The pretense debate☆
Introduction
Over the last 25 years, Alan Leslie and his collaborators have developed and defended a sophisticated and influential account of the psychological mechanisms underlying pretense, focusing primarily on pretense in children (German and Leslie, 2001, Leslie, 1987, Leslie, 1988, Leslie, 1994, Leslie, 2002, Leslie and Roth, 1993, Leslie and Thaiss, 1992, Onishi et al., 2007). In Mindreading, Nichols and Stich (2003) proposed an alternative account that borrowed a number of important ideas from Leslie, while arguing that one central feature of Leslie’s account is problematic and should be abandoned. Friedman and Leslie (2007) responded to Nichols and Stich (hereafter N&S), arguing that the N&S account of pretense is fatally flawed because it cannot account for the capacity to recognize pretense – a capacity which emerges quite early in childhood. In this paper, we have a pair of goals, one positive and one negative. The positive goal is to respond to the Friedman and Leslie (hereafter F&L) critique by showing how, with some elaboration and reconstruction, an account similar to the one proposed by N&S can address the pretense recognition problems posed by F&L. If our revised version of the N&S theory succeeds in meeting F&L’s objections, it might be thought that the debate has reached an impasse, at least for the moment, since there are two competing theories that can explain the available facts about pretense. However, our negative goal is to argue that this is not the right conclusion to draw, because the theory of pretense that Leslie and colleagues have developed and defended thus far is not a serious competitor to the theory we propose. More specifically, we will argue that the explanation of pretense that Leslie has proposed is importantly incomplete; it does not provide a satisfying explanation for some of the most obvious and important facts about the production and recognition of pretense.
Section snippets
Background: some shared assumptions about cognitive architecture
The N&S account of pretense is set out as a series of additions to a widely shared picture of the basic architecture of the cognitive mind. Though Leslie has explicitly embraced this picture only once, in a paper co-authored with N&S (Nichols, Stich, Leslie, & Klein, 1996), we think it is clear that Leslie, like many other theorists, assumes that this account of cognitive architecture is by and large correct, though far from complete.
Some points of agreement between Leslie’s theory of pretense and the N&S theory
Two examples of pretense recounted by Leslie, 1987, Leslie, 1994 have been widely discussed in the literature. In one of these, a child and her mother pretend that a banana is a telephone. The mother holds the banana up to her face and talks to Daddy. She then hands it to the child, who says hello to Daddy. In the other, a child and the experimenter have a pretend tea party during which imaginary tea is poured into a pair of cups, and then one cup is turned upside down and shaken. The child is
Some points of disagreement between Leslie’s theory of pretense and the N&S theory
By adding decoupled representations to his account, Leslie explains how pretenders avoid what he sometimes calls “reputational abuse” – the sorts of problems that would arise if pretenders treated representations subserving pretense in the same way that they treat the “primary representations” in their Belief Box. In addition to the quotation marks that serve to mark decoupled representations in his model, Leslie proposes to add “a second extension to the primary code” for the mental
Friedman & Leslie’s critique of behavioral theories of pretense
To explain their objections to N&S’s account of pretense, F&L consider a number of examples aimed at showing that the N&S behavioral account of pretense recognition leads to obviously mistaken predictions. In some cases the problem is that the N&S predictions would be far too broad. “The [N&S] theory,” they argue, “predicts that children will mistakenly deem non-pretend behaviors to be instances of pretense” (F&L, 2007, 111). In other cases, the problem is in the opposite direction – the N&S
The pretense game
According to Leslie and his collaborators, pretense is a mental state. We disagree. We think that pretense is best viewed as a sort of game – a game that requires no understanding of mental state concepts.
A critique of Friedman and Leslie’s explanation of pretense
If the explanation of pretense that we have offered in the previous section is plausible, then it might be thought that for the moment the debate has reached a standoff, since competing and significantly different explanations have been proposed – our “pretense game” account and the metarepresentational account offered by Leslie and colleagues – both of which claim to be able to account for the facts about pretense. However, we think that this is the wrong conclusion to draw. The debate is far
Objections, replies and clarifications
In the previous section we argued that, in its current state, Leslie’s metarepresentational theory of pretense is not a strong competitor for our pretense game theory because Leslie’s theory is gravely underspecified. Leslie and his colleagues have not told us how the theory explains some of the most obvious facts about pretense. They have not told us how possession of the PRETEND concept enables a child to understand what Mommy is doing in the banana-telephone pretense, or how it enables a
Conclusion
The main difference between Leslie’s metarepresentational theory of pretense and the theory offered by N&S is that on Leslie’s theory both the recognition and the production of pretense behavior require a child to form mental representations that include the mental state concept PRETEND. Thus “pretend play … [is] a primitive manifestation of the ability to conceptualize mental states” and “pretense is an early manifestation of … theory of mind” (1987, p. 424 & p. 416). On the N&S theory, by
References (29)
- et al.
The conceptual underpinnings of pretense: pretending is not “behaving as if”
Cognition
(2007) - et al.
Is young children’s recognition of pretense metarepresentational or merely behavioral? Evidence from 2- and 3-year-olds’ understanding of pretend sounds and speech
Cognition
(2010) Pretending and believing: issues in the theory of ToMM
Cognition
(1994)- et al.
Domain specificity in conceptual development: neuropsychological evidence from autism
Cognition
(1992) - et al.
15-month-old infants detect violations in pretend scenarios
Acta Psychologica
(2007) - et al.
Observers’ proficiency at identifying pretend acts based on different behavioral cues
Cognitive Development
(2004) Mindreaders: The cognitive basis of “theory of mind”
(2011)- et al.
Do humans have two systems to track beliefs and belief-like states?
Psychological Review
(2009) - et al.
False-belief understanding in infants
Trends in Cognitive Science
(2010) Mindreading in infancy
Mind and Language
(2013)
How do children represent pretend play
Children’s influences from “knowing” to “pretending” and “believing”
British Journal of Developmental Psychology
Child studies through fantasy
Pretense, imagination and belief: the single attitude theory
Philosophical Studies
Cited by (15)
Understanding pretense as causal inference
2023, Developmental ReviewOpaque windows in puppet studies: A response to Lillard (2022), Rakoczy (2022), Wellman & Yu (2022), and Yu & Wellman (2022)
2023, Cognitive DevelopmentCitation Excerpt :But it is only one among many theories of pretense. And it should be noted that this “cognitive theory” was criticized for its failure to explain how a child can recognize that other people are pretending (Friedman & Leslie, 2007) and thereafter replaced by a “pretense game” theory in which “pretending is an activity that is usefully viewed as a kind of game” (Stich & Tarzia, 2015, p. 9). As a contribution to the current discussion the boxological theory appears to attribute to young children precisely the ability to reason in the same way about pretend and real entities that we are questioning.
We don't play that way, we play this way: Functional Play Behaviours of Children with Autism and Severe Learning Difficulties
2020, Research in Developmental DisabilitiesCitation Excerpt :In a systematic review of the literature on play, Barton (2010) commented that the definitions regarding functional play are diverse, suggesting that at times functional play has been considered under the heading of pretend play and at other times as a separate category. This is possibly the result of an ongoing meta-representational debate regarding the thoughts the child holds during the play actions they complete (Jarrold, Mansergh, & Whiting, 2010; Leslie, 1987; Meini & Voltolini, 2010; Stich & Tarzia, 2015). Within the current study there is recognition that some meta-representation may be necessary to engage in some higher characteristics of functional play.
There's more to consider than knowledge and belief
2021, Behavioral and Brain SciencesActing and pretending
2024, Theoria (Sweden)
- ☆
We are indebted to Ori Friedman, Shaun Nichols and to two anonymous referees for their insightful comments on an earlier draft of this paper.