RE: [xmca] Correct unit of analysis for human activity (in games)

From: Helena Worthen (
Date: Fri Mar 10 2006 - 12:45:24 PST

Hello, Patrik (and Bill, and others):

Yes, it can have a large effect on a study if one chooses to shine the light
on a whole activity system as compared to the activity of a few individuals.
It can, for example, have the practical effect of expanding your research so
your work becomes unmanageable.

But if you are working from the premise that all activity takes place within
a system, you don't have to shine the light on all parts of the system. You
can choose the part of the system where what is most interesting to you is
happening. Your research, after all, is also object-oriented (purposeful)
human activity, and you need to keep your eye on your purpose.

When I read the three definitions of the unit of activity that you list
below (Mike's, James Wertsch's, and Yrjo's) they all sound as if they're
talking about the same thing. That's remarkable, I think, considering how
differently they are expressed. But when I read them all, I nod and say,

Actually, Mike's is kind of a cadenza played on the other two.

I take the idea of the "prime" or "correct" unit of analysis to mean the
smallest bundle of things that you'd want to be sure you were keeping in
mind when thinking about something. Like a molecule: "the smallest particle
of a substance that retains all the properties of the substance..."

Here's yet another description of the unit of analysis for understanding
human activity. It's from the introduction to Sawchuk, Duarte, and
Elhammoumi's new book, Critical Perspectives on Activity: Explorations
across education, work and everyday life. (Disclosure: I've got a chapter in
there, as does Yrjo and Paul Adler and others.)

"Activity in this tradition is not used in the everyday, common-sense way,
however. Rather, it is a specialized and, in fact, highly contested concept.
To begin with, it is defined as the minimal unit of analysis for the
understanding of cognitive development, human participation, and change. It
inherently contextualizes practice in cultural and historical terms. It is,
in our view, the most comprehensive analytic framework for analyzing human
practice and learning currently available. At its heart it affirms that all
human practice is mediated by symbolic, cultural and communal, as well as
material, resources or tools: It is through these forms of mediation that
human practice is understood as both dynamic and historical. This conceptual
approach allows important, integrated forms of analysis (p2)"

Note the emphasis on the purpose of an analysis. If you are interested in
studying change (for example, how some people learn something) you need a
minimal unit of analysis that conjoins practice (what is being done), and
culture and history. That doesn't mean that you have to do a full
ethnography or write the history of what you're looking at into your
dissertation. But if you skip an important aspect of either one of those,
your research will come out looking funny.

In my opinion, the aspects of culture and history that are most frequently
skipped are power imbalances related to race, gender and economics.

Helena Worthen
Chicago Labor Education Program
University of Illinois

-----Original Message-----
From: [] On
Behalf Of bb
Sent: Friday, March 10, 2006 10:13 AM
To: eXtended Mind, Culture, Activity
Subject: Re: [xmca] Correct unit of analysis for human activity (in games)

I could argue that, given the multiple timescales of the human condition
(lemke), and how culture plays a role in psychology, crossing timescales,
(e.g. the description of prolepsis in 'cultural psychology"), and the
description by Leont'ev of the "hierarchical structure of activity", plus
historical analysis included in Engestrom's expansive methodology, that one
would abductively think multiple units of analysis, crossing at least many
timescales, if not also some scales along other dimensions, could be
desireable and even practicable. This methodological potential seems
embedded in the CHA of CHAT.

Practically, multiply scaled units of analysis are useful only if such
methods coincide with the goals of the intended study.

I'm sorry I cannot add more -- it's just a passing thought, a humble
And I have no formal training, so take it for what it's worth.


On Friday 10 March 2006 3:55 am, Patrik Bergman wrote:
> Hi,
> Perhaps you have all gone through this debate before, but I will give it
> try: What do you consider being the correct unit of analysis for
> understanding human activity? I am reading the works of Yrjö Engeström
> quite a lot now and he states that the prime unit of analysis is a
> collective, artefact-mediated and object-oriented activity system. Others,
> such as James Wertsch, state the correct unit is an individual acting with
> mediational means. Meantime, Mike has for example stated that “Mediated
> action and its activity context are two moments of a single process, and
> whatever we want to specify as psychological processes is but a moment of
> their combined properties.”, which to me sounds like something in the
> middle of Engeström’s and Wertsch’s descriptions.
> PhD student that I am, I might be totally off track here, but it would be
> interesting to hear your views on this since I am thinking about this in
> relation to studying the online game World of Warcraft as a learning
> environment. As far as I can see, it can have large effects on a study if
> one chooses to study individuals and their meditational means, compared to
> also incorporating the whole activity system (including its history).
> Best,
> Patrik Bergman

xmca mailing list

This archive was generated by hypermail 2b29 : Sat Apr 01 2006 - 01:00:13 PST