology allows a multi-level analysis which attends to the micro level of talk, the wider meso level of the art task (the ways in which the teacher’s instructions shaped the focus of the interaction), and the material context. The following analysis moves beyond the exclusive focus on the micro level of talk within Discursive Psychology and explores the possible use ethnography has for researchers who adopt a critical realist approach.

Please note that in the following case studies the child responsible for creating the artwork included in the analysis is labelled child artist. The other children involved in the interaction (who are labelled using numbers) were working alongside the child artist creating their own artwork. Therefore, all the children involved in the interaction would have created their own piece of artwork during the lesson.

Case study one: Planning the creation of artwork, taken from a Year 1 class

The children of the class (5–6 years) had been instructed by the teacher to select small squares of different materials and attempt to paint the same image, using the same colours, on each of the materials. Children were given the freedom to select which image they wanted to paint. The interaction analysed in the extract below shaped the creation of a piece of artwork entitled “My Hamster Lily” (see Figure 1).

OPEN IN VIEWER

Figure 1.

A Year 1 child’s painting of “My Hamster Lily”.

This extract demonstrates how the simple act of planning which colour(s) to use in a painting is a flexible, socially negotiated process. Significantly, it is a child working at the table and not the child artist who initiates discussion concerning colour choice. This locates the planning process for the artwork outside of the child artist. Child one’s assertion that “you need grey” (line 1) places them in a position of power. They tell the child artist which colour they should utilise in their hamster painting. Child two’s question “do you need grey” (line 3) challenges child one’s authority and initiates considerable discussion. During consequent interaction two children work with the artist to negotiate the colour of hamsters. This presents the planning as a dynamic process which requires the combined efforts of three children.

In line 5 child one asks “hamsters are grey aren’t they” to build consensus around their suggestion that grey is the correct colour. The child artist’s reply of “yeah” (line 7) reinforces child one’s colour choice—jointly constructing hamsters as “grey” animals. Moreover, the artist’s suggestion that hamsters can be “black and grey” introduces a new concept as hamsters are re-conceptualised as creatures that are not monochromatic. This gives child two opportunity to introduce another colour combination of “black and white” (line 9). Each turn of talk opens up new possibilities as colour choice is defined and redefined. Planning is not a solitary process in which individual ideas are transferred to the page. Instead, it is a dynamic social process in which language is actively used to construct different options that may or may not be represented in the child artist’s painting.

It is important to note that during the negotiation of colour the artist is at the centre of discussion. In line 12 child one directly addresses the artist with their suggestions of “brown” and “black and white.” Significantly, in line 14 the child artist incorporates child one’s proposals of “white” and “brown” into their initial suggestion of “black.” This symbolises the joint construction and development of ideas with the child artist teaming up child one’s suggestions with their own initial colour choice. However, the child artist is cut short in line 15 by child two’s unrealistic suggestion of “purple.” This interjection is noteworthy; it departs from the focus on choosing realistic colours and opens up the possibility of creating an imaginary hamster. This attempt to move away from the creation of real hamsters is quickly rejected by the artist with a defiant “no” (line 17) before they list their final colour choice of “black and then grey.” Even though the artist was ultimately responsible for which colours they used, colour choice was a product of social negotiation.

Ethnographic comments: The impact of spatial relations on planning artwork

The inclusion of the still photograph above (Figure 2) allows an examination of the material context and how the arrangement of classroom space enabled social interaction. During the ethnographic phase of the research it was observed that in preparation for art lessons the physical space of the classroom was often changed. In classrooms that usually had desks arranged in rows the teacher and the first author would arrive early to push tables together. This movement allowed small artistic communities to be formed as children gathered around tables to create their work together. Chairs were often removed and children were allowed to stand to give them more freedom of movement. This was significant—it made painting a more vibrant, physical activity and the classroom ceased to be quite as restrictive. This created an informal atmosphere where children did not have to conform to the usual classroom rules of sitting quietly in their seats and getting on with their work.

OPEN IN VIEWER

Figure 2.

Year 1 children creating artwork together.

The children gathered around the table in Figure 2 are not afforded much space to create their artwork. This close proximity combined with freedom to stand and move around means that all artwork can be seen clearly and commented upon or critiqued by other children at the table. This situation was specific to art. In other lessons such as English, children’s work can only be seen by those working next to them—other children could not interject without being invited. During art lessons children’s work and the processes involved in creating artwork were public rather than private—teachers would allow the children to walk around the classroom and look at the work being created at other tables. This highlights links between the discursive and material context as the physical arrangement of the classroom shapes the ways in which children can interact. In this case the teacher’s manipulation of the classroom space facilitated the creation of small artistic communities and encouraged interaction between the children, making the activity more social and collaborative. This evidences the teacher’s power in the classroom and one of Parker’s (1992) material constraints on discourse—organisation of space. Had the teacher not moved the tables and allowed the children to walk around the classroom and talk to each other the interaction evidenced above would not have been possible. The art activity would have been a more individual endeavour.

Case study two: Scaffolding the development of skills from the position of teacher, taken from a Year 4 class

The children of the class (8–9 years) were using charcoal to sketch objects associated with a journey they had been on. The teacher had instructed the class to use techniques such as shading to give their work a 3D effect. The following extract shapes a sketch of a Nemo soft toy (see Figure 3).

OPEN IN VIEWER

Figure 3.

Year 4 child’s charcoal sketch of a Nemo soft toy.

This extract illustrates how observational drawing is not purely a cognitive activity in which children create artwork that represents their own unique perception of an object. In line one the child artist’s request for help with “how” to “shadow” incorporates another child into the artistic process. Following this, both children work together to create a shadow effect. Child one starts this process (line 8) by telling the artist what they “have to do.” This statement asserts child one’s expert position construing them as someone who has the knowledge and skill to “shadow” effectively. A power imbalance is created between the children—child one’s superior knowledge gives them authority over the child artist. This position is discursively maintained as child one gives clear, direct instructions such as “draw that outline but just a bit smaller” (line 9) to the artist. Use of these instructions highlights that the physical act of making marks on the page is socially negotiated.

The shadow on the artwork is not simply a reflection of the child artist’s perceptual ability or skill in translating what they can see onto paper. Instead, creating shadow is discursively managed as the child artist is talked through the procedure. The child artist’s hand movement is guided by language and therefore any marks made on the page cannot be considered a direct reflection of their perpetual ability or artistic skill. Instead, the child artist takes the position of a student whose skills are being guided by a more capable peer/teacher.

Significantly, in line 9 child one’s direction to “do it there” symbolises a shift away from the drawing procedure to where the outline should be placed on the page. Creation of shadow ceases to be discursively managed as child one physically indicates where the charcoal outline should be positioned. This extends beyond the discursive negotiation of art demonstrated in the analysis so far. Child one begins to take over creation of the actual artwork—physically asserting their position as expert. Child one’s shift from instructor to artist is evidenced in their assertion that “I’ll do it” (line 9). Despite this declaration, child one also asks “what d’you want?” (line 10) thereby incorporating the artist back into the drawing activity. The artist’s response of “just make a mark” (line 12) reclaims their physical control over the creation of the artwork. It repositions child one as an instructor—someone who guides the creation of shadow rather than drawing it themselves. Consequently, child one’s efforts are directed away from making marks on the artwork to keeping the child artist on the right track with verbal prompts such as “you are gonna do the shadow even smaller”(lines 15/16). The way that shadow is first of all perceived and then represented on the page is discursively constructed. The artwork cannot be considered the property of one child or indeed one artist. It is the result of two children working together to discursively negotiate the correct procedure for drawing shadows and physically negotiating the marks made on the page. The origins and production of art is collaborative in every sense from the scaffolding of skills to making the marks evident on the page.

Case study three: The joint perception and creation of colour within an artistic community, taken from a Year 6 class

The children of the class (10–11 years) were working on designing and making a papier-mâché mask that represented an emotion or a character from their imagination. The discussion below centres on the creation of the base colour required for the sea queen mask (see Figure 4).

OPEN IN VIEWER

Figure 4.

A Year 6 child’s sea queen mask.

This extract illustrates how this artistic community works collaboratively to perceive and create the right shade of blue required for a mask. In line 3 the child artist asks if the other children “like” the blue colour she has mixed. This statement incorporates the rest of the group into the process of deciding the suitability of colour. Colour is not construed as an objective concept. Instead, the perception and understanding of colour is something that is negotiated within the group. This is evidenced in line 5 where child two’s response of “what colour is that…” diverts the interaction towards reaching a joint consensus regarding perception of the colour. Child two’s suggestion that the colour is “white and” (line 5) re-conceptualises the colour as being white and not blue as the artist first suggested. The incomplete sentence invites the child artist to clarify their conception of the colour by finishing their statement. In response to this the child artist changes their conceptualisation of the mixture construing it as “white and blue” (line 7)—a colour which embodies both children’s perception of the colour. Therefore, colour perception is not simply a cognitive process in which light frequencies received by the retina are decoded by the brain—it can be negotiated in social interactions. The same colour is construed as “blue,” “white and,” and “white and blue”—it is a point of contention and not an objective concept that can be viewed in only one way.

When negotiating the colour this community followed democratic, collaborative processes where all comments were equally valued. In line 9 child three joins the interaction and directs joint attention towards the “nice” colour created by child one. This marks the start of a discussion centring on whether the colour will achieve the right effect rather than how it is perceived. In line 11 child one initiates discussion on his blue colour by working to establish if it is the “sort of blue” that the child artist “needs.” To achieve this, child one utilises an incomplete sentence—echoing a linguistic strategy used by child two in line 5. By asking if the colour is “too” without specifying what it has too much of, child one leaves this open for discussion. Significantly, it is child two and not the child artist who construes the blue colour offered by child one as “a bit strong” (line 13). Consequently, child one lightens their blue colour by adding a “bit of white” (line 15). Colour is first jointly constructed through discourse and then physically altered with the addition of white paint. This highlights the link between language and the creation of material objects.

It is important to note that the child artist’s only involvement in the colour mixing process is their approval in line 17—“yeah that’s nice I like it.” The child artist had no involvement in creating the colour used to paint the mask. Instead, a colour initially created by child one was modified through interaction with child two.

Ethnographic comments: The provision of art materials and their role in shaping art

This photograph (Figure 5) illustrates how the arrangement of classroom space and provision of art materials facilitates discussion. The children are gathered around a table in which paints and mixing palettes are placed in the centre. The children did not have unlimited, exclusive access to all the paints and mixing space that they required. Paint selection and mixing took place in the centre of the table and involved sharing the paints and negotiating who would have access to the palette space. This meant that painting and mixing paints became a social activity. Joint attention and activity converged at the paints and mixing palette as children co-operated to share out the resources and negotiate who could have what when. The provision of limited art resources enabled social interaction but constrained what could be said and created. Discussion was limited to the colours provided and what could be done with them. Social interaction was tied to the material world.

OPEN IN VIEWER

Figure 5.

Year 6 children creating masks together.

This case study highlights how the teacher used the organisation of space, tools, and materials to shape the children’s interaction and the artwork they created. The actions of the teacher constrained the kinds of colours that could be mixed and the material objects that could be talked about and used. Therefore, the teacher took an active role in shaping the kinds of artwork (material objects) created by the children. The provision of tools and colours was regularly used by teachers to control the kinds of artwork created in the lesson and ensure that it met the learning objectives or in some cases their personal preferences. For example, after an art lesson a Year 1 teacher explained how she never gave the children black paint as it prevented them from painting black outlines around the figures they painted or making a depressing painting that used too much black. This highlights an important link between the micro, meso, and material contexts. Lessons delivered by teachers were shaped by the art curriculum and the learning objectives it stated. Therefore, the macro context, in the form of Government policy, shaped the meso context of the classroom and the teacher’s instructions to the children. It also shaped the material context of the classroom, in relation to the art materials provided and the arrangement of the classroom. These wider contexts shaped the micro context of the interactions between children as they followed instruction and used the materials provided to create art which met specified learning objectives.

Discussion

This analysis has clearly illustrated that creating artwork in a classroom context is not an abstract, individual activity. Close attention to classroom interaction demonstrated that the artistic process is a collaborative endeavour which is discursively managed. Therefore, the artwork examined in the analysis cannot be considered to be the work of one child or a direct reflection of their inner world. This raises important questions for the dominant view within developmental psychology that art is a purely cognitive activity in which the child’s individual ideas flow onto the page. Each stage of the artistic process from planning (case study one), and making marks on the page (case study two), through to colour mixing and perception (case study three) was socially negotiated within small artistic communities. The artwork created was not an embodiment of a single child’s unique ideas or vision nor was it a direct reflection of their intrinsic skill as an artist. Instead, creating art was a flexible, dynamic, and socially negotiated process. The presentation of art in this analysis further builds upon S. Cox’s (2005) argument that the way that children configure their drawings is not simply developmentally determined, it is a purposeful activity. In line with this argument research in the area of child art would benefit from a move towards the creation of artwork and the functions artwork serve for the children who create it. An advantage of using Discursive Psychology is that it gives insight into how cognitive processes are formulated in talk and provides an opportunity to investigate the artistic process from the child’s perspective. However, the inclusion of ethnographic notes and photographs point to limitations of an exclusive focus on interaction when analysing the creation of children’s artwork.

This analysis evidenced two of the material constraints on discourse outlined by Parker (1992): organisation of space and physical coercion. The teacher’s manipulation of the classroom space (case study one) and the materials (case study two) highlighted how the material context enabled and constrained social interaction. Case study two demonstrated how physical intervention was utilised to encourage a child to follow instruction. This analysis has therefore demonstrated how the interactions between children were bound to the material world in which they were produced. Furthermore, language shaped the creation of material objects—pieces of artwork. The interplay between the discursive and material contexts puts forward an argument for researchers using a social constructionist framework to explicate “intimate links between discursive and material aspects of our existence” (Yardley, 1996 p. 501). Wetherell (1998) draws upon the work of Laclau and Mouffe (1987) and the example of building a brick wall to further explore this point. Building a wall can be an activity that is discursively managed by interactions such as “pass me the brick” and also involves the physical act of placing bricks on top of each other to form the wall—a material object. A different set of discourses beyond those relating to construction can also be assigned to the material object of a wall, for example, it could be construed as a barrier or conceptualised as part of a house.

This was echoed in the creation of child art as the children’s interaction helped to create a material object—a piece of art. However, their interaction was shaped by the organisation of the classroom space, the tools they were provided with, and, at times, physical intervention. This clearly illustrates the relevance of the material context which has been debated by Harré and Bhaskar (2005). Taking a radical relativist position, Harré argued that there is nothing beyond the micro context and the material context can be reduced to language. Bhaskar adopted a more realist approach and suggested that all experience is constrained by the material world. As evidenced in this analysis the material, macro, meso, and micro contexts shaped the creation of artwork and the children’s experience of the classroom. This highlights what critical realism can bring to a multi-level analysis. Attention to material contexts brings important insight into why the utterances analysed at a micro level were made and the ways in which the classroom environment and the teacher’s instructions shaped and limited children’s subjective experiences.

The case study approach presented in this analysis points towards the usefulness of ethnography for critical realism and the development of a multi-level analysis. An ethnographic methodology enables a range of data to be collected simultaneously in a naturalistic context; this allows the analysis to contextualise social interaction in the material contexts in which it occurred. As such the role of the material context can be evidenced using video stills and photography and possible links between discursive and material contexts can be explored in the analysis itself. Furthermore, the researcher’s immersion in the context they are studying lends itself more to the kind of multi-level analysis that Sims-Schouten et al. (2007) have argued is needed for a critical realist analysis. The analysis presented here was taken from a much larger project in which the English National Curriculum for Art was analysed using a Foucauldian-inspired analysis to examine the dominant discourses and subject positions available to teachers (Hallam, Lee, & Das Gupta, 2007)—interviews were used to explore the ways in which their teaching practices were shaped by the curriculum (Hallam, Das Gupta, & Lee, 2008), classroom observation was used to examine the teaching of art (Hallam, Das Gupta, & Lee, 2011; Hallam, Lee, & Das Gupta, 2011), and a combination of classroom data and interview data was used to investigate the ways in which teachers decide what is good and bad art (Hallam, Lee, & Das Gupta, 2012). The range of data collection methods associated with ethnography gives researchers the opportunity to explore the micro context of immediate social interaction, the macro context in relation to the discourses working to enable and constrain interaction, and material contexts. This opens up the possibility for a more systematic analysis of the interactions between these contextual layers.