Here is how Vygotsky answers your question, Andy, after stating that "word
meaning [concept] is [his] unit of analysis":
"'unit' is a product of analysis that possesses *all* the basic properties
of the whole" (T&S, 1987, p. 46, emphasis in the original).
And he famously illustrated this definition by speaking about the mistake
one makes when using too small a unit of analysis and trying to tell
properties of water by investigating the properties of oxygen and hydrogen.
In my own words, the word "unit", when used in the context of the expression
"unit of analysis" is the smallest aggregate of phenomena I need to consider
in my research to be able to say anything really helpful/useful and
I'm not sure what to make of your " historical accident, or a mistake, or
simply a trivial thing". Why should unit of analysis be any of those? Could
you, please, extend the set of possible choices buy adding, say, a rational
decision (that is, a decision made for an articulable reason).
Oh,.. now I can see, I think. You don't like the traditional divisions which
I seemed to be making while speaking about mathematical discourse,
scientific discourse, political discourse... You even asked whether my unit
of analysis is the same thing as "subject matter". Ok, so no, i'm talking
about *discourses*, which is ontologically quite different than the
(underdefined) "subject matter". And why the "disciplinary" division?
Because these discourses display the kind of inner cohesiveness (not
necessarily in the Halliday's sense of the word cohesiveness) - in their
word use, in their routines and meta-rules, in visual mediators, in their
narratives - that make them stand out as obvious units of analysis. Or, to
put it differently, when I start with a word, such as "number", and am
trying to investigate as much of its uses as necessary to see anything of
importance, I invariably end up, whether I want it or not, with looking at
the whole of formal and informal) numerical *discourse*, in any of its
Did I mange to make myself understandable?
PS. Of course, you may go on and ask what I mean by "the whole" of a
discourse. The boundaries are blurry, and I don't really mean I am checking
every piece of this rather elusive entity. But I do as far as necessary,
never excluding in advance anything that may be deemed as belonging to a
discourse in question.
From: Andy Blunden [mailto:firstname.lastname@example.org]
Sent: Saturday, April 23, 2011 4:22 AM
To: email@example.com; eXtended Mind, Culture, Activity
Subject: Re: [xmca] activity and reification
Anna, no-one took this up, but let me pursue it nonetheless.
I said I think we disagreee about what Vygotsky meant by "unit of analysis."
You concluded your third message in this exchange:
"...to speak about it as the use of word in discourse (not just a
single act, Andy; rather, a discursive activity with the word) ...
discourse (understood as a specific type of communication) is what
may usefully be taken as a unit of analysis in developmental (and,
obviously, historical) studies."
Leaving all other issues aside (I actually agree with most of what you said
in this message in response to Martin), what do you make of the word "unit"
in the term "unit of analysis"? Do you see it as a kind of historical
accident, or a mistake, or simply a trivial thing? I take it seriously, you
Do you see what I am getting at Anna? You seem to use the term to mean