Presentation of the Children’s Drawings of Gods Project

The international project Drawings of Gods: A Multicultural and Interdisciplinary Approach to Children’s Representations of Supernatural Agents, known in French as Dessins de dieux (DDD), and referred to in this volume as Children’s Drawings of Gods, has collected several thousands of pictorial representations of supernatural agents drawn by children from different countries. The project aims to conduct an international survey in order to uncover major trends in the ways that children graphically represent supernatural agents, and more specifically, figures of god. This project is resolutely open to receiving drawings from multiple cultural, religious, and linguistic horizons; this project’s field of action is not limited to Western Christian cultures. By means of a common protocol for collecting drawings from school-age children, eight countries (to date) have served as survey sites for data collection (Argentina, Brazil, Iran, Japan, Romania, Russia, Switzerland, and The Netherlands). The project has acquired drawings from these eight countries, as well as small number of drawings collected in Nepal (using the same protocol), and nearly a thousand drawings that had been previously collected in the United States (using a different protocol). In total, some 6500 drawings have been produced, digitized and integrated into the project’s database (for details, see below).

While the international scope of the Children’s Drawings of Gods project calls for important organizational management, its main ambition is to offer a rich and varied field that is conducive to an intercultural perspective (Brandt et al., 2009; Dandarova-Robert et al., 2016). In the past, the study of drawings of gods has been approached mainly from a developmental perspective, without much concern for interculturality (Harms, 1944; Hanisch, 1996; Ladd et al., 1998). This project aims to remedy this lack by analysing intercultural comparison and interreligious variation between the drawings of supernatural agents. Acknowledging the intercultural variation allows us to recognize the complexity of the problem and how it is reconfigured in more or less similar/dissimilar cultural contexts.

Why does comparison teach us so much? Comparison allows us to relativize the dominant conception of the anthropomorphic figure of God, which tends towards more and more abstract representations according to the child’s development. The attention paid to intercultural and interdenominational comparisons highlights the complexity of the figures of “god”. This complexity calls for a dialogue between disciplines, both within the sub-disciplines of psychology (psychology of religion, developmental psychology and intercultural psychology) and among the study of religions, cognitive sciences, social and cultural anthropology, and sociology of science.

A Work in Progress: From the Genesis of the Project to the Present

The project did not have high ambitions when it began. It began in a modest study conducted in 2000 by an undergraduate student, Carole Herren, under the supervision of Pierre-Yves Brandt at the University of Lausanne. Herren attended the course Introduction to the Psychology of Religion given by Pierre-Yves Brandt and was especially interested in the lecture on children’s representations of God. This lecture presented, among others, the study “The development of religious experience in children” by Ernest Harms (1944; republished in 1973). Harms based his study on more than 5000 drawings collected among children in public and private schools in the United States. This study can be considered as the princeps study of children’s representations of supernatural agents by the method of drawing. Children “were asked to try to imagine how God would look to them, if they were to picture Him in their mind, or to imagine the appearance of the highest being they thought to exist.” (Harms, 1944, p. 114). Herren collected 27 drawings from children between 8 and 13 who attended religious education classes in the Eglise évangélique réformée du canton de Vaud (EERV). This Swiss Evangelical Reformed church has its origin in the sixteenth century, when the canton of Vaud adopted the Reformation professed by Calvin. It is a reformed denomination, and has, in this canton, the status of a state church. The majority of the drawings collected by Herren depict a human figure, drawn from a front view. In some of them, the child has drawn only a human face. These few drawings were the occasion for an initial publication on the psychological roots of the Divine Face that find various expressions in many religious traditions (Brandt, 2002). In autumn 2000, Brandt enriched his course on children’s representations of God by supplementing his lectures with some the drawings collected by Herren. One year later, Yuko Kagata, a Japanese student attended this lecture. Looking at the Swiss drawings through the lens of her own cultural background, she was convinced that the proportion of anthropomorphic representations of gods would be drastically lower in the drawings of Japanese children, when compared to those collected in Western countries. After some discussion with Pierre-Yves Brandt, she decided to test this hypothesis and collected 142 drawings during two stays in Japan in 2003 and 2004. She wrote her master’s level dissertation on these data, under the supervision of Pierre-Yves Brandt and Christiane Gillièron at the University of Geneva (Kagata, 2006). Contrary to her expectations, she obtained a large proportion of drawings with anthropomorphic figures. In another very interesting finding, Kagata noted that almost half of the girls drew female divine figures (Brandt et al., 2009).

After examining this interesting data from an Asian country, a decision was made to collect drawings in Switzerland following the same procedure. Two students of the University of Lausanne, Anja Kniffka and Aurélien Schaller, collected 127 drawings in Swiss public schools and parishes in 2004 and 2005 in partial fulfilment of the requirements for their master’s level grades in the sciences of religions. The students modified the task slightly to reduce the anthropomorphism in the children’s representations of god that could be implicitly suggested by the instruction given to the children. For that reason, they did not ask children to imagine “god”, but instead suggested that they draw “all that comes to your mind when you think to the word ‘god’”. Posterior analyses, comparing the proportion of anthropomorphic representations in this sample with another sample collected in Switzerland in 2008–2010 (which used the same instruction that had been used in Japan) showed that the variation in the instruction had an impact on the results. The proportion of anthropomorphic representation in this Swiss sample from 2004 to 2005 is significantly lower than in the Swiss sample from 2008 to 2010 (Dandarova-Robert et al., 2016, pp. 349–350). For example, in the 2004–2005 sample, one boy (7 years old) drew a caravan, saying that the word “god” makes him think of family holidays in caravan. Without his explanation, this drawing produced in answer to the instruction to draw “god” would have led researchers to think that, for this child at the time of drawing, “god” had the physical appearance of a caravan. This illustrates the fact that the instruction to draw “all that comes to your mind” tends to induce an associative task (draw something that is associated with the concept of “god”) rather than a descriptive one (try to produce a drawing that illustrates the concept of “god”). This difference in instructions given during the collection process prevented a direct comparison between the drawings from the 2004 to 2005 sample and those collected in Japan in 2003–2004.

Later, in 2008, Zhargalma Dandarova joined the project and began to collect drawings in public schools and Christian Orthodox parishes in Saint Petersburg and in schools in Buryatia (Eastern Siberia). In sum, 754 drawings were collected in these regions between 2008 and 2015. In addition, 2008 was also the year when a new collection of drawings began in public schools and Protestant and Catholic parishes in Switzerland. Researchers ran a first wave from 2008 to 2010. During the same period, an open access, web-based database was launched under the link All drawings collected up to that time were scanned and uploaded to the database. In addition, we added to the database 993 drawings collected in 1987 in the United States by Kevin Ladd. Subsequently, new collections have been added. First came 400 drawings collected by Camelia Puzdriac in Romania between 2010 and 2013. Later, 302 drawings were collected during the “Mystères de l’Unil 2014”, a 4-day “Open Days” event in May 2014 during which school classes and families were able to visit the University of Lausanne. This Swiss sub-collection also contains drawings made by adults. Then, thanks to a 4-year research grant from the Swiss National Research Foundation (SNSF), Grégory Dessart completed a second wave data collection to complete the Swiss subcollection, and additional partners from other countries joined the project. Today, the database contains more than 7000 drawings, including 158 drawings collected in the Netherlands under the supervision of Hanneke Muthert and Hanneke Schaap-Jonker, 139 drawings collected in Brazil by Alberto Domeniconi Küntgen-Nery, 13 drawings collected in Nepal by Thierry Luginbühl, and 3032 drawings collected in Iran by a team of researchers under the supervision of Mohammad Khodayarifard. Recently, Ramiro Tau collected drawings in Argentina, and new drawings from Romania have arrived. The integration of these new collections into database is underway. With the exception of Ladd’s 1987 collection, all of the drawings from this database were collected following the same procedure and, apart from the Swiss sub-collection 2004–2005, with a similarly formulated instruction. Almost every chapter of this book presents studies that directly refer to the different sub-collections of this database.

Procedure and Instructions

Overall, the procedure for the participants includes four parts and takes between 30 and 50 min to complete:

  1. 1.

    children are asked to make their drawings;

  2. 2.

    once the drawing task is complete, children are asked to recall and write on the back of the sheet the instruction received at the beginning of the drawing task;

  3. 3.

    children are asked to describe their drawings on the back of the sheet (narratives);

  4. 4.

    children fill out a questionnaire.

Before presenting the main task, researchers ask the children to remain silent (to refrain from speaking any comments or questions aloud). Then they say:

Have you ever heard of the word ‘god’? Could you draw, please? You can draw anything that comes up to your mind when you think of the word ‘god’. Keep silent and do not let your friends to see your drawing. When you finish your drawing, raise your hand, please.Footnote 1 (Dandarova-Robert et al., 2016, p. 349)

The researcher takes care not to use masculine pronouns such as “he” when referring to god. Children are asked to raise their hand if they have questions, so that a researcher can speak to them quietly, one-on-one. These precautions are necessary to minimize the impact of one child’s representation of god on other children. For example, if one child asks loudly, “Can I draw Jesus Christ?” or “Can I draw a god in heaven?” it can affect the way other children compose their representation of god.

Children are asked to raise their hands when they have completed their individual drawings. According to our experience, the children do not finish their drawings at the same time, so they receive individual instruction for the second, third, and fourth tasks. As a second task, researchers then ask the children individually to restate the instructions provided to them in the drawing portion of the procedure. Researchers use the following prompt:

Do you remember what I asked you to do? What did I ask you to draw? Please, write the instructions I gave you on the other side of the sheet.

Then, as a third task, the children are asked to write a description of their drawing. Researchers use the following prompt:

Can you now provide a written description of your drawing to explain what you drew?

Another formulation sometimes used, was:

Imagine that you should describe it to a blind child. Can you write it on the back of the sheet?

After the children finish writing their description, they are asked to complete a questionnaire. With the younger age group (5–8 years old), the researcher interviews each child separately and records the child’s answers for the three last tasks. At the end of the four tasks, the researcher collects the drawings sheets and the questionnaires. He or she looks over each child’s drawing and description. If something is not clear in the drawing and/or in the description, the researcher asks the child to clarify it (for example, if the description contains no information about the figures drawn, or if the researcher sees the need to for additional information about some other details of the drawing). Likewise, the researcher verifies the completion of the questionnaire to assure that all questions have been answered and all necessary information has been provided (for example, date of birth, etc.).

In addition, metadata are collected. This can occur either before or after the drawing task has been completed. Children are asked to write their name, gender, age (date/month/year) the date of the data collection (date/month/year), and the name of the school, all on the back of their drawings.

As material, each child receives one blank sheet of A4 white paper, a box of water-resistant wax crayons (with 8–12 colours; these eight: blue, green, red, orange, yellow, brown, black, white; with the possible addition of these four colours: purple, grey, light blue and dark blue, light green and dark green). In some countries coloured pencils (with the 8–12 above-mentioned colours), and a pencil and an eraser were also provided to children.

Group size did not exceed ten children (one child per desk, in order to avoid the children seeing the drawings of their neighbours and communicating with each other during the session). Children and adolescents ranging in age from 5 to 18 years old participated.

Ethical Considerations

For this research, it is important to preserve the spontaneous composition of the representation. Therefore, it is important that the children do not know in advance precisely what they will be asked to do so that they will not discuss the task beforehand, either amongst themselves or with their parents. Consequently, the researchers provide an information sheet to the parents that presents the general aim of the research. It indicates that their children will participate to an international study in which the children draw pictures and answer some questions. Researchers also requested that parents grant permission for the researchers to display the drawings online, with the guarantee that only the first name of child would be required and that confidentiality would be protected. Parents had the option to ask that their child not participate, or to request that the access to their child’s drawing be limited to researchers only. Some parents did request this restricted access. Children also received general information regarding the task. More detailed information was provided to teachers (or school directors), but they were asked not to tell children the specific task of the study in advance.

Researchers informed children that they could decline to participate in the study or could withdraw from any part of the study at any time. In our experience, some children declined to make a drawing saying: “Drawing God is forbidden”, or “Only specially trained artists are permitted to draw god”, or “I do not know what god looks like”. In such cases, researchers asked the child to write on the back of the paper their reason for declining and requested that they fill in the questionnaire. Because such responses have empirical value for the research, researchers retained the blank sheets of paper with the explanations provided on the verso and added them to the database as part of the sample.

God, “God”, Gods, Supernatural Agent: A Note on Terminology

In this book, the term god sometimes begins with an uppercase letter G, sometimes with a lowercase letter g; it is sometimes presented in the singular, sometimes in the plural. The project, Children’s Drawings of Gods, is interested in children’s representations of the divine. When they receive instructions verbally, in some languages such as French or Japanese, for example, there is no difference in pronunciation between the singular and the plural. Furthermore, the oral pronunciation does not distinguish between upper and lower case letters. This is why, in general, we write “god” with quotation marks when we refer to the instructions provided to the children. The children have heard the word “god” and they have drawn representations of “god”. When we write “god” with quotation marks in reference to a French-speaking sample, it means that we refer to the instructions given to the children, which used the French word dieu. Similarly, when we refer to samples from other locations, the word “god” with quotation marks indicates the use of a word, in the translated instructions, that stood for god: kami in Japan, bog in Russia, khoda in Iran, deus in Brazil, dios in Argentina, etc. When we refer to the concept of god or to images of gods, we write the terms god, God, or gods without quotation marks. So, the expression “representations of gods” refer to representations of gods in general. Sometimes, we decided to write “representations of God” and not “representations of gods” because we refer to a cultural context where the possibility of having various gods is not plausible. This is especially the case in Iran where “khoda” is understood as the unique God. In the cultural context of Iran, when children receive the instruction to draw “khoda”, they understand that they are asked to draw representations of God and not of gods.

Finally, the term “supernatural agent” is the most encompassing. We did not use it with the children, since the instructions ask them to draw “god”, and also because it is too abstract for them. However, it is clear to some children that they have not really drawn a god. Instead, they drew a supernatural being that they would not spontaneously describe using the term “god”. When we want to emphasize that we are aware of this, we use the term “supernatural agent”.

Organisation of this Book

This book is composed of this introductory chapter, followed by 19 thematic chapters organised into nine parts.

The first part, “Towards an Integrative Model”, contains a single chapter entitled “Integrative Model of Children’s Representations of God in Drawings” (Brandt, Dandarova-Robert, Dessart, Muthert, & Schaap Jonker, Chap. 2, this volume). This opening chapter emphasizes one of the main goals of our research project: the integration of the results of different psychological studies conducted on children’s drawings of gods. This chapter draws much of its data from several other chapters of the book. Consequently, we could have placed this integrative chapter at the end of book, instead of placing it immediately after this introductory chapter of the book. We decided that it would be more stimulating to put it at the beginning of the book, to offer, from the outset, an overall vision that invites the reader to read the chapters on which this integrative chapter is based. To develop a detailed understanding of data assimilated in Chap. 2, it is worth reading the chapters that undergird this integration, specifically those on anthropomorphic and gender features (Part II), on emotional features and attachment styles (Part IV), and on the impact of the cultural context (Part V). Further, the information found in this chapter moves beyond this volume as it integrates not only the results published in other chapters of this book, but also the results published by other researchers outside of this book. The scope of this chapter serves both to integrate and to contextualize the project and its many facets.

The second part, “Focus on the Main Figure: Anthropomorphic and Gender Features”, consists of five chapters. It begins with a chapter entitled “Children’s God Representations: Are Anthropomorphic God Figures Only Human?” (Dessart, Chap. 3, this volume). This first chapter is a theoretical endeavour in which the author reviews the literature on anthropomorphism in children’s representations of gods and proposes a revised developmental model of children’s use of anthropomorphic features in god representations compared to human representations. The next chapter entitled “Humanness and Non-Humanness in Children’s Drawings of God: A Case Study from French-Speaking Switzerland” (Dessart & Brandt, Chap. 4, this volume) takes the model set out in the previous chapter and applies it to a sample of children’s drawings of gods collected in Switzerland. This chapter is followed by a chapter entitled “Construction and Transgression of Gender Categories in Representations of Divine Figures: A Cross-Cultural Study of Children’s Drawings” (Dessart, Dandarova-Robert, & Brandt, Chap. 5, this volume) that explores the dimension of gender features in children’s drawings of gods collected in four cultural areas: Japan, Switzerland, Buryatia (Russia), and Saint Petersburg (Russia). These same samples provide the data for analysis in the last two chapters of the second part of the book. Both of these chapters deal with the location of the god figure in the drawing. The chapter entitled “Where Gods Dwell? Part I: Spatial Imagery in Children’s Drawings of Gods” (Dandarova-Robert, Cocco, Dessart, & Brandt, Chap. 6, this volume) focuses on the background of the god figure. It shows that, regardless of the culture to which the child belongs, children tend to draw god either in a celestial context or without a background. Based on the ideas of embodied and grounded theory, the chapter entitled “Where Gods Dwell? Part II: Embodied Cognition Approach and Children’s Drawings of Gods” (Dandarova-Robert, Cocco, Dessart, & Brandt, Chap. 7, this volume) takes into consideration the spatial location of the god figure on the sheet of paper, and it shows that children tend to position their god figures in the upper part of the page.

The third part of the book, “Focus on Material Features”, is devoted to the use of computer vision algorithms for analysing the drawings. In the chapter entitled “Automated Colour Identification and Quantification in Children’s Drawings of Gods” (Cocco, Dandarova-Robert, & Brandt, Chap. 8, this volume), the children’s drawings from the same four cultural contexts described for part two, above (Japan, Switzerland, Buryatia and Saint Petersburg), were analysed in terms of colour preferences. Age, gender, and cross-cultural comparisons were conducted. The analyses highlight the privileged role played by the colour yellow in drawings of god. Blue and achromatism (grayscale) play complementary roles that vary across cultural contexts. The chapter entitled “Computer Vision and Mathematical Methods Used to Analyse Children’s Drawings of God(s)” (Cocco & Ceré, Chap. 9, this volume) draws on the same dataset. Analyses were based on features extracted from manually executed annotations (god position, anthropomorphic features) and features that were computed automatically (gravity center, colour frequencies, colour organisation). Then, numerical measures of differences between drawings were calculated from the data, and analyses based on these dissimilarities (multidimensional scaling and clustering) were conducted. The results in this chapter support the consistency of the findings presented in earlier chapters, but these authors analyse the data differently, by means of systematic numerical measurements tied to certain material features of the drawings. It is a first attempt to develop methods for analysing pictures that do not refer only to methodologies that rely on inter-judge convergence.

The fourth part of the book, “Focus on Emotional Features and Attachment Style”, extends the rather cognitive approach that more or less strongly underlies the previous chapters, to include the explicit consideration of the emotional dimension expressed in the drawings. It contains two chapters. In the chapter “Emotional Expression in Children’s Drawings of God” (Jolley & Dessart, Chap. 10, this volume), two artists evaluate the data from the Swiss sample, scoring each drawing for emotional intensity and valence. The emotional dimension of representations of gods had not yet been studied and these data have allowed us to expand our understanding, as shown in the integrative model presented at the beginning of this book. The same can be said of the other chapter in this part, entitled “Different Attachment Styles in Relation to Children’s Drawings of God: A Qualitative Exploration of the Use of Symbols in a Dutch Sample” (Muthert & Schaap-Jonker, Chap. 11, this volume). In this chapter, the authors conduct qualitative analysis on 24 drawings that were collected in the Netherlands. Results show that drawings from children with secure attachment seem to contain more positively connoted god representations. These results are accompanied by a strong literature review on the links between attachment theory and god representation. As a result of these findings, we have been able to enrich the integrative model that introduces this book with references to the attachment theory.

The fifth part of the book, “Focus on Specific Cultural Contexts”, contains two chapters devoted to specific cultural contexts that are not taken into account in the previous chapters. Both of these studies follow the same methodology of data collections, but the authors use, in part, different methods for analysing the data. For the research described in the chapter entitled “Iranian Children’s Drawings of God: Demographic and Contextual Considerations” (Khodayarifard, Pourhosein, Pakdaman, & Zandi, Chap. 12, this volume), more than 3000 drawings were collected in six different areas of Iran representing a variety of cultural contexts: Tehran, Savojbolagh, Sanandaj, Sari, Neyriz, and Tabriz. In Tehran, Neyriz, and Sari, people speak Persian. However, Mazanderani is the language informally spoken in Sari. In Savojbolagh, people speak both Persian and Azerbaijani. In Tabriz, they speak Azerbaijani, and in Sanandaj, they speak Kurdish. While Shia Islam is the official religion of Iran, people of Sanandaj are Sunni Muslim. The sample size allows a quantitative approach to the data and, thanks to the diversity of the six sub-samples, offers the possibility of intra-cultural comparisons within the Iranian society itself. The chapter “The First Discoveries and the Challenges of Researching Representations of Gods in a Continental Country such as Brazil” (Küntgen-Nery, Mendonça Torres, Guerreiro Vasconcellos, & Zangari, Chap. 13, this volume) also proposes an intra-cultural comparison. For this chapter’s research, drawings were collected from two different areas of Brazil: in the city of São Paulo, and among the Guajajaras Indians in the state of Maranhão. In this case, however, due to the size of the sample (n = 116), only an exploratory approach was possible. The results of these two studies are also partly included in the integrative model that introduces our book.

The sixth part of the book, “Focus on Non-Representability and Prohibition”, addresses the question of the irrepresentability of God. During each data collection event, when collecting children’s drawings of god, there are always some children who return a blank sheet of paper. Sometimes they explain that the task is impossible because no one has ever seen “god”, sometimes they state that it is forbidden to draw God. This debate is not new in the history of humanity. The chapter “Biblical Aniconism? Representing the Gods of Ancient Israel and Judah” (Römer, Chap. 14, this volume) introduces this topic in the context of the ancient Near East where aniconic representations of gods, like empty thrones, were found. The chapter “The Representation of God in Islam and its Prohibition: Strategies Used by Iranian Children When Asked to Draw God” (Astaneh, Chap. 15, this volume) analyses the Iranian drawings collected by Khodayarifard and colleagues, but from a perspective that specifically seeks to understand both the idea behind the prohibition and ways people (children especially) deal with the limitations imposed by such a prohibition. Astaneh shows that returning a blank sheet of paper is only one strategy, among others, chosen by children to avoid an iconic (or even an anthropomorphic) representation of God.

The seventh part of the book, “Focus on Comparison with Other Supernatural Agents” contains only one chapter entitled “Natural and Supernatural Agents: Children’s Representations of Gods and Dead Entities” (Tau, Chap. 16, this volume). This chapter presents two studies on children’s drawings collected in Argentina. In study 1, children were asked to produce a drawing related to the topic of human death. In study 2, children (a different sample) were asked to draw “god”, following the procedure described earlier in this introductory chapter. Both studies provide data on how children draw supernatural beings and place them in space, but the data were collected under two different sets of instructions. This allows the researchers to draw some comparisons between the samples, not only for differences due to the variation in instructions, but also for differences between the representation of “god” and the representation of supernatural beings (divine or not) after death.

The eighth part of the book, “Focus on the Research Process”, is composed of three chapters that further expand the interdisciplinary dimensions of the project. The chapter “‘Equipping Work’ and the Production of a Large-Scale Digital Infrastructure: An Ethnographic Inquiry into the ‘Children’s Drawings of Gods’ Project” (Vinck & Oberhauser, Chap. 17, this volume) adopts an ethnographic approach to describe how the data production and management had an impact on the evolution of our project. The authors include a reorientation of some of the research questions and even produce new ones. The chapter “Brief History of the Database ‘Children’s Drawings of Gods’ (2015-2019)” (Serbaeva, Chap. 18, this volume) complements the previous chapter by documenting the evolution of the project’s web-database. The chapter “Interdisciplinarity, Team Science, and the Next Generation of Researchers: The ‘Children’s Drawings of Gods’ Project Experience” (Darbellay, Chap. 19, this volume) takes a step back and proposes a broader reflection on the issues of an interdisciplinary project and the challenges faced by such an undertaking.

Finally, the ninth part of this book contains a conclusive chapter entitled “Interdisciplinary Approaches to Drawings of Gods: Challenges, Achievements and Perspectives” (Brandt, Dandarova-Robert, Cocco, Vinck, & Darbellay, Chap. 20, this volume) which discusses the contributions and limitations of the methodology of drawings for studying children’s representations. The authors then assess the benefits and limitations of interdisciplinary approaches that combine computer vision, database management, and developmental psychology. Finally, they appraise new perspectives of research on children’s drawings of gods in the fields of religious sciences and religious art, with regard to both the study of creativity, and the developmental psychology of norms.