Mapping as a Cultural and Cognitive Universal

Abstract

We hypothesize that nearly all humans, in all cultures, acquire the ability to read and use map-like models in very early childhood, and that this ability is a fundamental part of human ecological adaptation, comparable in many ways to tool use. Evidence pertaining to this theory should be sought in three kinds of research: studies in differing cultures of the development of young children's ability to use map-like models; studies probing for evidence of map-like modeling across the ethnographic spectrum; and studies probing for evidence of the use of map-like models in prehistory. We are pursuing all three lines of research. However, our main focus thus far has been on the developmental dimension of the problem. Here, we report evidence that supports the universality hypothesis from seven empirical studies carried out on mapping abilities of three- to five-year-old children in several Western and non-Western cultures; we offer a general ecological theory of the development of mapping abilities, a theory that appears to explain the evidence elicited and accords with the universality hypothesis; and we discuss the implications of this work for early childhood education.

Key Words:

• children's geography
• cultural universals
• development of mapping abilities
• ecology of map-like modeling.

Acknowledgments

The research reported here was supported in part by a grant from the U.S. National Science Foundation (SBR9423865, GRS-9906604, 9906418). Very important contributions to this research were made by Zhra Darvizeh, Silvia Elguea, Vanessa Eckert, Diane Hetherington, Dennis Kerkman, Gloria Lara, Nancy Middlebrook, Marites Pinon, Anne Reid, Jennifer Rice, Steve Sowden, Roy Surajpaul, David Uttal, and Monica Varela. We are also grateful for the research assistance of many other individuals: In addition to those named in the Notes to this article, we wish to extend special thanks to Len Hetherington, Joe Palka, Dhiru Soni, and Jennifer Stephens.

Notes

1. The generalization is intended to apply to blind as well as sighted people.

2. We are examining archeological and ethnographic evidence of map-like models and modeling, but space does not permit us to discuss this work in the present article.

3. Our concept of map-like model is compatible with most though not all definitions of “map” that one finds in the cartography literature. (See, e.g., Raisz 1948; Thrower 1972; Keates 1982. On the relation between this concept and other useful concepts, see, e.g., Harley and Woodward 1987a; Delano Smith 1987.) The map-like model would generally be seen as one of a number of kinds of representation that are, and should be, called maps. We think there is agreement that most ordinary, traditional, topographic scale maps have the characteristics listed by us as primary: a landscape viewed from an overhead perspective, reduced in scale, displaying features with signs. Raisz (1948, xiii) comments that “a map is, in its primary conception, a conventionalized picture of the earth's pattern as seen from above.” In this article, however, we are not defining “map.” We suggest that the culturally and psychologically basic or primitive way of thinking about and materially representing a geographic landscape is the map-like model. The notion of map syntax is also commonly used in the cartography literature (see, e.g., Keates 1982; Head 1984; Schlichtmann 1985; MacEachren 1995; but see Robinson and Petchenik 1976 and Robinson 1978 for a differing view).

4. These errors involve identifying a landscape feature seen on the air photo as something not at all realistic at the map scale or the map perspective: for instance, calling a tennis court a door, an error both of scale and of perspective. The term “scale error” is commonly used for what we call here “syntax error,” but some of the difficulties relate to perspective, not scale.

5. See Blades and Spencer (1994) on this problem; Matthews (1985) makes good use of this technique for somewhat older children.

6. This study was carried out by Steve Sowden in collaboration with the present authors and is reported in Sowden et al. (1996).

7. The fact that six children did not succeed at the navigation task does not imply that they were cognitively incapable of performing the tasks. Four-year-olds may be inattentive, may give playful responses, may be shy, may not want to cooperate, and so forth. At this age, rapid development of linguistic and social competence is taking place, and this may influence performance. It would be inappropriate to say, then, that fourteen of the children had the mapping ability but six did not.

8. This study was carried out by Mark Blades, Steve Sowden, and Christopher Spencer. A summary report has been published (Spencer 1998); an article fully describing the study is in preparation.

9. We discern a pattern in syntax-error reports that seems to suggest an underlying social cause (within this culture), rather than a cognitive limitation. In study 2, we noted that syntax errors occur mainly when the experimenter is probing for a specific response and are almost nonexistent when the children give free responses, although even in the specific-response probing the children much more frequently said “I don't know” than gave a syntax-error response. It seems probable that a child, wishing to respond to an interlocutor's question but not really knowing an appropriate answer to the question, will be tempted to give a pure guess or a playful response instead of an “I don’t know.”

10. This study was coordinated mainly by J. M. Blaut. The experiments were carried out by Steve Sowden in York, by David Stea and Silvia Elguea in Mexico City, by Roy Surajpaul and Dhiru Soni in Durban, by Zhra Darvizeh in Tehran, and by David Uttal in Evanston. See Blades et al. (1998). The research that Dr. Darvizeh undertook in Tehran was carried out entirely under her own direction and with no external funding. Dr. Uttal will report on his experiment in a separate article.

11. See Blades et al. (1998) for more detailed discussion of these methodological issues

12. Research has begun on the semantics of map-reading in preschool-age children, using child-defined map-signs.

13. This study was undertaken later than, and separate from, the cross-cultural study reported as study 3. The research was carried out in a village in which a long-term project to develop a model of community-based sustainable coastal management was under way under the direction of Dr. Anne Reid, Professor of Psychology at the Universidad Autonoma Metropolitana-Iztapalapa. The research was carried out by Anne Reid, David Stea, and two graduate students from the Universidad Autonoma Metropolitana-Iztapalapa, Monica Varela Cuevas and Gloria Lara Millan.

14. The study was designed and coordinated by Mark Blades and Christopher Spencer and was carried out by Diane Hetherington, Steve Sowden, and Sally Craddock.

15. Although oblique air photos were more legible than verticals, the success of the four-year-olds in working with the verticals was so apparent that the advantages of obliques, particularly in education, should be minor (and, of course, vertical air photos provide information that is much more map-like than do obliques).

16. This study was carried out by Steve Sowden in collaboration with the present authors and has not yet been reported in a publication.

17. We say “will” and not “can” because this experiment does not by any means plumb the cognitive abilities of three-year-olds and four-year-olds; we should anticipate better performance if we ask the children to do something specific and geographical with the toys, as we are doing in current research.

18. Gender differences were not found in the studies carried out in England, Mexico, South Africa, and the U.S.

19. Other syntactic and semantic transformations are involved in reading complex maps with complex projections and signs. There is also a grammar or syntax governing the map-signs and sign patterns. See Head (1984), Lyutyy (1985), and Schlichtmann (1991).

20. Doubtless there are important connections between protomapping and language development, connections that would work in both directions. For instance, the recognition of meaning in a toy or self-created play object seems likely to be one of the ways that a young child learns about semantic meanings in general, since semantic transfer must take place between words and material objects such as toys and between toys and the larger things they signify.

21. In addition, four-year-olds play action games (such as blind-man's-buff) that require the forming of a cognitive map of a territory, but this seems to be a form of behavior rather distant from protomapping. A young child often observes and avoids walking on cracks in the sidewalk. Does this bear some resemblance to mapping? At what age does it appear?

22. See, e.g., Hirschfeld and Gelman (1994), Elman et al. (1996), Pinker (1994), and Wallace (1989). We may recall that Erwin Raisz (1948,; 3) suggested that mapping is “an inborn aptitude of humanity.” Although Blaut (1991) discussed the possibility that there may be an innate map-acquisition device homologous with Chomsky's (1965) postulated language-acquisition device, this remains an unsupported speculation.

23. Muir developed a successful first-grade map-skills curriculum based on air photos (see Muir and Blaut 1970). The research on air-photo interpretation by young children has influenced England's National Curriculum for geography, Key Stage 1, for children ages five through seven (Department for Education 1993; Spencer 1998).

24. There is a fourth direction—study of the brain structures and functions involved in map cognition and behavior—but we have not yet found a way to work in this area.