What it is like to be a pickpocket

Introduction

In this study, I present (a)socio-cognitive skill of a special kind from the perspective of cognitive ecology. I focus on pickpocketing, using scenes from Pickpocket (1959) by Robert Bresson as a kind of illustrations. This is not a movie study. Instead, I use the film excerpts to better demonstrate how the cognitive dimension of pickpocket activity can be seen in the context of newer trends in cognitive science.

In the literature of cognitive science, there are known cases of the use of meaningful thought experiments, such as the famous case of Otto in the context of the extended mind hypothesis, described by Andy Clark and David Chalmers (1998). Films have also been referred to in research conducted on perception, e.g., Joseph D Anderson (2002) used clips from Mission Impossible (1996), Titanic (1997), and other movies in his work. In this article, I will treat the selected scenes from Pickpocket as a screened thought experiment. Thought experiment is the use of a hypothetical scenario or situation to increase the understanding of certain phenomena. It is not indispensable to prove research hypotheses, but is very helpful. In this light, the excerpts from Bresson’s movie allow me to better show some possible mechanisms of the interaction of pickpockets with their socio-material environment. This approach can greatly help in understanding ecological research on cognition.

I use this specific movie due to its suggestive impact on the onlookers. Movies show phenomena more effectively than neutral, selective recordings of research, although (or for this reason) the former are something more. Our susceptibility to the influence of the cinema is part of the situating of human cognition in culture. While this is not enough to treat the study as a transdisciplinary approach, I believe it can inspire cooperation between filmmakers and cognitive researchers.

My account differs from both ecological research on film perception and the dominant approach in ecological philosophy and psychology. I use the concept of distributed cognition, which I additionally treat as a kind of mechanism-based explanation of unique, extended cognitive phenomena (and therefore not an explanation based on universal laws or functionalism). In this context, I place great emphasis on the phenomenon of affordance. However, my use of this category is closer to its use in studies of design and human–artifact interaction than to its meaning in Gibsonian tradition.

I assume that the pickpocket’s activity is based on the skillful and atypical use of affordances of things and social affordances in a distributed cognitive system. Using the aforementioned theoretical tools and the film scenes, I try to show the perspective of such an atypical and clever “user” of a given environment (including places, people, and things). Rather than analyzing the presumptive state of the pickpocket’s mind, I focus on the analysis of the agent’s interaction with the environment. I present the advantages of such an approach compared to research involving real pickpockets and the “neuroscience of pickpocketing.” While I do not diminish their value, I wish to demonstrate the possibilities of enriching the conclusions of these studies with an additional perspective.

Cognitive ecology and affordances

As I mentioned before, the general framework for my account is cognitive ecology. According to Edwin Hutchins (2010), cognitive ecology is the study of cognitive phenomena in context (biological, social, cultural). In general, it focuses on the web of heterogeneous interactions in a cognitive ecosystem. The crucial heuristics here is that socio-cultural conditioning has to be taken into account not at a later stage of analysis, but at its beginning. In the context of cognitive ecology, I am interested in two ecological approaches: distributed cognition theory and the concept of affordances.

The first one—developed by Hutchins (1995, 2014), David Kirsh (1995), Nancy Nersessian (2006), and others—describes cognitive systems that usually go beyond the boundaries of an individual agent. These systems include the interactions between human and non-human components, such as artefacts. One has to keep in mind that the boundaries of the systems are not fixed in advance, but depend on the scale of the cognitive structure under investigation, and at the same time on the given research problem. From my point of view, distributed cognitive system can be viewed as a system that is based on mechanisms (Wachowski, 2018). The mechanisms are more or less durable and more or less tightly organized. It is important to be able to distinguish between genuine components of the cognitive system from causal background conditions (see, e.g., Kaplan, 2012). In general, according to distributed cognition theory, our cognition is permanently situated not only in biology but also in culture. Human perception and action always take place within cultural practices.

Affordance is a key concept in ecological psychology. It was coined by Gibson (1966; 1979) and subsequently developed and modified by a range of theorists and researchers (see Dotov, Nie, & de Wit, 2012). According to ecological approach, perception and action are tightly coupled and mutually constraining. One can say that affordances are relational action possibilities provided for the acting agent by objects (including other people) in her/his environment. The perceiving of affordances is a process of perceiving not value-free, but value-rich ecological objects. Thus, from a human perspective, pavements are for walking, stairs are for climbing, chairs are for sitting, sometimes stones are for sitting too, sometimes chairs are for breaking out a window, and so on. From a monkey’s perspective, trees are for climbing more than stairs. There are also social affordances like gestures, facial expression, gaze and the body in general, affording, for example, a common action like kissing, quarreling, having sex, escaping, or giving help. Of course, mistakes and misunderstandings can happen. In sum, affordance is a kind of cognitive shortcut that skips concentration and analysis and is based on the dependencies between structured environment and agents.

Gibson’s theory has resulted in many of valuable studies. In the context of my article, attention should be paid to the analysis of the relationship between affordances and intentionality in perception occurring in a meaningful human world (Heft, 1989), as well as later studies on the role of the social context in the perception of affordances: in this light, affordances can be shared, we perceive some (most?) of them together with other(s), “behaving with,” sometimes in cooperation (Costantini & Sinigaglia, 2012; Rączaszek-Leonardi, Nomikou, & Rohlfing, 2013; Richardson, Marsh, & Baron, 2007). An important account of the concept of affordances in everyday activity is provided by Alan Costall (2012). He describes cultural affordances as established and widely acknowledged use-meanings of things. According to him, we ought to try to understand the objects within a network of relations among different people, but also other objects drawn into a shared practice (p. 92). As Costall and Richards (2013) claim, “affordances are not simply discovered, but nor are they mentally projected upon inherently meaningless things. They are negotiated” (p. 91). These negotiations are situated in our cultural ecosystems.

The term affordances also has a second, parallel life. It has proved to be influential in areas of design (McGrenere & Ho, 2000). The concept of affordance was introduced in these fields by the cognitivist Donald Norman. For him, this term “refers to the perceived and actual properties of the thing, primarily those fundamental properties that determine just how the thing could possibly be used” (Norman, 1988, p. 9; see also Norman 2013). In studies on design, there is an important issue worth our attention: analysis of structural dependencies between affordances and perceptual information. William Gaver (1991) distinguishes perceptible affordances (e.g., affordance of the door) and false affordances (like in the case of a dummy door handle) as well as hidden affordances (affordance of the handle hidden in the structure of the door). Another important analysis the researcher made concerns affordances for complex actions. He distinguishes sequential from nested affordances. The former refer to a situation in which acting on a perceptible affordance reveals information indicating new affordances (like in the case of opening a door: a sequence from grasping the handle to full opening); sequential affordances are revealed over time. Nested affordances are grouped in space. In the case of the door, “only by seeing the affordance of pulling the handle as nested within an affordance of pulling the door can opening the door be a perceptible affordance” (p. 82). Moreover, he explores issues of designing artifacts that support collaboration; affordances described by him are possibilities not only for individual action but also for social interaction (Gaver, 1996).

In sum, the approaches within cognitive science, presented here, seem to have one important idea in common: we are thinking not only with our brains; we are thinking with our eyes and hands, things, as well as with other people (see Latour, 1986). In this context, the category of affordance seems to be a good way to describe our cognitive relations with the environment.

In my analyses, I distinguish perceived, false and hidden affordances. This is not strictly in line with Gaver’s proposal. I have adapted his general classification; however, I also use it for body affordances and social affordances. For this reason, “hidden affordance” is, according to me, an affordance that provokes a particular action or behavior, but is not made aware by an agent, especially since it is usually nested in a set of other affordances that may be perceived or false.

Loners and magicians

Many of us have probably heard of the famous illusionists like Great Tomsoni, David Copperfield, or perhaps the younger deception specialist Apollo Robbins (TED Talks, 2013), who cooperates with scientists, security agents, filmmakers, and others. In general, there are various ways of studying the pickpocket’s craft, from studies on misdirection to the neuroscience of pickpockets (or rather the neuroscience of magic tricks).

On the one hand, I am not sure that the existential dimension of being a pickpocket is irrelevant to her action and perceived world (Greydanus, 2018). Probably one of teachers of all pickpockets is fear; their motivations are anti-social. The pickpocket (or a group of pickpockets) works in hiding—a fact that could have phenomenological significance. A controlled pickpocket becomes just a magician. It is difficult to say how much the activity space of the latter is similar to the thief’s space.

In this light, one can analyze the movie Pickpocket by Bresson. Its main character cannot make contact with other people. He is an isolated loner who avoids contact with others, except for (paradoxically) acts of theft. He was told: “You’re not in the real world …  You don’t share any interests with people” (see J. Clark, 2012).

On the other hand, I don’t ask about the nature of human attention, how the pickpocket’s brain works, or how the pickpockets’ victim brain works. I ask about how pickpockets’ activity is situated in their cultural ecosystem. So, what don’t the previous and current studies explain?

Let’s look at the phenomenon of misdirection described by Susana Martinez-Conde and Stephen L. Macknik (2008). Misdirection is the practice of diverting the onlooker’s attention away from a secret action. “In the lingo of magic, misdirection draws the audience’s attention toward the ‘effect’ and away from the ‘method’, the secret behind the effect” (p. 73). From the perspective of cognitive studies, on can distinguish “overt” misdirection, redirecting the onlooker’s gaze away from the method to a particular object, and “covert” misdirection: thanks to this subtle magicians’ technique, spectators may be looking directly at the method and remain unaware of it. Usually, researchers describe two kinds of covert misdirection. One of them is change blindness, when observers do not notice changes—expected or unexpected—by looking at the scene at any one instant in time (Rensink, 2005). In turn, during inattentional blindness people fail to notice unexpected objects that are fully visible in front of them (Simons & Chabris, 1999). Additionally, some psychologists describe a third kind of cognitive illusion: choice blindness. General speaking, studies on this phenomenon show how people justify their choices, including the choices they haven’t actually made (Hall & Johansson, 2009; Martinez-Conde & Macknik, 2008, p. 73).

Of course, misdirection techniques can also be used by pickpockets. However, the research on the phenomena mentioned above don’t reveal much about magicians’ or pickpockets’ perception in action and even about the manual techniques used.

Some information about the latter can be found in studies by Martinez-Conde and Macknik (2008). These researchers, in collaboration with Robbins, suspected that two kinds of motions are essential to effectively misdirecting the mark. According to them, if pickpockets want to attract spectators’ attention to the entire path of motion, they sweep out a curved path, and, in contrast, if they want to reduce spectators’ attention to the path and quickly shift the attention to the final position, they trace a fast, linear path. The sharp curves and corners seem less predictable and redundant and can be therefore novel and informative; curved trajectories seem less redundant, so more salient than straight ones (p. 78; compare, e.g., with Yeager, 1990). A later study (Otero-Millan, Macknik, Robbins, McCamy, & Martinez-Conde, 2011) shows that research on magic misdirection can shed new light on neuropsychological principles of perception and cognition.

In spite of all, pickpockets do not work in a vacuum. They do not steal human minds, but real things of value. The pickpocket’s craft is related to the infrastructure of his environment. For this reason, to understand pickpockets, we need to understand their interactions with the specific cultural ecosystem in which they work. The affordance-based perspective described in the two previous sections seems to be the optimal approach to the ecological studies on pickpocket’s perception.

The mind loves the hand. The hand loves the crowd

To better identify important aspects of the cognitive engineering of this kind of thieves, I analyze some aspects of pickpocketing presented in the movie Pickpocket by Bresson. Why have I decided to use excerpts from the old movie, apart from the theme? On the one hand, the director often shows antisocial or simply non-social behavior. He maintains aloofness and emotional distance between the viewer and the film. Actors appearing in his movies are non-professional, their acting is minimalist. Viewers get something like action schemes, emotionally “discolored” moving images (see Wachowski & Bogucki, 2017). On the other hand, an especially interesting and complex example of human–environment interactions is shown in Pickpocket. This is a great show of the cognitive engineering of the pickpocket craft.

We need to bear in mind that Bresson’s movie is not a documentary or a faithful reconstruction of a crime story. This is why the scenes it contains may be exaggerated and larger than life (see J. Clark, 2012). What is important is that they present pickpocketing techniques in the context of the opportunities and constraints of a given environment.

In the movie, one can observe a young thief Michele. He is practicing his antisocial skills that need a social context. Maybe the pickpocket doesn’t share any interests with people, as Jeanne, a neighbor of his mother, told him, but he shares many affordances with them. Let’s see what the consequences are.

In the context of my analysis of the movie, some questions are raised. Is the pickpocket’s perceived world structured in a different way? Can we identify the same affordances shared among thieves and their victims? Can we identify special affordances for pickpockets? How do they use things when controlling others’ perception? Which artifacts (including places) are “well designed” from the point of view of pickpockets?

It is hard to say if our socio-material world is designed for honest people, or for a more multifaceted society. Some people can perceive and use apparent and hidden features of this world—and at the same time various affordances—for their own purposes. It is obvious that we all perceive some affordances together, but not necessarily in cooperation, as we have different intentions. Our thief is often focused on some hidden or undesirable properties of things or constellations of things, like space between clothes, slipperiness of wallets, or a crowd in the subway, as well as, of course, on the possibilities offered by his own hands. Michele successfully uses different kinds of affordances in his craft and at the same time constantly pretends shared activity, using gaze contact or body contact, and hiding his true intentions. There is a constant tension between focusing on own action and managing others’ perception.

To answer the question about what a thief manipulates in the immediate environment of the victim (including the surface of the victim’s body), we need to realize the nature of the thing which is our property and which we keep close to protect it from others to a greater or smaller extent. The movie helps us to understand this. It is not the remembered bits of information we take with us when we leave home. Neither is it something we keep an eye on all the time. Every so often, we try to perceptually refresh the feeling that a given thing is still with us either by touching (directly or through clothing) or looking at it. Theoretically, such refreshing could be all we ever need, if it was not necessary to use the object, change its location or sometimes just confirm visually (by looking at a visible fragment) that the thing has not disappeared. As a matter of fact, the same applies to things we keep at home. By the way, do we certainly know how our thing looks exactly?

I think that this temporary and fragmentary refreshment of the feeling that the thing is still present often hinges upon one or two perceptible affordances which a skilled pickpocket would have no trouble replacing with false affordances. For a long time, the owner of a handbag or wallet needs to access only these affordances: a woman just needs to be aware that a compact, smooth object of a certain small weight is pressed against her side by the elbow, whereas for a man it would be enough to know that a flat rectangular object carried in an inside pocket can be touched through the fabric of his jacket. This fragmentary, perceptual control can easily be engaged for a time with objects of similar properties, creating the seeming affordance of a handbag (something easily carried when tucked under the arm) or wallet (a convenient container for money that is easy to keep under the outer layer of clothing and easy to take out if necessary). But this is not all. As he is stealing, a pickpocket would also use false affordances as various “smoke screens”—from the simple newspaper he seems to be reading, but which is in fact masking his hands on the victim’s property, to the obvious use of a crowd (often artificial when there is a group of pickpockets operating) to the subtle use of social affordances like giving a helpful hand to a man boarding a train (while taking the watch off his wrist). Such arranged social situations abound in interesting applications of special types of sequential or nested affordances: a pickpocket would insert use of hidden affordances into sequences or groups of perceptible or false affordances as in the case mentioned above. Sequential and nested affordances are also related to the complex and subtle procedure of identifying and using subsequent hidden affordances that emerge during the theft, for example, when a pickpocket assesses the technical properties of the victim’s handbag, then opens it, rummages inside and smuggles selected objects from the handbag into his own pocket. It would be hard to talk of any predictable sequence of activities both when it comes to the handbag and its contents as well as the place where the theft takes place. Smart thefts are improvisations involving the victim and in the presence of potential observers.

Does a handbag have different affordances for the owner and the pickpocket? To a large degree, yes. We might even say that these are two different objects. What is important for the owner is that the bag is convenient to carry by hand, offers a lot of room, closes and opens easily, and has extra inner compartments. It might also be a new bag from a prestigious producer, making it worthwhile to expose it publicly (if the bag bears a visible logo of the producer, it is good having it on display). The same elements may also be relevant for a pickpocket. But, to him, they afford something different: easier access to the inside of the bag in the absence of the owner’s attention, shape and size to be replaced by a dummy, less protection by the owner, etc.

Cooperation between pickpockets, on the other hand, requires sharing certain affordances, mainly hidden, and behaving in a specific fashion or “behaving with,” as has already been mentioned; this is also used by thieves working alone—faking collective interest in a scene on the street, discussion, a violent quarrel, or an artificial crowd.

So it can be said that the pickpocket mainly manipulates hidden and false affordances. The former are the properties of things and their material context which the thief has learned to operate fluently, partly thanks to his well-trained hands. The latter—false affordances—are chiefly falsely used (i.e., with a hidden intent) social affordances. We see this, for example, when the pickpocket offers a helping hand to a passenger climbing steep steps in order to board a train. While grabbing and pulling him into the train, he surreptitiously steals a watch from the victim’s wrist. Note that apart from the hidden and clever theft, the key element here is focusing shared attention on the act of providing help and the creation of emotional involvement in the grateful victim. The use of social affordances is equally important in situations that are characteristic of a non-centralized, distributed socio-cognitive system, like in the case of group activity. A group of pickpockets working together needs a pretext. The simplest example of this is a fake crowd in a doors or a passageway, where a purposely distributed group of thieves is effective when each of them plays a role, faking social affordances if necessary.

Of course, the scenes from this movie don’t reflect the full repertoire of possible roles that affordances can play in the interactions between pickpockets and their surroundings. For example, in the movie Favourites of the Moon (1984), the mother of a novice thief suggests that he gets rid of stolen pockets, whose affordances can lead the police to the trail. In another movie, Breakfast at Tiffany’s (1961), there is a funny (but worth reflecting on) scene where two people use full visibility as a camouflage affordance, leaving the store wearing stolen masks on their faces. As for non-movie examples, it is worth remembering how useful modern technology can be for a thief: they can, for example, use a mobile phone, a pair of headphones, or a tablet computer in a crowd to communicate with another pickpocket. It is also worth paying attention to the role of recursive thinking in the interactions between thieves and policemen: specially trained policemen, in plain clothes and hidden in the crowd, use affordances which are attractive for pickpockets; the latter in turn learn to recognize such policemen; more experienced policemen recognize such advanced thieves thanks to the “advanced” affordances used by them, and so on. ¹

All this begs the question: do social affordances constitute the necessary components of these wide mechanisms or merely the background conditions in the described mechanisms of pickpocket activity? I think that both are true. Certain acts of theft are better shielded by means of social affordances, because they distracted more effectively. Others are based on the manipulation of social affordances as the indispensable means to mask the crime. However, this requires empirical verification of constitutive relevance in the mechanisms (see remarks on the Otto-Inga thought experiment in Kaplan, 2012; see also van Eck, 2019). My short study only provides some clues.

Conclusions

If the assumptions of cognitive ecology are right, the question about what it is like to be this or that agent acting routinely is also a question about the cognitive structure of her or his socio-material environment. The meaningful example of the pickpocket was intended to show the potential and limitations of research in this field. I have shown this with help of the concept of affordance and distributed cognition, using excerpts from Bresson’s movie as a kind of illustrations. Pickpockets do not work in a vacuum but are a special kind of users of urban spaces, and thus they are experts in local cultural practices and using everyday things in the city.

If we want to find out how technically advanced thieves are today, we ought to look carefully at modern security techniques. If we want to understand the activity of pickpockets to some extent, we should try to look at everyday things from their perspective. An affordance-based approach seems to be very useful in studies on different “socio-cognitive engineers.” It requires integrating certain findings of ecological psychology and studies of design, including insightful knowledge of human–artifact and human–human interactions.