A Huli aphorism says, Real don't make children's talk, or the games agoba kiruba [which hand is it in?] or mbola tola [exploding mud balls]. In much the same way that Huli men don't like children, in anthropology a view has prevailed that real anthropologists don't study children's (Norbeck 1974; Schwartzman 1976; New 1994). Children's voices have by and large received only perfunctory consideration in ethnographies and the profile of child as a topic within mainstream anthropology is rightly characterized as minuscule (Chick and Donlon 1992:236). Indeed, monographs on children's remain extremely scarce. Moreover, in respect to that unique genre of variously referred to as makebelieve, sociodramatic, symbolic, fantasy, or representative (cf. Fein 1981; Singer and Singer 1977; Singer 1973; Feitelson 1977), systematic theorizing or crosscultural research is nearly invisible. There appear to be several reasons for this persistent neglect of children's pretend play. Investigating naturally occurring sociodramatic is notoriously problematic as episodes are fleeting, the actors highly mobile, and the physical presence of an adult observer unacceptably intrusive. In this regard, there are few programmatic guidelines that might assist endeavors. In part, too, how research intercalates with other levels of data so as to constitute a theory-building part of the discipline remains an enigma. Analysts have largely been content to subsume their interests in child within the broader rubric of socialization-what functions serves in producing competent adult communicators-rather than generate a set of issues or problems directed specifically at for play's sake. Perhaps, also, anthropologists have been too ready to defer to the vast developmentally oriented literature in psychology. Such studies have focused on issues related to cognitive, affective, and symbolic competencies (Werner and Kaplan 1963; Golomb 1977), personality factors as variability in individual levels of predisposition (Fein 1981; Singer 1973), the evolution of language styles (Dixon and Shore 1993; McCune-Nicolich 1981; Rubin and Wolf 1979; Garvey and Kramer 1989), and sequential trajectories in forms from early sensorimotor and sociodramatic to games with rules (Piaget 1962; Robinson and Jackson 1993). Too often, however, discussion is immured within the context of data provided by Western, Englishspeaking children subject to laboratory conditions where behavior has been variously facilitated, modeled, or suggested (McCune-Nicolich and Fenson 1984). The few anecdotal accounts there are of make-believe within the crosscultural literature have thus assumed an importance for theorists that is, to be blunt, quite out of proportion to the levels of ethnographic detail or reliability offered. Anthropological interest here has been less one of addressing issues as model of mind can best account for the development of pretend (Goncu 1989:341) than in assessing (1) what is the universal status of children's fantasy play, and what saliently similar patterns does interaction evidence? and (2) are subsistent cultural variations in behavior systematically linked to or determined by identifiable cultural practices or institutions? There are no clear answers to these questions and much disagreement exists about the validity of available findings. Analysts who indicate the low or nonexistent status of representational in many cultures (e.g., Feitelson 1977; Golomb 1977; Fein 1975; Pan 1994; Sutton-Smith 1972) - what Schwartzman (1984:49) calls the deficit hypothesis - repeatedly invoke the ethnographic cases of the Nyansongo, a Gusii community in Kenya Levine and Levine 1963:173); Manus children, whose play as the most matter of fact, rough and tumble, non-imaginative activity imaginable (Mead 1930:96); so-called disadvantaged Middle Eastern and North African immigrant children (Smilansky 1968); and positively dangerous generalizations that such is the way the New Guinean plays . …