Generalizations

 

Generalizations in Activity Theory: origins, process, typology and methodological implications

by Marco Antonio Pereira Querol

mapquero@gmail.com

Understanding the process of generalization has both practical and methodological implications. From a practical point of view, knowing what a generalization is and how it is produced is essential to be able to foster it. From a methodological and scientific point of view, it is important to understand the different types of generalizations that exist and the process by which they are generated, in order to align them with data and analytical methods that ensure scientific validity.

This post seeks to answer the following questions: What is the origin of knowledge in Activity Theory (AT)? How are generalizations stored and transmitted? How are they produced? What types of generalizations are there? What kind of generalization does a Change Lab (CL) seek to produce? And, finally, what are the methodological implications for the generalizations produced in a CL to be scientifically valid?

AT is based on a distinct epistemology and seeks to produce a specific type of knowledge, or generalization, called theoretical generalization. The way of producing this scientific knowledge differs significantly from other perspectives, which can generate uncertainties in researchers from different backgrounds. I begin by discussing the epistemology in AT. Below I only repeat and synthesize the ideas of other authors, for a complete view of generalizations in Activity Theory see Pihlaja (2005).

Epistemology in AT

Epistemology is a word used by philosophers that refers to the study of the origin of knowledge. Therefore, epistemology in Activity Theory means how AT understands where knowledge comes from, how it is created and formed.

In Activity Theory, the origin of knowledge is human activity, in which the subject transforms the object. Knowledge is generated from practical actions in which the subject transforms his environment.

The philosopher Ilyenkov (1977) proposes a path to understand the emergence of human thought and the process of creating knowledge and concepts. According to him, the first step occurs when the individual is faced with an object (a thing or object still as raw material). Then, the subject acts on this object, exploring it, measuring it, evaluating it (an action). From this practical interaction, an idealized projected mental image of what the object can be is formed (a concept that may or may not be expressed in a word). In the sequence, the individual returns to act on the object (an action), now guided by this concept, transforming it into a concrete result, a product of work (a thing, or object as a result or product). This path can be summarized by the formula: thing → action → word → action → thing. Although the actual process itself is much more complex and does not always follow this linear order, Ilyenkov's central idea is that the concept or idea only arises from the object and the subject's practical interaction with it, and cannot precede practice.

To transform an object raw material into a product or result, the subject considers both its external and internal properties — those that enable its transformation. These properties are not revealed without the transformative practice or outside of it, and require a certain reflection.

The object is only transformed through the application of what Davydov (1990) calls certain modes of activity. By repeating the process of transforming the object several times, individuals discover what are the essential modes of activities, that is, the conditions necessary to transform it – revealing its properties.

The process of discovering the object's properties takes place in two main circumstances. The first is the reproduction of the object's transformation process occurs repeatedly, occurs in more or less variable external conditions and situations. That is, each time the object is reproduced, the conditions may vary. The second is that the transmission, from one person to another and from one generation to another, of knowledge about the modes of activity that make this transformation possible, uses models and patterns. These two circumstances require individuals to identify and establish only the essential and really necessary conditions to reproduce a given form of movement of objects. Incidental conditions are "filtered out", leaving only those that actually and necessarily determine the modes of activity.

Signs, words, and tools as generalizations

By repeating an operation several times under varying conditions, the individual discovers the properties, conditions, and modes of activity that lead to the movement or transformation of an object—that is, it produces a generalization. This generalization, at first, is mental and cognitive, remaining within the individual's internal sphere. However, faced with the need to transmit it and cooperate, the individual creates signs or words that represent it. Therefore, both signs and words are, as pointed out by Vygotsky (1978), generalized reflections of reality — that is, generalizations used in everyday life.

Signs and representations are not only products of an action or experience, but also mediators in the very process of its construction. In other words, generalizations are used to produce new generalizations—that is, the generalization process itself is mediated. Human beings are not "empty boxes": from birth, they absorb signals and tools that mediate not only their external actions to transform the environment, but also their thoughts. Human perception resorts to previous knowledge, of both individual and social origin. We use concepts, models, and theories to interpret the world and make our own generalizations. Among human beings, a generalization takes the form of a cultural artifact, transmitted from person to person and from generation to generation. In this process, it evolves and develops.

Leontiev (1933) points out that generalizations are not only present in signs and words, but are also incorporated into tools. Tools implicitly carry knowledge about what they are, how they work, what properties of the object allow them to be transformed by them, and how this transformation occurs. To learn to use a tool is to learn to perform the operation that it was created to mediate (Leontiev, 1933).

In short, tools as well as signs and words carry generalizations of operations, events, or objects. These generalizations mediate both communication actions, which enable cooperation, and direct actions to transform the external world and internal cognitive actions related to the individual's own thinking, reflection, and self-control. Generalizations not only originate from human activity, but also develop in it. Without generalizations, communication, cooperation, or the use of tools would not be possible.

Type of generalizations

Perceptual-functional generalizations

As proposed by Luria (1976), the type of generalization predominant in a society or community depends on the dominant mode of activity in which its individuals are involved. In societies where handicrafts and manual work prevail, perceptual-functional generalization tends to predominate, in which ideas and concepts are linked to a specific practical and concrete situation. In this type of thinking, an object is always associated with the practical function it performs in a given context. For example, the hammer is associated with the function of driving a nail by being attached to nails and wood — rather than being categorized as "a tool" in the abstract sense.

Compared to perceptual-functional thinking, in more abstract thinking, the object is removed from a specific situation and inserted into a general category, such as "tools" or "sharp objects", or even classified according to attributes such as general form or function. Abstraction allows generalization to be applied to different cases, enabling the individual to establish logical relationships independent of a practical context. On the other hand, situational generalization is more specific to a given circumstance: "in a situation X, one must do Y". This type of generalization is usually incorporated into habits, automated operations and transmitted through gestures, speech or images.

Perceptual-functional generalization would tend to predominate in societies with more artisanal work, characterized by the practical manipulation of objects, while more abstract generalizations would be more frequent in technologically advanced societies, with a more complex division of labor and exchanges mediated by money (Pihlaja, 2005).

Empirical generalizations

Davydov (1990) points out that, in industrialized societies, with greater division of labor and formal education, the mode of empirical generalization tends to predominate. This type of generalization is characterized by abstraction from visual and observable characteristics of an object or phenomenon. Empirical generalization is centered on immediate, observable, and measurable reality, interested in aspects such as number, quantity, quality, properties, and measurement. It is therefore based on directly perceptible sensory data.

Empirical generalizations are based on the observation and comparison of the external properties of objects. When observing an object, the individual identifies similar properties and, from this, creates a class of objects with common characteristics. This process is based on finding and isolating certain stable and recurring properties of this class of objects (Davydov, 1990).

The process of forming an empirical generalization involves identifying something invariant in objects or processes and assigning a word to that invariant. This common characteristic, once named, is used to recognize objects within a given set.

For example, an individual observes a series of objects and notices that they all have a rounded shape. The invariant, in this case, is the rounded shape, which is given a name to designate the category — for example, "balls".

The content of an empirical concept corresponds to the common characteristics present in a class of objects. This type of generalization makes it possible to classify, systematize and establish connections between phenomena or beings. In addition, it allows associating successive events and establishing cause-and-effect relationships through similar methods (Pihlaja, 2005).

As Pihlaja (2005) points out, this type of generalization follows a path from the concrete to the abstract, in which, based on observation, a linguistic abstraction is elaborated. Here, "concrete" refers to an object that can be directly observed, while "abstract" refers to aspects or characteristics common to various objects.

The creation of empirical generalizations follows inductive reasoning: one starts from one or several concrete cases to later construct an abstract logic that groups objects into classes.

It is important to highlight that empirical generalizations are not limited to descriptions of something; They can also include explanations. However, these explanations are based on observations of sequential events and on cause-and-effect relationships, i.e., causality in a sequence, determining from empirical observation modes of activity that allow the movement or transformation of an object. One of the ways to achieve this is through the establishment of causal relationships between events.

According to Engels, quoted by Davydov (1990), the observation of a sequence of events, one after the other, does not constitute, in itself, proof of causal connection. The only way to demonstrate such a connection is through experimentation. By deliberately repeating a sequence of events, it becomes possible to eliminate accidental or occasional connections, which may occur by coincidence but are not part of the causal relationship. Variability is essential for learning to occur. Marton (2000) points out that variation allows the subject to distinguish what is essential and what is irrelevant. Thus, the individual is able to identify the internal, universal and essential connections between the observed events.

Experimentation in scientific activity recreates the conditions necessary to produce a certain movement or transformation, collects data and analyzes them statistically. Experimentation is not exclusive to science: human activity and work itself have, in their essence, an experimental character. By recreating, through an action, the conditions of reproduction of a movement or event, the individual is able to discover the properties responsible for its occurrence. In this way, a generalization is formulated—a kind of universal law: whenever such conditions occur, the movement or event will be produced.

Davydov (1990) also draws attention to the fact that the process of experimentation is always part of a broader social context; that is, objects and events are part of the individual's experience based on social needs.

Theoretical-genetic generalizations

A theoretical-genetic generalization identifies the general principle that explains the emergence or production of a phenomenon (Pihlaja, 2005). It is elaborated from the analysis of the internal relations of the structure of a system, seeking to identify the essential relationship that leads to its manifestation. This type of generalization is constructed through the transformation of objects, allowing us to understand how certain structural elements are articulated to generate a phenomenon. Table 1 compares the main characteristics of a theoretical generalization and an empirical generalization.

Table: Comparison between abstract-empirical and theoretical-genetic generalization (Pihlaja, 2005)

Abstract-empirical generalization	Theoretical-Genetic Generalization
It is produced by comparing objects and their representations.	It is produced from the analysis of certain relationships within a structured system.
It identifies the formally general ownership of particular objects in order to assign them to a particular formal class, regardless of whether or not they are connected to each other.	It identifies the essential relationship between things, which constitutes the genetic basis of all other manifestations of the system.
It is based on observation, reflecting the external properties of objects and relying completely on visual conceptions.	It is based on the transformation of objects, reflecting their internal relationships and connections.

 

In Activity Theory, an example of theoretical-genetic generalization is the identification of the contradictions that gave rise to the system and guided its development, as well as the fundamental principles that made it possible to resolve these contradictions. This unit — formed by the contradictions and the principles that solve them — is called the germ cell, because it represents the smallest basic unit that contains, in condensed form, the essential elements of the system.

The Method from Abstract to Concrete

In everyday usage, "concrete" is understood as something that can be perceived, felt, measured, or touched, while "abstract" is that which exists only in the mind. In materialist dialectics, however, these concepts take on another meaning: concrete refers to an entire system, interconnected and contradictory by nature; abstract is something extracted or isolated, which represents only an aspect or part of a whole (Ilyenkov, 1982).

To understand an element of a system, it is necessary to analyze it in relation to the total system of which it is a part, understanding the role it plays and knowing the historical process that led to its emergence and its current form. Thus, understanding the emergence of an object requires reconstructing the events, contradictions, and specific relations that gave rise to it.

In materialist dialectics, the method from the abstract to the concrete is used to produce theoretical-genetic generalizations. It is a method that involves a historical analysis of an object or phenomenon, in which the researcher seeks to understand its development from the original contradictory relationship – or germ cell – initially single and isolated (and, in this sense, abstract), to its current mature and complex form.

As Miettinen (2000) points out, the study begins with the analysis of the system as a whole – that is, the concrete one – investigating its functioning, its elements and its manifestations. From this global view, the basic germ cell (for example, the contradictions and the principles that resolve them) are abstracted. Then, it returns to the whole, now progressively enriching the abstraction, incorporating more specific elements, phenomena and relationships derived from the germ cell.

Summary

We started this post by talking about epistemology in Activity Theory. From this perspective, the origin of knowledge is in human activity: it is through the transformation of the environment that knowledge is produced. This knowledge about the world, called generalizations, is transmitted from one person to another through language.

Signs and words, as well as tools, function as deposits of generalizations of reality. As they are passed down from person to person and from generation to generation, these generalizations transform, evolve, and develop.

Next, we saw that there are at least three types of generalizations: perceptual-situational, empirical-abstract, and theoretical-genetic. Perceptual-situational generalization refers to knowledge directly linked to a specific practical and concrete situation. For example, a description of something based on the practical function it has in an operation, action, or activity. Empirical-abstract generalization, in turn, involves classifications, descriptions, categorizations, or explanations produced from direct empirical observations, and is elaborated through processes such as quantification, measurement, and qualitative or quantitative description of a phenomenon. Finally, theoretical-genetic generalizations aim to understand the internal relations of a system that led to the emergence of an object or phenomenon, that is, to explain why and how something emerged from its internal contradictions.

Generalizations in Activity Theory and the Change Lab

In short, it can be said that AT is a set of concepts and theories that allow us to discover the internal properties — the essential internal relations of a system — that allow the transformation of an object or phenomenon. In other words, it is about understanding how and why a system became what it is, investigating its origin and the forces that transform it.

What kind of generalization does a Change Lab (CL) seek to produce? The answer is: it depends. If the objective of a CL is to promote expansive learning — that is, to expand the object-purpose of the activity and generate a new model — it is essential to understand the historical contradictions and the principles that aim to solve them, that is, to identify the germ cell, which constitutes a theoretical-genetic generalization.

However, it is not a question of judging which type of generalization or method is superior or more desirable; Both are necessary, as they complement each other. Empirical-abstract generalizations need theoretical-genetic generalizations to understand the reason for what is being observed. In the same way, theoretical-genetic generalizations depend on empirical-abstract generalizations to enrich themselves and address the concrete. Thus, the question is not which type of generalization to use, but how to integrate them in a complementary way.

In CL, both types of generalizations and methods are employed. Generally, it begins with empirical generalizations, based on the observation of the activity and the analysis of concrete data. Then, historical analyses are carried out to produce abstractions that allow us to understand the contradictions and identify principles to solve them — a process that corresponds to a theoretical-genetic generalization. To enrich the model, empirical-abstract analyses are developed. From them, a new abstract model is built, which is gradually enriched and concretized.

Methodological implications

What methods should the researcher use in a research based on Activity Theory? Obviously, the method depends directly on the research questions. For example, if the objective of the research is to identify the types of manifestations of contradictions in an activity or the types of actions that the agents perform, one tends to produce empirical-abstract generalizations. On the other hand, if the research seeks to understand why a system of activity is the way it is, analyzing its internal contradictions and its possibilities of development, the result will be a theoretical-genetic generalization, and a method such as a historical analysis will be necessary.

In general, it can be said that the greatest potential of an CL and of Activity Theory lies in the production of theoretical generalizations capable of explaining why an activity system is the way it is — that is, of revealing the historical contradictions that configured it and that guide its development. From the methodological point of view, this knowledge emerges only during the process of transformation of the object. Hence the relevance of longitudinal data analysis, which may include, for example, historical analysis, the identification of disturbances, innovations and initiatives, as well as the examination of social interactions and discourses.

The scientific validity of the theoretical-genetic generalization produced from Activity Theory and ML lies in the in-depth analysis and understanding of the principles underlying the activity system. Its greatest potential lies in producing an understanding of the origin and trajectory of this system, as well as in enabling the construction of a new model for its development. To this end, experimentation is promoted through real-life interventions.

Interventions thus become a central strategy for reproducing transformation processes that would otherwise be rare or difficult to record under natural conditions. They allow, to a certain extent, to test hypotheses, control variables and differentiate redundant external conditions from essential internal conditions. The intervention is configured, therefore, as a form of experimentation in reality, capable of generating knowledge through its own transformation. The resulting generalizations empower interventionists and practitioners, allowing them to collectively design a more sustainable future for their activities.

References

Davydov V.V. (1990). Types of generalization in instruction: Logical and Psychological problems of Structuring School Curricula. Reston, Virginia: National Council of Teachers of Mathematics.

lyenkov, E.V. (1977). Dialectical logic: Essays in its history and theory. Moscow: Progress.

Ilyenkov, E.V. (1982). The dialectics of the abstract and the concrete in Marx's Capital. Moscow: Progress.

Leontyev, A.N. (1933). Notes on consciousness I and II. Multidisciplinary Newsletter for Activity Theory. (3–4), I–VIII.

Luria, A.R. (1976). Cognitive Development, its Cultural and Social Foundations. Gambridge: Harvard University Press.

Pihlaja, J. (2005). Learning in and for production: An activity-theoretical study of the historical development of distributed systems of generalizing (Doctoral dissertation, Helsingin yliopisto).

Marton, F. (2000). Variatio est mater studiorum. Magyar Pedagógica, (100/2),127–140.

Miettinen, R. (2000). Ascending from Abstract to the concrete and Constructing a Working Hypothesis for New Practices. In V. Oittinen (Ed.) Evald Ilyenkov's Philosophy Revisited. Helsinki: Kikimora.

Vygotsky, L.S. (1978). Mind in Society: The development of higher psychological processes. Cambridge: Harvard University Press.