Talking About Thinking

Language, Thought, and Mentalizing

Leda Berio

• Publisher: De Gruyter
• Year: 2021
• ISBN:
  9783110748550
• Edition: 1
• Topic: Filosofía
• Subtopic: Epistemología en filosofía

Our ability to attribute mental states to others ("to mentalize") has been the subject of philosophical and psychological studies for a very long time, yet the role of language acquisition in the development of our mentalizing abilities has been largely understudied. This book addresses this gap in the philosophical literature.

The book presents an account of how false belief reasoning is impacted by language acquisition, and it does so by placing it in the larger context of the issue, how language impacts cognition in general. The work provides the reader with detailed and critical literature reviews, and draws on them to argue that language acquisition helps false belief reasoning by boosting the ability to create schemata that facilitate processing of information in some social contexts. According to this framework, it is a combination of syntactic clues and cultural narratives that helps the child to solve the classic false belief task.

The book provides a novel, original account of how language helps false belief reasoning, while also giving the reader a broad, precise and well-documented picture of the debate around some of the most fundamental issues in social cognition.

Part I: Preliminaries

Chapter 1: Introduction

1.1 Talking about thinking and thinking about others

The question of what makes humans able to communicate, has fascinated researchers since the very first philosophical and scientific inquiries. This is not surprising, since other animals’ communication skills, while impressive, do not match natural human language.

Not only does natural human language seem to be a unique communicative tool, it also appears to come with a set of fascinating cognitive skills. Being able to communicate verbally entails an ability to use words, and also an ability to understand how our message will be received, guess how the interlocutor will understand it, and, in general, relate to and interact with other people in a meaningful way. In a sense, communicating is the enterprise of making two minds, two internal worlds, interact. If that was not enough, language allows us to discuss, evaluate, and predict other people’s behavior in a verbal form, to build narratives around other agent’s behavior, and to discuss them with our peers. This book is, in a way, about how communicating verbally and understanding others are deeply intertwined human abilities.

The ability to understand other’s behavior, and to do so in the light of one’s mental states, is often called mindreading or mentalizing in philosophical literature. It entails, or so the traditional story goes, the ability to attribute beliefs, desires, and thoughts to other agents. This ability has long been thought to be related to language: on the received view, language develops as a consequence of the fact that humans have mentalizing skills. In a nutshell, since we are able to understand what goes on in other people’s minds, we have the skills to develop a language that allows us to communicate about it. This has been strongly influenced by the idea that mentalizing skills are possibly innate and modular, i. e. they are mostly present at birth, and only marginally dependent on experience.

More recent theories and empirical discoveries, however, seem to cast some doubt on this picture. On the one hand, apes seem to be capable of some (albeit limited) mindreading (Call and Tomasello, 2008; Suddendorf and Whiten, 2003). On the other hand, growing evidence suggests that language might be what aids the development of mentalizing (Diessel and Tomasello, 2001; de Villiers, 2005; Hale and Tager-Flusberg, 2003; Wellman et al., 2001). This seems to turn the traditional picture upside-down: perhaps it is not that our ability to communicate stems from our ability to understand others, but that our ability to understand others originates from our ability to communicate.

This is connected with another important issue in philosophy of cognition, which is that of the role of language in general: what is the role that natural language occupies in our cognitive architecture? Does acquiring language shape and rewire our brain and cognitive skills dramatically, or should we, rather, think of language acquisition as a domain-specific step in our cognitive developmental path, something that is only used for the specific purpose of communicating and barely entails cognitive changes in the rest of the system?

This book offers a partial reply to these questions – one relative to the development of mindreading skills and their relationship with language. I take an empirically motivated and philosophically argued stance against the traditional picture: I argue that the interaction between pre-verbal mentalizing abilities, language acquisition, and more sophisticated forms of mindreading is a dynamical and complex interaction. It is not the case that our mentalizing ability precedes and enables language tout-court; on the contrary (and this is the second stance I take), acquiring language provides a substantial boost to our abilities to understand others.

Arguing for this position, I simultaneously argue for a specific relation between language acquisition and our ability to understand others, and for a picture of cognition that sees language as a powerful cognitive tool. Far from being just a communicative device, the role of language in this picture is that of enhancing pre-existing abilities and potentially enabling new ones.

1.1.1 A brief background

Research on social cognition has been animated for a very long time by a crucial study (Wimmer and Perner, 1983), in which, for the first time, the false belief task paradigm was used: experimenters realized that 3-year-olds, but not 4-year-olds, reliably failed at a task in which they had to attribute a false belief to an agent in order to pass. The underlying idea was to test whether children were able to understand that a character leaving the room before the object of interest was moved was not supposed to be able to know, upon return, that the object had changed location. The first interpretation of this finding, which was replicated several times, denies to preschoolers under four the ability to correctly understand that agents can have false beliefs. The study generated a long, heated debate between, initially, Simulation Theory and Theory Theory of mentalizing (Gopnik (2001); Perner (1991); Gordon (1986) and others, summarized in chapter 4), followed by a long series of empirical studies aimed at replicating or reconsidering the initial conclusions (Onishi and Baillargeon (2005); Baron-Cohen et al. (1985); Zaitchik (1990); Scott et al. (2011); Scott and Baillargeon (2009) and many others, summarized in chapter 4), and then an ongoing attempt to achieve clarity in the current literature and to formulate accounts that take into consideration all of the available data (Goldman (2006); Perner et al. (2015); Apperly and Butterfill (2009); Apperly (2011) and many others (chapter 5). On the one hand, there is a debate around whether or not children under the age of four are endowed with the representational and conceptual abilities necessary to attribute false belief; on the other hand, another fundamental question is “What exactly are the representational abilities involved?”

The debate on the role of language in cognition also has deep roots. Since the proposal of the Whorf-Sapyr hypothesis (Whorf, 1956), the research landscape has changed several times, going from a complete rejection of linguistic relativity to a more open-minded attitude towards the claim that language can shape our conceptual world. As I explore in the first chapter, the question has been discarded from the very start by so-called communicative views of language, according to which natural language is naught but a tool to communicate thoughts that are in another format (Fodor, 2008). This kind of view sees propositional thought as completely independent of language. On the other side of the spectrum, several theories attribute to language a pivotal role in “rewiring cognition” (Karmiloff-Smith, 1994), or providing the possibility for propositional thought per se (Tillas, 2015a). This debate is entrenched within several questions: how does a specific language shape the conceptual categories humans have (Lupyan, 2009, 2012), how are the concepts acquired in the first place (Carey, 2009), what is the underlying role of cultural practices (Gopnik, 2001), and what is the correct way to consider cognitive architecture (Carruthers, 2012)?

Finally, another domain of philosophical and psychological analysis is concerned with the more specific problem of how much of our social cognitive abilities is related to the development of language skills, ranging from theories that consider language as constitutive of some of the abilities involved in false belief reasoning (de Villiers, 2005), to theories that stress the cultural role of social narratives (Nelson, 2005; Hutto, 2008b), to theories that lean towards a middle ground between the two positions and argue for a role for both social development and linguistic acquisition (Garfield et al., 2001), and to many other positions besides (Van Cleave and Gauker, 2013; Gordon, 2007; Bermúdez, 2009; Montgomery, 2005). This debate has extended to linguistics (de Villiers, 2005), as it involves the analysis of how language-specific structures influence on the development of cognitive abilities.

This work’s leading question is concerned with the extent to which language acquisition influences our cognitive abilities, with a focus on false belief reasoning in relation to the acquisition of mental state verbs. This might appear to be a very specific question; however, as the reader will see, there are many intertwined issues and problems which come together to form an answer to this question, and it is on the literature listed above that the work in this book relies.

The set-up of this book is funnel-like. It starts with a rather general set of issues (theories of language and cognition) and works towards a very specific one one; this not only reflects my personal path of inquiry, but also avoids the risk of obtaining an account of the influence of language on false belief reasoning that is completely disconnected from a general view of cognition. I will come back to this topic more than once in the upcoming chapters. This does not mean that I give a final answer to the question of what the exact relation between linguistic and non-linguistic cognition is in general; it means however that I give an account of how mental state terms help the child developing abilities to pass the false belief task in a way that does not disregard research concerning language and cognition, but rather is compatible with it. At the same time, a fundamental concern in this book is to give an account of the role of language in social cognition that also fits with the developmental trajectory of the empirical literature that focuses on pre-verbal skills; this is important, because, as will be seen in chapter 5, many rival accounts fall short of this objective. In this sense, while the scope of the book might appear narrow, there is a conscious effort to make my account fit with the bigger picture in two ways: in relation to the development of other skills that are central for mentalizing, on the one hand, and in relation to the more general problem of how linguistic coding and non-linguistic coding interact in human cognition, on the other.

For these reasons, I start with an overview of the issues underlying the interaction between language and thinking in chapter 2. I argue in this chapter that the most fruitful perspective is pluralism, i. e. a perspective according to which the problem of the relation between language and thought is tractable only once we define with more clarity the domains of our inquiry. On the one hand, we have to choose the cognitive domain to explore; on the other hand, it is useful to think about which kind of linguistic information is most likely to be relevant to that domain. This is the underlying reason why the rest of the book focuses on the very specific question, whether acquiring mental state terms (described in chapter 3) has an impact on how children pass the famous false belief task in its explicit variation. The answer to this question is mostly given in chapter 4, and it is one that builds upon the empirical data and the theoretical contributions, which are further analyzed in chapter 5. However, this book is set up to answer a related, but different question as well: not only if mental state terms play a role, but also how. The answer to this question is given in chapter 6, which contains my theoretical proposal. In chapter 7, I indicate a path for further research and draw more general conclusions.

1.2 Methodological note

What makes this work a philosophical enterprise is the fact that I constantly relate the issues at hand to the underlying theoretical questions about human cognition and mind; what makes this work an empirically-informed philosophical enterprise is that I give constant priority to experimental data, taking seriously the importance of experimental paradigms, design, and analysis choices. While this is not meant to be a psychological inquiry, and a very modest amount of attention is dedicated to statistical analysis of data, I do put empirical reviews at the center of this work. This is not only true for psychological data, but also for language: the reader will realize, while reading this book, that a lot of importance is given to linguistic analysis. This is the case for two fundamental reasons: on the one hand, it is rare that philosophy, psychology, and linguistics tend towards the same research themes in such an evident way as in language and mentalizing research. Studying the influence of language learning on mentalizing tasks necessarily means considering language acquisition literature as well as cognitive psychological data. On the other hand, an issue that is of fundamental importance for me and my work is that of cross-cultural and cross-linguistic analysis. While the cross-linguistic data available is limited, I make an effort through this work to specify which findings, heavily supported by research with English-speaking participants, can indeed be extended to other populations, and which of the linguistic features are shared by other languages. As I further elaborate in chapter 3, this is of fundamental importance in order to make today’s psychology and philosophy as impartial as possible. Hence, when achievable, I report on cross-linguistic data available in the literature: this sometimes includes reporting on technical linguistic phenomena, but I do provide the theoretical tools that are necessary for understanding the underlying issues. While not being a linguistics book, this work is meant to be informative for philosophers, linguists, and psychologists alike, in providing conceptual tools that lie at the interface of these discipl...