Reading :: Transforming and Understanding

 Transforming and Understanding: An Introduction to Cultural-Historical Activity Theory

By Yannick Lemonie

I’ve been waiting for this book to be translated into English, and as soon as the English version was released, I ordered it. Lemonie teaches ergonomics and conducts developmental research in Paris, where he uses the change laboratory methodology, and he has developed his understanding of CHAT in close contact with Yrjo Engestrom and Annalisa Sannino (who wrote the Foreord and Postface, respectively). 

As the title suggests, Lemonie intends this book to be an introduction to CHAT — more specifically, Yrjo Engestrom’s version of CHAT. The book has three parts:

• Part 1, “Four Generations of Cultural-Historical Activity Theory,” covers the history of CHAT in the “generations” that Engestrom has popularized.

• Part 2, “Four Central Concepts for Analyzing Activity,” examines the object of activity, the concept of contradiction, expansive learning, and development as qualitative reorganization. 

• Part 3, “The Interventionist Dimension of CHAT,” focuses on the change llaboratory methodology and contrasts formative interventions with other interventionist approaches.

There’s a lot of overlap between Part 1 of Lemonie’s book and my book Triangles and Tribulations, which will be published next month. But they represent two different approaches to CHAT, with Lemonie’s book more or less seeking a single history leading to a single CHAT, while my book understands CHAT’s history and concepts as much less coherent and centralized. But put those differences aside for a moment. Lemonie’s book excels in Part 1, systematically examining the “generations” through detailed research into how Vygotsky, Leontiev, Engestrom, and Sannino understood and developed CHAT. In each chapter, he provides tables describing the different periods in the development of the person whose work is being examined. 

In the chapter on Vygotsky, he covers mediation (including the collapse of physical and psychological tools), the instrumental method, and analysis by unit, discussing how these set up the second generation. In his chapter on Leontiev, he overviews Leontiev and the Kharkov school, including their split with Vygotsky, and provides background on the rift between Leontiev and Rubinshtein (whose category of activity deeply influenced Leontiev). The chapter on Engestrom’s third-generation CHAT covers the problems of mainstream cognitivism, the activity system and activity networks; he also defends the activity system from critiques by Blunden and Bakhurst. Finally, the chapter on the fourth generation centers on Sannino’s work with utopias; it covers the runaway object, utopias, and Sannino’s work with homelessness policy. Lemonie covers a lot of ground in these four chapters, but he does it well, giving us an overview of the concepts being developed in each generation.

Part II brings us to four key concepts in CHAT. 

Chapter 6 covers the object of activity, which is that to which an activity is oriented. Here, Lemonie discusses how Leontiev’s understanding of the object as both an independent thing and “a mental image” (Leontiev 1974, p.11, qtd on p.189) differs from Rubinshtein’s in that it overcomes the immediacy of the stimulus-response relationship (p.189). Since every activity is oriented to an object, the object delimits the activity (p.200). Lemonie then draws on Leontiev, Engestrom, Kaptelinin, and Stetsenko to explore controversies over what counts as an object (pp.202-204). 

Chapter 7 moves to the concept of contradiction, discussing its status as a dialectical concept and providing various examples from the literature. He discusses Entestrom and Sannino’s recent work on discursive manifestations of contradictions (p.223) and draws on Ilyenkov to understand contradictions not as problems but as systemic causes (p.227). And he covers four forms of contradictions (p.234).

In Chapter 8, Lemonie covers expansive learning. (To my surprise, he characterizes my book Topsight as an example: p.250). Throughout, he of course draws heavily on Engestrom’s work.

Chapter 9 focuses on development as qualitative reorganization, drawing on Vygotsky, Engestrom, and Stetsenko.

Finally, in Part 3, Lemonie turns to the interventionist dimension of CHAT. 

Chapter 10 focuses on the Change Laboratory methodology. Here, Lemonie draws on Engestrom and Sannino to baldly state that CHAT is an interventionist approach (pp.300-302). In fact, he quotes Yamagata-Lynch’s 2010 book as saying that CHAT has been applied in North America as descriptive rather than as a tool for changing practice (pp.299-300), a stance that he rejects. Instead, he describes the change labs methodology as one premier way to apply CHAT in the way he thinks it should be applied, as interventionist. He honestly does a great job here, providing a compact, clear, well illustrated discussion of change labs.

Chapter 11 then discusses formative interventions vs. other interventionist approaches: action research, design-based research, and participatory design. These discussions are again compact and thorough. 

Overall, this is a valuable book for those thinking through CHAT and especially for those interested in interventionist approaches. I do have a couple of disagreements with it, which I’ll discuss in a bit of detail, but I want to emphasize that these are small disagreements compared to the substantial contributions this book has made.

The first disagreement is in something Lemonie says about an article I published about participatory design in 2005. (I can’t believe that was 20 years ago.) The article about how PD techniques were picked up and “translated” (in the ANT sense) as they moved from Scandinavia to the US. I used the ANT term “betray” to describe how these techniques were taken up in North America and used in a way that did not accord with their use in the Scandinavian context — the techniques were meant to empower participants in Scandinavia, but were used in the US to improve usability. Lemonie says: ”While he playfully invokes the term ‘betrayal’, he immediately underscores that any form of translation is a form of ‘betrayal’, thereby attempting to weaken the term. However, the idea he defends is that the methods and techniques employed are adapted to the American context.” But “I would argue that this is a betrayal in the sense that, by adapting to this context, PD methods have lost the objective of truly transforming the status quo,” instead becoming tools for productivity “without providing workers with the power to challenge the existing order” (p.374). 

But this is the whole point of the sociology of translation! As Latour liked to say, there is no transportation without transformation, no translation without betrayal — innovations (such as PD’s use of prototyping) don’t just spread unaltered, like particles, but instead are altered as they are taken up. In my article, I show how the conditions under which PD was developed in Denmark (especially the strong unions that had a legislatively enshrined seat at the table with management) simply did not exist in the US, meaning that PD literally could not be taken up in the same way. 

Furthermore, I disagree that Scandinavian PD aimed at “truly transforming the status quo.” Quite the opposite: the status quo was being threatened by rapid computerization, and the research team partnered with union representatives to develop interfaces that could preserve parts of the status quo. Their explicit purpose was to identify tacit craft practices and build interfaces that could still leverage them so that the deep experience of workers wasn’t lost. They feared that without these interventions, that tacit knowledge would become useless and experienced typesetters would be replaced by young computer scientists — which would have been a complete shift of the status quo! To preserve the status quo, the union provided political (not functional) representatives: not average workers, but union reps authorized to stand in for the union as a hierarchical organization. They doubled down on the existing hierarchy.

 I think this point is important more broadly as well. As I argue in Triangles and Tribulations, CHAT itself has been translated many times. To take one example, in Chapter 8, Lemonie discusses how “Engestrom employs the concept of the zone of proximal development in a distinct manner from Vygotsky” (p.258), not as an individual’s growth working in a dyad, but as collective growth as the collective dialogues about potential solutions. Similarly, in Chapter 10, Vygotsky’s principle of double stimulation (which was applied to individuals) is applied to collectives. These follow more generally from the fact that in developing the concept of the activity system, Engestrom moved from Leontiev’s concept (a singular subject) to collective subjects. All of these are translations, picking up an insight and using it differently, and we might consider how Vygotsky might have viewed these translations as betrayals! But if Engestrom had not translated these concepts to make them relevant for larger social groups, CHAT might have languished rather than being the subject of Lemonie’s book. 

And this brings me to the second disagreement. Lemonie goes all in on the argument advanced in different ways by Sannino, Engestrom, and Stetsenko: that CHAT is an interventionist theory (p.300), requiring interventionist approaches (p.302). He argues that the Vygotsky Circle’s method was fundamentally interventionist (p.302) — that is, a fundamental part of the method involves acting on the object to ascertain its properties (p.301) — and methods built on it must follow a four-step process:

1. “analyze the connections between the whole and the parts in the present”

2. ”historicize the analysis by examining the preconditions for the most important present-day connections identified in the first step in the context of past contradictions”

3. “project[] the main social contradictions from the past, through the present, to their resolution and beyond in the future”

4. ”organiz[e] … the actions necessary to achieve the future” (p.304)

In this context, he criticizes Yamagata-Lynch’s “appropriation” of CHAT primarily for descriptive research (p.300). (He might as well be criticizing me as well, since I’ve often used CHAT for descriptive case studies.)

This argument, as mentioned above, has been advanced by various people. I don’t fault Lemonie for repeating and elaborating on it. However, restricting CHAT to interventionist research is a bad move. 

First, we find models for descriptive research throughout CHAT’s history. Leontiev’s forbidden colors experiment is not interventionist — at least, not in any meaningful sense. Leontiev created different conditions in which he did or did not supply colored cards to children doing the task, and did or did not provide minimal instructions. The point was to see how they took up the cards. Similarly, Luria and Vygotsky’s Uzbek expedition was not interventionist: They selected conditions (illiterate, semi-literate, literate) and tested participants to find out whether they could see optical illusions and how they conceptualized groupings of tools and objects. We could go on and on. These might have been “interventionist” in the sense that they used double stimulation, but they generally involved someone with more power (the researcher or a more culturally advanced participant) and someone with less power (a child, an illiterate peasant with the Soviet security apparatus looking over their shoulder). They did not involve guaranteeing agency to the participants. Certainly we can find work by the Vygotsky Circle that might involve collaborating with participants in ways that emphasize their own agency, but this work took part in clinical settings — Vygotsky working in defectology, Leontiev and Luria working in rehabilitation centers. 

Even when we get to Engestrom’s work, we find early studies that are primarily descriptivist. His 1987 Learning by Expanding sets the tone by laying out the activity system and applying it to literary cases such as Huck Finn — describing the case but certainly not intervening in it. His 1990 and 1992 collections again describe cases such as judging and medical work, but generally do not lead to interventions. His 1996 collaborative article with Escalante describes the failings of a kiosk, but does not involve interventions with either the users or the manufacturer. Yes, Engestrom argues for an interventionist orientation throughout this early work, but his publishing suggests that these descriptive cases are themselves legitimate products of research. And as I discuss in Triangles and Tribulations, these early descriptive cases were the models for the CHAT work we took up in North America! 

Second and related, descriptive research is the first step toward intervention. The above four-step process begins with descriptive research — as it should! Without this precursor step, interventionist research is a hammer looking for a nail: it assumes a problem rather than exploring to determine whether one exists. This is an important limiter. Take for instance my 2012 article on coworking. When I heard about coworking in 2008, I became interested in who coworks, why they cowork, and what they thought coworking was. Four years later, I published a CHAT analysis, one that noted ongoing contradictions and speculated on how coworking would develop. If I had started by assuming a problem, I might not have investigated at all, or I might have come up with some highly stereotyped problem based on an incomplete understanding. Fieldwork involves entering the worlds of one’s participants so that one can develop insight about the activity from their view, and without that step, a researcher may not be able to adequately understand the lived problems that they face — or even whether they have problems that they find severe enough to require intervention.

This brings us to the third issue: interventionist research assumes a problem that can be localized, articulated, and solved through intervention. The problem is the nail that the hammer of interventionist research seeks. But there are real drawbacks to seeing the entire world through a problem-solution frame — particularly one that pressures researchers to identify solutions that they and the participants can implement together. That sort of solution suggests a limited scale, budget, and timeframe. (The counterpart to these limited solutions is that researchers often look for a common and familiar root to the problem. Lately for Engestrom and Sannino this root has been capitalism, and the solution has thus been alternatives to capitalism.) This state of affairs can lead researchers to project a basic problem onto the study rather than learning from participants. 

I think this issue is exacerbated by research funding, which tends to already be problem-oriented in interventionist research. Thus researchers come into the research primed to seek a specific problem, the one described in their funding proposal, and guide their participants to a solution that can be achieved by the end of the funding cycle. Consider for instance Sannino’s work with the Housing First project, which begins with the question of how to implement policy dictates, but avoids any questioning of those dictates. The researchers have assumed the institutional goals of the public policy and are focusing on ways to enact them, not question them. 

This is a bit of an irony. I find in my discussions with European colleagues that they tend to think of US research as thoroughly capitalist, serving the values of capital (this is how I interpret Lemonie’s criticism of my PD paper). But I typically don’t seek funding for my studies. When I was curious about coworking, for instance, I just started making appointments to interview people and tour spaces. I didn’t have to articulate a problem, a timeframe, a budget, or a deliverable, and consequently I allowed myself four years to explore coworking, and I published when I felt that I was done. This leisurely pace and wide-ranging exploration would not have been possible if I had accepted the yoke of funding. 

Fourth, in Engestrom’s CHAT, research assumes that problems are primarily in the corners (the mediators: instruments, rules, and division of labor). This is a big shift from Vygotsky and Leontiev, who mainly focused on the individual participants as they took up instruments. As I’ve argued elsewhere, this shift from subjects to mediators reflects the scaling up of scenes of research from dyads to organizations. With dyads, it’s easier to alter the subject (e.g., to teach a student how to use a cultural mediator). With existing groups such as workplaces — whose participants have some autonomy, including the ability to exit the workplace), it’s easier to alter the mediators. To use the example of Housing First again, Sannino reports that some staff at the housing unit were unhappy with changes to rules, so they quit, presumably to be replaced by others who could get with the program. The change lab focuses on establishing agreements — housing agreements, care agreements, etc. — little social constitutions that are established by change lab participants and then presented as an existing agreement to others who later enter the organization. 

Since this interventionist research assumes that the problems are in the corners, solutions tend to alter mediators rather than taking on more fundamental changes such as subjects (are we hiring the right people?), objects and outcomes (are we working on the right thing for the right reasons?), or communities (should we add or cut stakeholders?). That puts some powerful conditioning on the problem space. Compare this to Lemonie’s criticism that PD in the US has “lost the objective of truly transforming the status quo,” instead becoming tools for productivity “without providing workers with the power to challenge the existing order” (p.374). 

So that’s my brief for looking beyond interventionist research in CHAT — a brief that was provoked by Lemonie’s arguments, but really responds to more general CHAT arguments. Despite these minor disagreements, Lemonie has put together an excellent book, and I highly recommend it to those who are interested in CHAT, especially the change labs approach.