Essentialism and cultural belief in folkbiology 1 Running Head: ESSENTIALISM AND CULTURAL BELIEF IN FOLKBIOLOGY Folkbiological reasoning from a cross-cultural developmental perspective: Early essentialist notions are shaped by cultural beliefs

Two experiments examine the evolution of folkbiological reasoning in children (4 to 10 years of age) and adults from four distinct communities (rural Native American, rural majority culture, and suburban and urban North American communities). Using an adoption paradigm, we examine participants’ intuitions regarding the inheritance of properties and the mechanisms underlying the transmission of kindhood. Across all communities and ages, there was a strong biological component underlying reasoning about the inheritance of properties. There were also differences in children’s intuitions about the mechanisms underlying kindhood: Native American children were more likely than their counterparts to consider blood as a candidate biological essence. This suggests that as children search to discover the underlying essence of a biological kind, they are guided by broad essentialist notions that are shaped by discourse within their community. Essentialism and cultural belief in folkbiology 3 Biological Reasoning from a Cultural and Developmental Perspective To interpret the natural course of everyday events, we depend heavily upon intuitive notions about the objects and events in the physical world, about the entities and processes in the biological world, and about the rather specific beliefs and predilections of our fellow human beings. These intuitive ideas, dubbed naïve physics, naïve (or folk) biology, and naïve (or folk) psychology, respectively, have been the focus of active investigations by cognitive and social psychologists alike. At the same time, developmental psychologists have considered the origins and emergence of these notions (e.g., Astuti, Solomon & Carey, 2004; Hirschfeld & Gelman, 1994). The present paper is concerned primarily with the emergence of intuitive understandings of living things (folkbiology) in individuals raised in distinct communities in the US and ranging in age from four years to adulthood. Developmentalists have long been interested in children’s acquisition of knowledge about living things and the biological world. In his detailed observations of children’s reasoning, Piaget (1964) proposed that young children have a rather unstable notion of the concept alive, judging virtually anything that appears to move on its own (e.g., clouds, the wind, heavenly bodies) an animate being, and imbuing it with intentions and even beliefs and emotions. In Piaget’s view, this childhood animism is illogical and atheoretical, and ultimately must give way to the more mature, truly biological system of organization that characterizes adult thought. In 1985, Carey’s work revitalized interest in this issue, bringing new experimental evidence to bear in support of a new claim. Like Piaget, Carey argued that the concept alive undergoes substantial developmental change, that children’s understanding of this concept differs radically from that of adults, and that a truly biological concept of living things is not available until children reach roughly 9-10 years of age. Where Carey’s view parted company with Piaget’s was in her characterization of the young child’s mind. In her view, although young children’s thinking differs radically from that of the adult, it is neither illogical nor unstable. On the contrary, she proposed that young children's understanding of living things is initially embedded in a folk psychological, rather than a folk biological, explanatory framework. Two points follow from this assertion. First, Essentialism and cultural belief in folkbiology 4 that as a consequence of their folk psychological framework, young children view humans as the prototypical animal; and second, that children must undergo a fundamental conceptual change if they are to move from a folk psychological (in which humans are at the very center) to a folk biological framework, in which they come to view humans as one animal among many. Carey’s bold claims served to stimulate a large body of research. Currently, there appears to be consensus within this active field of research on three points: First, that young children do indeed have intuitive theories that guide their reasoning about biological phenomena; second, that these theories may indeed be distinct from their theories concerning psychological phenomena (for extensive reviews, see Carey, 1999; Gelman, 2003; and Inagaki & Hatano, 2002); and third, that these early theories differ systematically from both the more elaborate theories held by adults and the more formal theories that form the core of Western science curricula. But there is also ample ongoing debate, which has focused primarily on characterizing the scope of children’s intuitive biological theories, and this calls up the issue of domain-specificity. Some have argued that when reasoning about biological phenomena, children do not invoke domain-specific biological causal mechanisms until about seven years of age (see Carey, 1995; Carey & Spelke, 1994; Johnson & Carey, 1998; Johnson & Solomon, 1997; and Solomon, Johnson, Zaitchik & Carey, 1996). Others have interpreted the evidence differently, arguing that children’s reasoning about biological phenomena is guided from the start by skeletal principles or naïve theories that invoke domain-specific biological causal mechanisms (see Atran, 1987, 1990; Atran, Medin, Lynch, Vapnarsky, Ek & Sousa, 2001; Gelman & Wellman, 1991; Hatano & Inagaki, 1994; Keil, 1989; Sperber, Premack & Premack, 1995; Springer, 1995; and Springer, 1996). The concept of biological essentialism has figured largely in these recent discussions. Biological essentialism, put simply, is a commonsense assumption (a) that each and every biological entity has an underlying essence, or internal causal force, and (b) that this essence is responsible for maintaining the identity of that entity within its species, even as it causes the entity to grow, change form, and transmit features from one generation to the next (Ahn, Kalish, Gelman, Medin, Luhmann, Atran, Coley & Shafto, Essentialism and cultural belief in folkbiology 5 2001; Atran, 1998; Atran, Estin, Coley & Medin, 1997; Coley, Medin, & Atran, 1997; Gelman, 2003; Gelman & Wellman, 1991). Inherent in this notion, then, is a commitment to tracing the essence of unique individuals as well as the essence of kinds (or species) of individuals. The notion of biological essentialism involves causal mechanisms that dictate how and why biological (but not non-biological) phenomena unfold and is, in this sense, domain-specific. What is very interesting, especially from a developmental vantage point, is that this overarching “in principle” belief in an underlying biological essence has been shown to guide reasoning and learning even in children and adults who lack specific information about what that essence might be (Gelman, 2003; but see also Strevens, 2000). The idea is that, even when they are naïve with respect to what this essence might be, adults and children nonetheless believe that an (as yet unknown) underlying essence does exist, and that it is responsible for the identity of an individual within its kind and therefore responsible for the appearance, characteristic behaviors, and ecological preferences of individuals within the kind. This in principle belief, which has been described as an essence placeholder (Medin & Ortony, 1989), is perhaps best characterized as a skeletal principle, for it guides the acquisition of knowledge within the biological domain, but does not specify precisely the content of the knowledge (see Gelman & Williams, 1998 for a discussion of the role of skeletal principles in conceptual development). The idea of an essence placeholder fits well with the evidence that an overarching belief in biological essentialism may be universal, but the more precise intuitions about what that essence actually might be varies rather dramatically across cultural settings and historic time. To the best of our knowledge, a range of vital biological (e.g., blood, heart, liver, brain, and even mothers’ milk) and spiritual (e.g., the soul, karma, mind) elements have been considered, at one time or another, as harboring the essence of an individual (Stoler, 1997; Zimmer, 2004). The idea that biological essentialism is a skeletal principle also fits well with the observation that the precise range of entities that are considered to be biological also varies across cultural settings and historic time. For example, the Itza Maya appear to consider the sun and even the “forest spirits” to be biological entities. In other cultures, these entities are not conceptualized as biological and therefore are not expected to be governed by a Essentialism and cultural belief in folkbiology 6 biological essence. This point is important. Although there is developmental and crosscultural variation in the boundary conditions on the entities that are considered to be biological (and of these, on the entities that are judged to be alive) and in the precise candidate essences, an overarching belief in essentialism appears to be universal and to guide the reasoning of young children. Returning to the developmental issues at hand, one important question has been whether young children invoke specifically biological causal mechanisms in reasoning about entities in the biological world (see Gelman, 2003 for an excellent review; see Keil, 1989, and Inagaki and Hatano, 2002, for discussion of several candidate mechanisms). To get to the core of this issue, several experimental tasks have been designed, but one in particular – an adoption task – has been especially popular. This task focuses squarely on children’s developing intuitions regarding the inheritance of properties from one biological entity to another (see, e.g., Astuti, et al., 1994; Astuti, 2001; Atran et al., 2001; Bloch, Solomon & Carey, 2001; Johnson & Solomon, 1997; Solomon et al., 1996; Sousa, Atran & Medin, 2002; and Springer, 1996). Typically in this task, a participant is told a story about a baby who was separated at birth from its biological mother and then reared exclusively by an adoptive mother. The participant is then presented with a series of follow-up questions, designed specifically to assess their intuitions about whether the baby, once grown, will resemble its biological or adoptive parents in its physical and behavioral characteristics. For example, in a story involving a baby who was born to a pig, but then raised exclusively by cows, participants are asked to judge whether the baby, once grown, will exhibit a) the physical characteristics of the biological or adoptive parent (e.g., a short curly tail or a long straight tail, respectively) and b) the behavioral characteristics of the biological or adoptive parent (e.g., oink or moo). In addition, participants may be asked to make a judgment regarding kindhood (e.g., whether the baby will grow up to be a pig or a cow). Performance on this task has typically been interpreted as follows. If participants favor the biological over the adoptive parent in attributing characteristics to the offspring, this is taken as evidence of a specifically biological mechanism of transmission, favoring the influence of nature over nurture. If participants favor the adoptive over the biological parent, this is taken as evidence of the importance of the environment, favoring nurture Essentialism and cultural belief in folkbiology 7 over nature. Some researchers have gone further to argue that evidence for a truly domain-specific biological causal theory requires a differentiated pattern of responses, in which physical properties (which are presumably transmitted by biological mechanisms) are attributed to the biological parent, and behavioral or psychological properties (which are presumably transmitted by environmental mechanisms) are assigned to the adoptive parent (Astuti, et al., 2004; Bloch et al., 2001; Solomon et al., 1996). Although this line of argument is plausible, it is not unimpeachable. There are logical reasons to question the position that the differentiated pattern is a litmus test for a truly autonomous biological theory. Just as it is naïve to assume that development is guided by either nature or nurture, so it is naïve to argue that attributions regarding properties are guided either by biological or environmental mechanisms (see Sousa et al., 2002 for a similar argument). The two are not mutually exclusive. Some clearly physical properties (e.g., bone density, muscle strength, height, and weight) are influenced heavily by the environment as well as by biology; some behavioral properties (e.g., activity level and diurnal rhythms) are influenced heavily by biology as well as by environmental happenstance. Following this logic, if the expression of any given property may be under both biological and environmental control, a failure to distinguish between physical and behavioral properties in the inheritance of properties may, in fact, reflect participants’ appreciation of the necessary interaction between nature and nurture. In the face of these interpretive issues, what should count as evidence for an autonomous theory of biology, one that is organized around distinctly biological causal mechanisms? In our view, what is required in theory is evidence of causal reasoning that operates in a broadly coherent manner upon entities unique to the biological domain, and what is required in practice is that we move beyond the inheritance task, to examine in greater detail the causal mechanisms underlying the transmission of kindhood. This is precisely the strategy adopted in two recent research programs, and the strategy that we advance in the experiments reported here. Focusing on the Yukatek Maya, Atran et al. (2001) uncovered a strong biological component underlying the inheritance of properties and the judgments of kindhood. Both adults and children (ranging from 4 to 7 years of age) revealed a robust tendency to attribute properties of the birth parent to the offspring. This “birth parent bias” was observed for all properties, Essentialism and cultural belief in folkbiology 8 whether they were behavioral or physical, familiar or novel. In other words, the differentiated pattern was not observed in this population (even in adults). Crucially, this experiment went one step beyond examining intuitions regarding inheritance of properties. In addition, the experimenters asked participants to reason about the mechanisms responsible for the transmission of an individual’s kindhood. After each participant had answered questions regarding the inheritance of properties (described above), the experimenter presented one last scenario. This scenario was designed to tap into participants’ intuitions regarding the mechanism underlying the transmission of kindhood, and focused on the possibility that blood was essential to kindhood. Participants were told that when the baby was still young, it had become very sick and that to heal the baby, a doctor had to take out all of its blood and replace it with blood from the adoptive mother. Participants were then asked what kind of animal the baby would grow up to be. Adults and children alike continued to favor the birth parent, suggesting that in their view, an individual’s kindhood is transmitted by birth and is unlikely to be changed even by an intervention as radical as a blood transfusion from the adoptive parent. This suggests that for the Yukatek Maya, blood is not judged to harbor the “essence” of an individual. A subsequent study by Sousa et al. (2002), involving urban children and adults in Brasilia (Brazil), reported converging evidence for a strong biological component underlying property inheritance. Brazilian children and adults showed a strong birth bias for all properties, and this was especially pronounced for familiar as compared to novel properties. But when queried about the effects of the complete blood transfusion, there was an intriguing effect. Although Brazilian adults, like the Yukatek Maya, denied that the blood transfusion from the adoptive parent would influence the individual’s kindhood, Brazilian children performed quite differently. They judged that the blood transfusion could indeed change kindhood. Sousa et al. (2002) linked this finding to the considerable discourse concerning blood within Brazilian urban culture. They speculated that, as a consequence of this discourse, Brazilian children hear a great deal about blood and therefore may be more likely than children in other cultures to seize upon blood as a plausible candidate biological essence, as a mechanism that underlies the transmission of kindhood from one individual to another. Essentialism and cultural belief in folkbiology 9 In the present studies, we sought to pursue this very intriguing effect in several directions. A first goal was to broaden the empirical base and to this end, we recruited participants from four distinct communities. Two of these communities are situated in adjacent areas in rural Wisconsin, one European American and the other Native American (Menominee Tribe). Participants from these communities were included in Experiments 1 and 2. The other two communities are situated in adjacent towns in Illinois, one urban and the other suburban. Participants from these communities were included only in Experiment 2. We presented adults and children from each of these communities with three adoption scenarios, examining their judgments regarding the inheritance of physical and behavioral properties. Perhaps more importantly, we pursued questions concerning the mechanism underlying the transmission of kindhood. To address this question, we solicited participants’ intuitions regarding two potential candidates for essence: blood (Experiments 1 and 2) and nurture (Experiment 2). The Native American Menominee population was especially important in this enterprise, because it offered an opportunity to test Sousa et al.’s interpretation of the Brazilian data. Menominee tribal membership is based on blood quantum. Because both federal and tribal regulations depend on formal tribal enrollment, blood quantum measures have important practical consequences in the lives of individuals, families, and for the community at large. For example, hunting and fishing are frequent activities for Menominee adults and children, and the products of these activities are very visible within the context of family life. However, hunting and fishing regulations draw a threetiered distinction between enrolled Menominees, “descendents” (individuals who have at least one Menominee parent but who do not themselves meet the blood quantum requirements), and non-tribal residents (e.g. a non-Menominee spouse of an enrolled Menominee). Enrolled members are allowed to take more fish and game and enjoy a longer season than do their descendents, and descendants are permitted more fish and game and a longer season than their non-tribal counterparts. As another example, the Menominee tribal school receives financial aid from the Bureau of Indian Affairs, but this aid is based on the number of children who are enrolled Menominees; no funds are provided for descendents or non-tribal residents. As a result of this federal regulation, the Essentialism and cultural belief in folkbiology 10 parents of any child who is a descendent (or non-tribal counterpart) must pay tuition if they wish to send their child to the tribal school. In short, tribal membership status, which is based strictly on blood quantum, has very tangible and practical implications for Menominee families. Because tribal membership status has an impact on parents and children alike, there is considerable community-wide discourse concerning blood (quantum). This provides an intriguing research opportunity. If young children do indeed have an overarching in principle belief that there is an underlying biological essence that is responsible for kindhood, and if they are as yet naïve with respect to the precise content of that essence, then they should be especially sensitive to the discourse of the adults in their communities as they seek to identify plausible candidate essences. As a consequence, community-wide discourse about blood should elevate the plausibility of blood as a candidate essence for children. If this is the case, the Menominee children (like their Brazilian counterparts in Sousa et al., 2002) should be more likely than children from the non-Native US communities to judge that a complete blood transfusion is relevant to an individual’s kindhood. To anticipate, the results of the two experiments reported here suggest that across diverse cultural communities, (1) there is a strong biological component underlying people’s reasoning regarding the inheritance of traits, (2) there is a strong bias to attribute an individual’s kindhood membership on the basis of the kindhood of its biological parent, (3) this bias emerges in early childhood, and (4) it is accompanied by skeletal notions about causal biological mechanisms that underlie the transmission of kindhood from one individual to another.