We are great pretenders. We start early, and we pretend in a variety of basic and more advanced ways for many different purposes.

When pretending, we act or treat something or someone as if it has a feature or features other than those that the said something or someone is taken to possess. Although it remains an open question whether other animal species engage in pretense, it is well documented that humans typically begin to engage in and can recognize acts of pretend play from a very young age. The developmental psychological literature distinguishes between several distinct forms of pretending in which children engage and their ways of recognizing acts of pretense as these capacities appear in different stages of typical child development within certain populations (see Lillard et al., 2014., Liao & Gendler 2020 for a recent summary of these findings).

Pretending isn’t just child’s play – it also features in many everyday adult human activities. People pretend when they act as if they themselves are something they know or believe they are not, or have features they know or believe they do not have – such as acting as if they are a wizard, or as if they are reliably honest. We pretend when we enact certain roles for various purposes. We pretend when we adopt certain personas in our efforts to deceive, trick, and impress others. We also do so, without deceit, when we adopt certain personas in order to entertain others, as actors do when delivering theatrical performances and as narrators do when embellishing and enriching their storytelling by acting out the parts of various characters. And we do something similar in our self-directed acts of pretense. For example, without deceit, we might knowingly pretend to be someone other than who we are in order to effect a change in ourselves or to encourage or push ourselves to achieve certain outcomes.Footnote 1 Suffice it to say, there is a wide spectrum of pretend practices and pretense comes in many forms and at various levels of sophistication.

How can we best understand pretense in all its variety? This paper makes the case that, when compared to its current rivals, radical enactivism, or RE, offers the required nuance needed for understanding of the full array of the many, various forms of pretense. In particular, its multi-storey account of cognition, which holds that psychological attitudes can be both contentless and contentful, enables it to appropriately account for both the most basic and most advanced forms of pretending. By comparison, other accounts of pretense, which are promising in key respects but offer more uniform accounts of pretense, are condemned to fail in this regard.

The action of the paper unfolds as follows. Section one briefly reviews the current state-of-the-art in theorizing about pretense, revealing that there are reasons to steer clear of representationalist theories of pretense for two main reasons. Firstly, said theories arguably over-intellectualize the phenomena to be explained. Secondly, said theories have yet to adequately address serious problems associated with their foundational explanans – the positing of mental representations.

Section two then provides an overview of RE’s account of pretense – an account which assumes our acts of pretense are out-in-the-world, embodied activities and not in-the-head mental phenomena. Drawing inspiration from Wittgenstein, RE focuses on the social-cultural practices that surround and give life to acts of pretense and attends to the nuances and differences in the varieties of pretense instead of attempting to find the essence that encompasses every form of pretense.

Section three compares RE’s proposal with that of a new contender which has just arrived on the scene – praxeological enactivism, or PE. PE agrees with RE on many points, however it is argued that PE goes wrong to the extent that it embraces a brand of idealism and to the extent that it assumes that even the most basic varieties of pretending take the form of interpretive sense-making.

Finally, section four responds to Langland-Hassan’s (2021) challenge, which holds that any enactivist account which does not acknowledge the contentful attitudes of pretenders cannot provide an adequate account of pretense. It is shown that RE can meet this challenge, but that it does so by rejecting certain internalist and essentialist commitments. With respect the latter, some counterexamples are provided to challenge Langland-Hassan’s own mentalistic recipe.

It is concluded that RE is well placed, by comparison with other existing accounts, to provide crucial resources for understanding the full range and variety of pretense. In particular, RE avoids the pitfalls of representationalist theories of pretense while also providing a means to combine the best elements of PE and Langland-Hassan’s proposals, without taking on board their shortcomings.

1 Challenging Representationalist Assumptions

Philosophers and psychologists alike have devoted considerable attention to trying to understand pretense in recent decades (Leslie, 1987, 1994; Lillard, 1993, 1994; 1998; Nichols & Stich, 2000, 2003; Carruthers, 2002; Gómez, 2008; Currie & Ravenscroft 2012; Currie 2006; van Leeuwen 2011; Ma & Lillard, 2013; Mitchell 2002). These theorists seek to understand: What is it to pretend? How do we recognize acts of pretense? What enables us to carry off acts of pretense? Despite concerted efforts, a received account of pretense has yet to emerge. As to be expected, different theorists disagree with rivals on the fine details of their specific proposals.Footnote 2

There is one main issue upon which most of today’s prominent theories of pretense consistently agree – namely, that, one way or another, pretending always involves mentally representing of some sort or other.Footnote 3 Lillard (1996) epitomizes and makes explicit the standard line of reasoning that motivates widespread acceptance of the assumption that mental representations of some sort are necessary for pretense.

if one is pretending to be a kangaroo, one mentally represents oneself as a kangaroo inside one’s head and often also makes sequelae of that representation externally perceptible, for example, by hopping. Whereas making the sequelae of the representation perceptible is optional (one might be a kangaroo that is sitting still, not in any way evoking its “kangarooness”), the mental representation is crucial. To hop without it is not to pretend to be a kangaroo, no matter how kangaroo-like one’s hopping is. (Lillard, 1996, p. 1718 emphasis added. For a more recent statement along similar lines see Ma & Lillard 2013).

Notably, Lillard (1996) starts from the assumption that mentally representing is necessary and sufficient for pretense, whereas any outward manifestations of pretending behaviour sponsored by the posited mental representations are entirely optional. To think of pretense as constituted by individualistic, mentalistic phenomena is to frame our explanatory needs in a very specific way. If we start by defining pretense as constituted by mental representations then, in a quite self-fulfilling manner, it will seem logically inevitable that only a certain type of explanans will do: only explanations positing in-the-head mental representations of some sort could possibly answer our needs when it comes to explaining pretense.Footnote 4

Why think that acts of pretense must always and everywhere involve mental representations? This conclusion is, of course, a no-brainer if it is assumed that, in general, cognition – even of the most basic variety – consists of operations involving content-bearing, truth-evaluable mental representations.Footnote 5

If we start with the notion that our basic forms of engaging with world offerings necessarily involves representing those offerings in contentful, truth evaluable ways then it is easy to understand why such great weight is placed on the fact that in young children seem able to imagine “what is not the case” in pretend scenarios. It would seem that the child must know the truth, using one set of mental representations, and yet, simultaneously, be able to represent the very same set of facts as being otherwise, using a distinct mental representation or set of mental representations. Moreover, they must have a means of keeping track of these two sets of representations without confusing their reality – always being able to tell apart the reality from the fiction.

To do this, it is standardly thought that pretenders must somehow keep track of which of their mental representations and attitudes are about the actual world and which are about imaginary things that exist only in their pretend worlds. Without some means of doing so, the thought goes, the pretender may mix them up to disastrous effect.

How is such confusion avoided? Those who adhere to representationalist starting assumptions tend to posit extravagant sub-personal mental machinery, Possible World Boxes and their ilk, to do this specialized quarantining work (see, e.g., Leslie 1987, Nichols & Stich, 2003).

It has been claimed that the mechanisms that enable pretense may be the true source of our metarepresentational abilities—and that even the simplest acts of pretense involve “an early manifestation of the ability to understand mental states [and hence represent] the specific innate basis of our commonsense theory of mind” (Leslie, 1987, 412). Why think this? As Semeijn (2019) points out, pretend play, “typically requires children to anticipate behaviour of playmates that is … based on false (pretend-)beliefs” (p. 114). If we assume all of this, then young pretenders must be capable of representing what another represents to be the case and must be able recognize what is so represented as a representation. Having such capacities is necessary in order for the child to know that the other does not take what they represent to be true. On the basis of such reasoning, some theorists conclude that “pretending necessarily requires a theory of mind” (Baron-Cohen, 1999, 265).

This is all pretty demanding stuff for children of just 2-years of age. In order to pretend, such children would need to be relatively cognitively sophisticated, and knowledgeable about the nature of pretense itself and what it involves. In this vein, some theorists go further still, adding a hyper-intellectualist assumption into the story – they assume that the child must operate with the concept of pretense in order to pretend (Leslie, 1994).

The need for explanations of these intellectualist and hyper-intellectualist sorts follows swiftly for anyone who makes the fundamental assumption that cognition is, even at its roots, always and everywhere, contentfully representational. For those who assume this, it is inevitable that the target explanandum of pretense is characterised in intellectualized terms – and that characterisation sets up a certain understanding of what is required for the explanation of pretense. Against this backdrop, positing mental representations will seem unavoidable in any adequate explanation of pretense.

But to insist that mental representations are logically necessary for pretence is nothing but a philosophical-snake-oil-operation. We are entirely free to drop the starting assumption that pretense is constituted by and can thus only be explained by positing mental representations if we drop the assumption that cognition, always and everywhere, involves content-bearing mental representations. Let’s call said assumption the Cognition always Involves Content, CIC, assumption. There are good theoretically-based reasons for dropping the CIC assumption – especially when it comes to thinking about the most basic forms of cognition. Invoking the Hard Problem of Content, Hutto & Myin (2013) and (2017) supply detailed, sustained arguments, that put pressure directly on the CIC assumption that the most basic forms of cognition are contentful.

And here’s the rub: if we characterize pretense differently from the start, then different accounts of what is needed to explain it, those proposing de-intellectualized explanans, become credible. The key move is to avoid over-intellectualizing the phenomena to be explained in the first place. For example, there is no need to explain how pretenders manage to keep track contentful mental representations of what they know to be the case. Nor is there a need for purported explanations as to how they keep these apart from what they are only pretending to be the case if we do not start by assuming that our basic cognitive grip on the world always and everywhere involves truth-bearing contentful mental representations.

If we drop the CIC assumption, we might, for example, characterize pretense in wholly dispositional terms. If so, to borrow one of Langland-Hassan’s (2020) examples, we can understand what goes on when Sally pretends to be a lion in terms of her manifesting “two different sets of dispositions: dispositions to act like a five-year-old girl (ascribed, in part, by attributing her a belief that she is a girl) and dispositions to act somewhat lion-like (ascribed by saying that she is imagining that she is a lion). Having the latter set of dispositions does not amount to being disposed to act exactly like a lion—only to mirror some salient qualities of lions in one’s behavior … There is no conflict in Sally’s having both sets of dispositions simultaneously” (Langland Hassan 2020, p. 164).Footnote 6

In light of empirical findings, there is great advantage in taking this de-intellectualized line on early forms of pretense. It allows us to dissolve the so-called developmental puzzle of pretense. That puzzle of pretence arises because it is empirically well-established that children as young as 2-years-old can reliably recognize acts of pretense and distinguish them from sincere actions. However, going by standard verbal false-belief findings, 2-year-old children are not yet able to attribute mature mental state concepts and contents, certainly not explicitly and verbally (See, e.g., Mitchell & Neal 2005, p. 175). To assume that they can do so implicitly and non-verbally, at their young age, when pretending, makes them appear anomalously precocious given the other empirical findings about what children that age are able to do. If we assume children as young as two are capable of making meta-representational attributions when pretending, then why are they seemingly unable to do so in other settings?

In a recent analysis of the state-of-the-art, Semeijn (2019) reminds us that “Leslie’s theory cannot, without additional assumptions or qualifications, explain the early emergence of pretend play relative to the ability to pass explicit false belief tests” (p. 125). Leslie and others have sought to explain why this might be so. However, their proposals are not convincing, and they look ad hoc. For example, Leslie posits an additional cognitive mechanism to account for the seeming gap between cognitive capacity and performance, yet as Semeijn (2019) notes, so far “there is no direct evidence for this additional cognitive mechanism” (p. 125).

Motivated by these failures to accommodate the empirical findings, Langland-Hassan (2020) proposes that young children might be getting by using quite different means when engaging in pretense. For example, he holds that rather than deploying sophisticated theory-of-mind machinery to attribute false beliefs to others they may instead be distinguishing acts of pretense, in large part, by noticing signature cues from others. This sort of account could be augmented by assuming that young children are capable of mind minding – in ways that allow them to be sensitive to and to keep track of the contentful attitudes of others without thereby having to ascribe contentful attitudes, as such, to said others (see Hutto 2017a). Moreover, young pretenders could be navigating their initial games of pretending merely by having some grip on the basic rules of games of pretense rather than making contentful attributions about the mental states of others.

Accordingly, following this de-intellectualized construal of what pretense involves, we need only assume that children can identify games of pretense and enter into and master them “in the same way the child learns the rules to any game—such as kickball, or freeze tag” (Langland-Hassan, 2020, p. 159, see also Weichold & Rucińska 2021). Crucially, as Langland-Hassan observes this, “does not require the representation of another’s mental states” (p. 159).

And we can de-intellectualise our account of what pretense involves further still. For it need not be assumed to rest on the prior possession and use of a concept of pretense. It is surely enough to say that children are beginning to get a grip on the concept of pretense once they are able to reliably detect and start to play games of pretense.

Langland-Hassan (2020) proposes that “In one sense, we certainly have ascribed the child the concept pretend, to the extent that being able to detect and play such games constitutes understanding pretense. In this (behavioral) sense of ‘pretend,’ the child fully understands that the parent is pretending—and indeed that the parent is acting as if pretending” (p. 160).

Yet, in sticking to a de-intellectualized characterisation of pretense, it might be questioned whether having only a basic capacity to recognise and enter into games of pretense, impressive as it is, suffices for having ‘a’, let alone ‘the’, concept of pretense. We might reasonably demand more from anyone who can be said to have command of said concept (see Hutto 2008, p. 131 for a general proposal along these lines). For example, we may reasonably expect the person will be able to make reliable inferences about pretending, or that they can explain what pretending involves, or that they can answer questions about it. Surely, even if we allow that 2-year-olds are beginning to get a first, tentative grip on the concept of pretense, we ought still to demand much more from them before attributing to them anything like a full understanding of pretense.

Putting all of this together, we have good reason to be open to accounts of pretense that assume, firstly, that pretense is a kind of practice-constrained, embodied activity and not a wholly and solely in-the-head mental representational phenomena, and, secondly, that when it comes to explaining the special kind of cognitive activity involved in pretense we should try to avoid theories that assume pretense requires having and ascribing contentful mental representations from the get go.Footnote 7

2 Radical Enactivism, the Primacy of Practice, and the Varieties of Pretense

It would seem that if we are to do justice to the phenomena of pretense, in all its variety, we need to come at it in a completely different way. RE proposes thinking of pretense as a kind of embodied activity rather than as a state, or states, of mind. Moreover, it assumes that pretense is a practice-constrained embodied activity that, first and foremost, depends upon – and needs to be understood against the backdrop of – the particular socio-cultural practices that give it life. In short, when it comes to thinking about pretense, RE’s proposes giving pride of place to practices: this defines its preferred approach to this phenomena.

Accordingly, it is assumed that pretense is not an off-the-shelf basic inherited mental state, or psychological capacity, or set of such capacities. Thus, even though being able to pretend depends on one’s being able to exercise certain basic cognitive capacities, when pretending these must be exercised in specific sorts of ways, in tune with the forms and norms of the specific socio-cultural practice in which one is engaged.Footnote 8

In embracing the primacy of practices, RE makes a strong assumption about the origins of pretense: it holds that before children can pretend, they must be first introduced to, and to some extent master, a special kind of socio-cultural practice. On this view, a child’s initial ability to pretend therefore does not depend solely and primarily on their individual and basic imaginative capacities but rather on the way those capacities are put to use when they are “being taught by adults and by imitating the pretense acts of adults” (Rakoczy, 2006, p. 115). RE maintains that the mastery of practices of pretending proceeds individual pretending, not the other way around.

That said, one can accept the primacy of practices thesis without going so far as to maintain that “pretense is an essentially social phenomenon” (Rakoczy, 2006, p. 113), or that sociality is “constitutive of pretense” (Rucińska, 2019 p. 2, emphasis added). A more modest assumption suffices: one need only accept that “early pretend play arises through cultural learning in social contexts in a similar fashion as other action forms do” (Rakoczy, 2006, p. 113).Footnote 9

The signature idea of the primacy of practices approach that RE recommends is that practices of pretense should be recognized as occupying the chief place in our analyses of various forms of pretense. Practices form part of the explanans and the explanandum of pretense. Accordingly, we should focus on features of specific kinds of pretending practices before we attempt to understand or characterize any signature cognitive capacities required for such pretending and the familiar psychological profiles that are typically associated with it, and which may needed to fully explain how such acts of pretence are carried off in particular cases. The suggestion is that when it comes to clarifying the nature of pretense we should focus first on features of various practices of pretense – including games, deceptive activities, theatrical performances – and not primarily, and certainly not exclusively, on the psychological mindsets of individuals.

To conceive of acts of pretense as inherently tied to the emergence of certain kinds of socio-cultural practices should make us immediately sceptical of the idea of finding the essence of pretense – or in assuming the pretense has an essence that in any way resembles that alleged of natural kinds proper. We are likely to find a range of cases that lack a common core of the sort that can be captured by any hard-and-fast recipe of necessary and sufficient conditions defining what counts as pretense in every possible instance.

In this respect, RE’s approach differs from that of Langland-Hassan’s (2020, 2021), who seeks to capture the essence of pretense in a reductive recipe. On Langland-Hassan’s analysis, an episode of activity will count as pretending if and only if the right individual mindset is present. By his lights, the true hallmark of pretense is the presence of a certain combination of intentions and specific beliefs about what one is doing such that S must intend to make some x y-like, while believing that x will not, in the process, be made into a y.

The essential ingredients for pretense, according to this proposal, are a combination of intention and belief – mixed together as per the above formula. It is that particular mindset, made up of standard mentalistic components and nothing else, that distinguishes genuine cases of pretending from those in which one accidentally or unintentionally acts as if p.Footnote 10 If this analysis of pretense is right, then whenever one finds pretense one always find this particular arrangement of mentalistic attitudes.

Taking a leaf out of Wittgenstein’s book, RE sets out with descriptive and clarifying naturalistic ambitions rather than seeking to provide a metaphysically robust, reductive analysis of pretense (see Hutto 2013, Hutto & Satne, 2018a; 2018b). Carrying through on such ambitions requires looking closely at the many and various forms of pretense and the many different patterns and ways they figure and manifest themselves in our lives and practices. The art of being duly attentive to such patterns requires not imposing a priori theoretical assumptions and constructions on what one finds, or in assuming that all forms of pretense operate in the same way or share the same basic features (See Moyal-Sharrock 2021, esp. Chapter 1 for a fuller discussion). If we are to get real about pretense, we should heed Wittgenstein’s advice: “Don’t think, but look!” (PI 66).

When we look at pretence, we find sameness with variation. Thus, as Weichold & Rucińska (2021) observe, in order to engage in certain sophisticated forms of pretense in may be necessary to explicitly think of oneself as pretending, and to engage in the pretense with explicit intention or purpose, say when one is deliberately deceiving someone or to performing a particular part in play – but this need not be true of all cases. Consider the pretend play of young children. Some might participate in a game of pretence and act accordingly with deliberate intentions to do so, but there will also be cases in which children spontaneously respond to afforded opportunities and begin to pretend without any such prior intentions.

Borrowing an example from Langland-Hassan (2020), consider a child – let’s call her Kirstin – who, in the middle of a pretend tea party game, checks under her bandaged finger to examine a cut. Let us assume that Kirstin’s checking on her cut is not an action that she undertakes with the intention or for the purpose of playing the tea party game. Still, it might get incorporated into the game all the same. Imagine that another child in the game, Anthony, sees Kirstin’s action and incorporates it into their game. Let’s suppose that Anthony picks up on Kirstin’s action and runs with it. What if Anthony proposes that Kirstin’s checking under her bandage constitutes her summoning of genie to join their tea party. Suppose Kirstin, who hadn’t thought of this before, goes along with this proposal. It seems that her checking under her bandage, which was not done for the purpose of playing this game, has now changed status and become part of the tea party game.Footnote 11

Here it is noteworthy that Wittgenstein uses pretence as a key example of something that can only be understood by attending to the pattern it weaves in our lives. When we look at a particular case of pretence it reveals itself to have, not a single common core or essence shared with all other such cases, but certain overlapping similarities with other cases of pretence even if it has special features of its own. He tells us, “For pretence is a (certain) pattern within the weave of life. It is repeated in an infinite number of variations.” (LW I, 862). Or again, “Seeing life as a weave, this pattern (pretence, say) is not always complete and is varied in a multiplicity of ways. But we, in our conceptual world, keep on seeing the same, recurring with variations. That is how our concepts take it. For concepts are not for use on a single occasion” (Z 568).

The advantage of focusing on the differences as well as what remains the same in our practices of pretense is that we are able to do justice to what various kinds of pretence involve without assuming that there is one common mindset or aspect that informs or applies to all instances of pretending.

Insofar as we are able to identify anything that is common to all acts of pretence, we might, at most, say that they are all a matter exercising embodied capacities to act in certain ways within some constrained practice that we count as pretending. But to say this is at best only to gesture at the sort of thing pretence is as we find it in our lives rather than to try to offer a precise analytic recipe of the essential ingredients that we find in every instance of pretending.

3 Praxeological Enactivism, and It’s Limitations

How does RE’s assumptions and approach compare with those of Praxeological Enactivism, or PE, as advanced by Weichold & Rucińska (2021)? On the face of it, RE and PE have some core similarities. PE, as the name indicates, seeks to connect enactivism to practice theory – a name for a set of ideas associated with thinkers like Wittgenstein, Heidegger, Bourdieu and others.Footnote 12

PE also strongly follows RE in seeking to reconceive pretense by challenging familiar cognitivist assumptions that characterise pretense as a mental representational phenomenon in the first place. Like RE, PE understands pretence as manifesting in various forms of embodied activity. Finally, PE also follows RE in thinking that to such understand embodied activities we need to give attention to the particularities of the surrounding practices that give them life and purpose.

Where RE and PE part company is with respect to PE’s proposal that “all phenomena of pretense consist of in alternative sense-making” (Weichold & Rucińska, 2021, emphasis added). In line with this proposal, PE offers us the following formula: “To pretend means to make sense of one’s surroundings in an alternative way, crucially varying from how the world is perceived in the context of ordinary practices” (Weichold & Rucińska, 2021, emphasis added). PE thus makes an essentializing move that RE rejects for a number of reasons.

PE defines enaction as “the constant, interactive process through which an organism engages in sense-making, or interprets its surroundings as a meaningful world of relevance” (Weichold & Rucińska, 2021). It holds, moreover that an “enacted dependent world of relevance exists only for [a given] specific living body with its unique way of interpreting” (Weichold & Rucińska, 2021, emphasis added).

Given these commitments, PE appears to be a paradigm case of a sort of enactivism that risks “falling prey to a revamped version of idealism” (Rolla & Figueiredo, 2021).Footnote 13 In particular, in claiming the worlds brought forth are unique to, and only exist for, individuals due to their unique history of interactions, PE seems to embrace a solipsistic version of mind-dependent idealism that has decidedly “anti-realist undertones” (Rolla & Figueiredo, 2021). This is because it focuses too strongly on the organism side of the equation when it comes to making sense of the idea that organisms ‘bring forth a world” (Rolla & Figueiredo, 2021).

Taken at its word, on this score, PE’s position proves awkward when it comes to making sense of its own proposed account of pretense. To see this, it is worth examining the discussion of the example of a young girl pretending to be a lion, provided by the defenders of PE. 

According to our suggestion, she is enacting a certain narrative or a public script for lion play: lions find caves to hide into, or rocks to climb upon (see also Rucińska 2016, 2019). Her play is spontaneously guided by the affordances of her environment as she is making sense of it while at play: the chair affords climbing on to, and therefore, “climbing a rock” play, while the table affords hiding under, therefore, “hiding in the cave” play, etc. If the girl’s mother interrupts the play because the dinner is ready, the girl can switch from play context to non-play context smoothly: the chair turns back into something to sit on, while the dinner table is again the place to eat dinner (Weichold & Rucińska, 2021).

It is easy enough to make sense of the child’s ability to take advantage of the environmental affordances in different ways and to respond to them in quite different ways in reply to changes of context. This much of the PE proposal is in good order. Problems arise, however, when advocates of PE try to make sense of the metaphysics implied by this account. They tell us, for example, that “it is not the case that the kitchen table is always “really” or “truly” a kitchen table” (Weichold & Rucińska, 2021).

According to PE, the child makes sense of the kitchen table in one way when participating in the practice of having dinner with her family, but she makes sense of it in a different way when using it as a cave in her games of pretense. The trouble is that these different ways of making sense of the table have direct implications for its metaphysical status according to PE. Thus, we are told that there “is no independent reality that needs to be represented in the mind … in order to engage in pretense; we propose that we enact our realities instead” (Weichold & Rucińska, 2021). Taken one way, this is just a reminder that mental representations are not required for pretense. Yet taken together with the above claim about shifting realities, PE, as formulated, appears to imply that a kitchen table literally becomes something else when we enact our realities differently, and what it becomes depends directly on the particularities of the sense making in question.

RE adopts a quite different stance on this issue. According to RE, the table is there throughout – whether the child is enjoying dinner with her family or playing lions. Thus, should the child or onlookers claim that Sally is using the table as if it were a cave in her pretense they will be speaking truly.Footnote 14 Moreover, there are many other true things that we can continue to say about the table in question. Crucially, the table does not vanish from existence when a child pretends that it is a lion’s den. It is there, independently of us, even though it would not have existed at all had we, to use the collective ‘we’, not engaged in other practices that enabled us to create and manufacture such artefacts.

When it comes to understanding the relevant possibilities talk of ‘enacting worlds’ and ‘sense-making’ can mislead. Acts of pretending alone do not have the power to make things or unmake things. Pretending that a table is a battleship neither brings a battleship into being nor removes a table from being.

According to RE, the table has the same sort of status as the duck-rabbit drawing and its variants that Jastrow and Wittgenstein made famous. The original drawing appeared in a German magazine 1892 – it, and the versions that followed, are human artefacts. They exist independently of us, even though when seeing one of these figures under a given aspect we cannot describe or characterise the figure neutrally to the extent that we are trying to capture what is characteristic of seeing it under said aspect. By the same token, RE holds that Sally is engaging with the table and experiencing its aspects differently when she treats it as a lion’s den for the purpose of her game and engages in embodied activities and practices which suit that sort of engagement.

Considering their position on color perception highlights where RE and PE differ on this matter. Weichold & Rucińska (2021) tell us that “Humans do not experience light of certain wavelengths, but colors.” RE, by contrast, holds that we do experience light of certain wavelengths, but that we experience that light by experiencing colors. It is precisely for this reason that we can learn more about what we experience and why there are many ways that we experience and investigate the light of certain wavelengths through a variety of means.

Prima facie, it seems that PE is committed to saying that color perception “is a product of the organism’s mind. Organisms would bring forth an internal world” (Rolla & Figueiredo, 2021). Yet, as Rolla & Figueiredo (2021) observe, “to the extent that we understand enaction as the bringing forth of a world internally, it doesn’t seem like [enactivism’s promised] middle way between realism and idealism … is tenable”.Footnote 15

RE’s position on the idealism-realism question is substantially in tune with what Rolla & Figueiredo (2021) propose. Drawing on the notions of niche construction and social niche construction, they argue that the metaphor of ‘bringing forth a world’ be understood in terms of the selection and refinement of patterns of sensorimotor engagements through which we are able to interact with aspects of our surroundings and reliably engage with its environmental structures. To understand ‘bringing forth a world’ in this concrete sense is to think of enaction as:

the coupling between organism and its environment which leads to the organism physically altering environmental structures in order to make them advantageous for its survival. In some cases, ours for instance, those processes may lead to the construction of a social world too, that is, an environment characterized by sociocultural practices and institutions. We claim that enactivism is compatible with the idea of an independent reality of physical structures and relations without implying a pregiven world [in the sense that the world does not come already, somehow, pre-specified].Footnote 16

Finally, there is something else of concern about which RE and PE differ. The defenders of PE hold that “what humans experience and act on are always worlds that are understood by them, that are meaningful to them” (Weichold & Rucińska, 2021). In line with this, Weichold & Rucińska (2021) tell us that in both pretense and non-pretense cases, “humans bring forth, or enact, their surroundings as understood worlds” (emphasis added), and that “whether they are engaged in pretense or not, human beings are always acting on a world which is already understood by them in a certain way” (Weichold & Rucińska, 2021). In this vein, PE conceives of enaction as the process of our “subjective understanding of the world, or sensemaking” (Weichold & Rucińska, 2021, emphasis added).

Yet, as argued in the previous section, it seems quite possible for someone to engage in acts of pretending using the most minimal script-like knowledge and without any need for interpretation or understanding as these terms are standardly understood. Certainly, the advocates of PE demonstrate, through their own examples, that young children neither need to know nor understand much about the subjects matters that are the focus of their pretense in order to join in games of pretense and get them going. To say that all pretenders must ‘understand’ what they are doing seems to mischaracterise what is involved in certain acts of pretense – and to do so for the sake of keeping faith with a global, antecedent theoretical commitment.Footnote 17 Ergo, it looks as if PE is violating Wittgenstein’s injunction to ‘look, don’t’ think’ in its treatment of such cases.

4 Dealing with Langland-Hassan’s Mentalistic Challenge

For all that has been said so far, it might be thought that enactivist accounts of pretence have a crippling limitation. After all, Langland-Hassan (2021) argues that only accounts of pretense that are sufficiently mentalistic will be capable of accomodating bona fide case of pretense (Cf. Langland-Hassan 2012, 2014, 2015).

In order to address this charge, it is important to understand the true force of this challenge. As we have seen, enactivists characterize pretense as a special kind of embodied activity. But what kind of activity, exactly? Pretending is not just any old activity. Borrowing one of Langland-Hassan’s (2020) examples, hobbling about one leg after having stubbed one’s toe is an embodied activity. We can imagine Henry doing this and accidentally, thereby, making himself look as if he is a peg-legged pirate. But Henry’s acting in the aforementioned way does not suffice for Henry to be pretending to be a peg-legged pirate. What else is needed? Perhaps, it might be thought that Henry must also be imagining what-it-would-be-like to be a peg-legged pirate while acting in ways that make him look like one?

Yet, that too is not enough. The mere addition of mental imagery, however appropriate, is not sufficient to close the deal in this case. For example, while hobbling around on his stubbed-toe, Henry might notice in the mirror that he is moving around like a peg-legged pirate. Noticing this fact about his behavior might cause him to recall how Robert Newman, the actor portraying Long-John Silver in the eponymous film about that famous pirate, moved about. Henry may note the similarities. He might have all sorts of further mental imagery. He might think about how other peg-leg pirates hobble about. All of this might flood Henry’s mind during his own hobbling about. Yet the addition of this sort of mental imagery, however much of there might be, would hardly convert his embodied activity into an act of pretending.

What is missing? It might be thought that for the addition of such mental imagery to make his activity into an act of pretending Henry needs to invoke mental imagery of the aforementioned sort in order to help bring off the effect of looking like a peg-legged pirate. Would the activity in question count as an instance of pretending if Henry had, at some level, opted to use said mental imagery in order to guide his performance – namely, if he started hopping about as such for the purpose of making himself like a peg-legged pirate?

Of course – and here’s the rub – for that to be the case, Henry would, it seems, already have had to have had the intention of making himself behave like a peg-legged pirate, at least in certain salient respects. Thus, it turns out, what matters for some activity to be an instance of pretending is that the person has the right kind of intention for engaging in the relevant embodied activity.Footnote 18

Even if having contentful prior intentions is not necessary in all cases of pretense, as per the discussion of section one, surely there are many parade cases of pretense that not only happen to involve having such intentions but in which having said intentions is definitive of those cases of pretense.

If so, what’s an enactivist to do? Langland-Hassan (2020) identifies a possible way by which enactivists might answer this challenge. As he notes, it is perfectly possible even for radical enactivists to admit contentful attitudes in their accounts, including contentful intentions and beliefs even while rejecting the existence of contentful mental representations tout court.

That particular combination of commitments is possible because the fact that S’s harboring a contentful attitude – say, S’s having the belief that P – does not logically entail the existence of contentful mental representations – say, a mental representation with the content p. The proposal that contentful attitudes depend upon the existence of contentful mental representations is just one possible theory about their nature.Footnote 19 For this reason, it is perfectly possible to “talk of beliefs and intentions, without countenancing interpretations of those states that commit one to mental representations with corresponding semantics” (Langland-Hassan, 2021).

RE is already fully committed to the view that folk psychology only has light-duty ontological commitments (Hutto, 2008). Indeed, RE absolutely insists on the importance of contentful attitudes when it comes to understanding certain sophisticated kinds of imagination, memory, and other cognitive phenomena (see Hutto 2015, 2017b, Hutto & Myin 2017). The situation is no different with respect to pretense: RE acknowledges that in some central cases, pretense necessarily involves contentful attitudes.Footnote 20

In underlining this point, it is important to note two important caveats. The first is: committing to the existence of contentful attitudes does not imply, not even tacitly, a commitment to attitude-internalism. Taking stock of individual mindsets is entirely compatible with recognizing the importance of social, contextual, environmental factors when it comes to characterizing such mindsets. In particular, we can assume that understanding someone’s contentful attitudes may be crucial for understanding some case of pretense without assuming that “what accounts for the difference between pretense and non-pretense will be … something in the head of the pretender” (Langland-Hassan, 2021).

Locating the difference between pretenders and non-pretenders by looking at differences in their individual mindsets of is, if RE is correct, not a matter actually looking inward, at what is going on inside their heads.Footnote 21 RE rejects any attempt to locate or understand contentful attitudes by focusing on what happens in the heads of individuals. Indeed, RE maintains that we must look beyond what is ‘in the head’ of individuals if we are to characterize their mindsets. It is not possible to characterize the internal mindsets of individuals, neither from the inside nor the outside, without considering the person’s history and wider contexts in which they find themselves.

The second caveat is: while it is important to acknowledge that understanding a person’s contentful attitudes may be indispensable for understanding certain acts of pretence, doing so in no way commits RE to the much stronger thesis that it is possible or desirable to try to specify the necessary and sufficient conditions of pretense in wholly mentalistic terms.

Langland-Hassan (2020, 2021) has such reductive ambitions. But there is reason to suspect that, like many a reductive formula of the analytic variety, the particular recipe he proposes is vulnerable to counterexamples. For example, consider a case of a lacklustre toolmaker – call her Teresa. Teresa is a first-rate, expert toolmaker. Let’s suppose that she sets to her task with a firm intention to make a piece of wood into a tool. Perhaps, she is even duty-bound to do so. But let us also suppose on this occasion, it so happens, that her heart just isn’t in the job. She’s simply not in the mood to make this piece of wood like a tool to sufficient degree to produce a serviceable tool. Moreover, she knows it. Her efforts are utterly half-hearted and lacklustre: she just can’t be asked. Still, she carries on with the work. Teresa’s mindset is still such that she intends to make the item tool-like in relevant respects. Yet, taking stock of own her lack of enthusiasm, she simply does not believe that she’s going to produce a viable tool on this occasion. Prima facie, Teresa appears to satisfy Langland-Hassan’s basic mentalistic criteria for pretense – she intends to make some x y-like while believing that x will not, in the process, be made into a y. Yet, despite this, we would be hard-pressed to say that Teresa is just pretending in this case to make the piece of wood tool-like. So it looks like being a pretender requires something more than the particular combination of attitudes specified by Langland-Hassan’s (2020) mentalistic recipe.

Or consider another case - the familiar practice of pretending to be something one is not in order to be or become different than one is. Consider the case of Katrina. She is a nervous flyer but has to take lots of long-distance flights for her job. She knows that flying is safer than many other things that she regularly does but she can’t shake off her irrational fear of it. So, Katrina settles on the following trick: She decides to try to pretend to be the sort of person who loves to fly rather than fears it. Moreover, she enters into this pretense for the express purpose of making herself into the sort of person who loves to fly. Independently of her actual chances of success, she is extremely confident that this method will work for her. Thus, she intends to make herself y-like, but she does so precisely because she believes she will become y-like in the process. Katrina’s ‘fake it until you make it’ approach to overcoming her fear of flying looks like a bona fide form of pretense practice but it seems to violate Langland-Hassan’s (2020) mentalistic formula.

Perhaps these counterexamples can be dealt with. Even if they can, it is likely that more will crop up. Indeed, if the history of philosophy has taught us anything at all it is that analytic philosophers are pretty good at cooking up counterexamples to defeat each other’s proposed reductive analyses of important phenomena. It would be rather surprising if Langland-Hassan’s (2020) efforts in this regard proved more untouchable than other more famous but ultimately failed attempts to provide a final analysis of, say, knowledge or causation. Still, in any case, the discussion of section two already supplies independent reason for steering clear of this kind of essentialism. And that is precisely what RE does.

5 Conclusions

On the basis of the foregoing analyses, it should now be clear that RE is in a position to give an account of the full range of cases of pretense. It avoids over-intellectualising the phenomena after the fashion of representationalists; it avoids the problems faced by PE; and it is able to answer Langland-Hassan’s mentalistic challenge. In particular, it looks as if radical enactivism can provide what is needed if we are to keep it real when trying to understand and do justice to the many and various practices of pretense.