Why Use These Theories?
My object of research, the academic interview, exists within
the at least partially definable ecological space of academia. For this reason, theories that hold
ecological views about social action are likely to be relevant. Cultural-historical activity theory (CHAT)
and its predecessor, activity theory (AT), are two reasonable choices. Arguably, actor-network theory would also be
a valid candidate, but a secondary motivation for me in choosing the activity
systems theories is to examine how the first generation theory, AT, compares
with its more elaborated version, CHAT.
The fact that both are still used as theoretical frameworks makes it
reasonable to examine them side-by-side even though, in fact, CHAT evolved from
AT. I am also interested in examining
Paul Prior and his fellow authors’ (2007) attempt to adapt CHAT as rhetorical
theory. Because Prior et al’s version
turns out to be something of a Franken-theory with elements that are
challenging to translate into an analysis, I will not apply it. However, I will make a few comments on it at
various points.
About the OoS
The
published academic interview, my object of research, can be found in a number
of places. The most obvious is in
academic journals. A recent search, for
example, quickly turned up two 1995
interviews of Charles Bazerman, one in Composition
Studies (Crawford, Smout, & Bazerman) and one in Writing on the Edge (Eldred & Bazerman). Interviews also receive book-length
treatments. An example is Conversations with Anthony Giddens, composed
of seven thematic interviews with the prominent sociologist (Giddens &
Pierson, 1998). Another is An Invitation to Reflexive Sociology,
which is described as “Pierre Bourdieu’s ‘discourse on method’” (Bourdieu &
Wacquant, 1992, back cover) but has as its second and longest section an
interview of Bourdieu by Wacquant. Finally, it is possible to find academic
interviews that are primarily distributed online. Online journals can include interviews, such
as McNely’s (2013) interview of Spinuzzi as described in my first case study. It
is also possible to find interviews online that are transcriptions of oral
interviews, such as, for example, the 18 visiting speaker interviews included
in Florida State’s Rhetoric and Composition department site.
In spite of the fact that, as we
can see, the interview is a robust form that is published in different media
and is distributed in several ways, very little research has targeted this
genre. One of the few is Arnold’s (2012) look at Michel Foucault’s interviews
as a particular genre within his larger body of work. Although Arnold doesn’t examine the academic
interview as a genre, she does examine the ways that Foucault used the
interview as a space to make theoretical statements that transcended his
specific published works. Another look
at a specific scholar’s interviews is Chakraborty’s (2010) discussion of how
Gayatri Spivak’s interviews brought together “the scholarship with the persona”
(p. 623). Chakraborty also examines the way that such interviews allow public
intellectuals to “use the interview as a modus vivendi to produce intimate interlocution
with their chosen constituency” (p. 627). In both Arnold’s and Chakraborty’s
discussions, however, the focus is primarily on the scholars rather than the
genre. In terms of looking at how genres
are situated, Tachino’s (2012) study of intermediary genres might serve as a
model. Though Tachino is looking at a completely different activity system,
namely the judicial system, the concept of “a genre that facilitates the ‘uptake’
of a genre by another genre” (p. 455), almost certainly applies in the case of
the academic interview. One outcome of
an academic interview is almost certainly increased uptake of a scholar’s
published works. A similar case is the
use of published interviews as a mediating genre for avant-garde artists to
discuss their art, as discussed by Bury & Scott (2000).
Mediation: A Core Concept
The idea of an intermediary genre
leads us to the idea of mediation, a concept at the core of both activity theories,
AT and CHAT. As Russell (1995) notes, activity
theory owes a significant debt to Russian psychologist Vygotsky. Vygotsky emphasized that “the structure and
the meaning of the tool in a community… strongly influence the actions that people
accomplish, and as such the cultural tool is a strong semiotic component of the
learning process” (van Oers, 2010, p. 8). Since Vygotsky’s theories were connected to
learning, unsurprisingly they have frequently been used by educational
theorists. However, because “activity
theory analyzes human behavior and consciousness in terms of activity systems,”
or in other words, “goal-directed, historically situated, cooperative human interactions”
(Russell, 1995, p. 53), it is obvious that these theories can be applied to many
settings. For example, activity theory
(AT) has been employed by at least two writing scholars Russell (1995) and
Nowacek (2011). Both use the “mediated
action triangle” that Yamagata-Lynch (2010) identifies with first generation
activity theory. However, Yamagata-Lynch
points out that “many CHAT scholars now encourage investigators to engage in
new work within an interventionist framework using third generation activity
theory” (p. 23). This version correlates
with the framework that Spinuzzi applies in his 2003 study. The idea of mediating artifacts continues as
a Vygotskyan core concept that remains at the heart of both versions.
When applied to my object of study,
the interview genre fits slot of “tool” or “mediational means” in activity
theory and of “instrument” in cultural-historical theory. In other words, it is
posited as a core node in each case.
Incidentally, the central role of a mediating tool or artifact is
obscured in Prior et al’s theory of CHAT because mediators are not nodes in this
version. In fact, the authors discuss
mediation as a concept but do not include it explicitly in the conceptualization,
“Remapping Rhetorical Activity: Take 2.”
Instead, they argue, “We take mediated activity and mediated agency as
fundamental units of analysis. In those terms, everything in the three maps
(literate activity, functional systems, and chronotopes) is about mediation”
(p. 22). Clearly, this
mediation-is-everywhere stance is a significant departure from the more focused
version of AT/CHAT. To be honest, many terms
in their list, such as communities, ecologies,
activity, and socialization, are
so general that it is hard to imagine the idea of mediation having the
theoretical potency that it has when it is associated with a tool or
instrument. While it is true that in a
sense, these abstract categories are mediating, in another way, the conceptualization
moves mediation from node to connecting channel. This idea of connections as inherently
mediating draws Prior et al’s version of CHAT closer to actor-network theory. At the same time, the theory lacks the
precision of ANT because Latour (2005) insists on careful tracing and labeling
of connections. To see abstract categories as mediating seems to fall into the
everything-is-social pit that Latour labors so hard to banish.
How the Network is Structured
The network formulation associated
with AT and CHAT, i.e. the nodes, structuring of nodes, agencies and
relationships within and between nodes, is where the first generation theory
(AT) diverges from the more elaborated version (CHAT). I will take each in turn. Activity theory is
usually represented by a triangle with the “mediational means” at the top, the
“subject” on the left, and the “object” or “objective” on the right. An additional arrow emerges from the corner
of the triangle pointing towards “outcome(s),” meaning that human agents use
mediational means to work towards goals in the activity system, and this should
generate hoped-for outcomes. There are a
couple of implications here. First, it
is the human agents who have agency.
Tools are used intentionally by these agents to work towards conscious
results. Other objects and proximate
groups have little relevance within this model. However, subjects, tools, and
objectives are assembled together in the drive towards outcomes.
Since in my application of the
theory the academic interview is the mediational means, or tool, as we have
said, the interviewer and interviewee work together constitute the subject, the
agents who work together to accomplish a goal of illuminating a scholar’s
ideas, motivations, influences, personal idiosyncrasies and other interview
content. The hoped-for outcome, one imagines, would be
the increased circulation of the scholar’s ideas within the disciplinary
community. Of course, there may be
supplementary outcomes, such as reinforcing the values of a discipline,
educating interested parties outside the discipline, mentoring disciplinary
novices, and so forth.
Although Spinuzzi (2003) describes
his framework as “activity theory” (p. 36), the diagram on page 37 demonstrates
that his version bears a great resemblance to what is usually described as cultural-historical activity theory, or
CHAT. In an important expansion of
activity theory, Engeström (1987), one of the theory’s main proponents, added
an additional row to the bottom of the original triangle to draw in more of the
ecological context of the activity system.
This is what is now typically known as CHAT. The first additional node is rules, followed by community, and finally, division
of labor. These are stretched out to
make up the bottom row of a larger triangle, with lines drawn between each
point. Spinuzzi’s version has two minor substitutions,
namely, to replace subject with collaborators,
and rules with domain knowledge. However, the appearance of the diagram is
altered. Instead of embedded triangles,
each node is pulled out to constitute vertices in a hexagon. Since both diagrams connect the points to
multiple other points, the result is not fundamentally different, but it does
suggest that Spinuzzi wants points to be visually and logically equivalent. This contrasts with Engeström’s
version where the arrangement might imply that in some way the subject is
caught between rules and mediating tools, for example.
To apply the model to academic
interviews, we add to the previously-mentioned tools, subject, and object, the
domain knowledge of the discipline within which the interview takes place,
which is normally also the discipline of the scholar being interviewed. The content knowledge and research objects of
the discipline would be included here as well as, arguably, the discursive
formulations of the discipline in a Foucauldian sense, that is, what can be
talked about and how. The community
clearly would be the disciplinary practitioners and other scholars working
within the discipline. The division of
labor can capture the point that one “collaborator,” to use Spinuzzi’s term, is
the scholar being interviewed and the other is the interviewer, usually
another, less established scholar in the discipline, who, although named, is
nevertheless rendered less visible in the greater light of the luminary being
interviewed. Perhaps, too, we can extend
the division of labor to include other actors in the publication process, such
as journal editors, archivists and webmasters and so on. If this is the structure of the network, we
can see that agency has now been dispersed since collaborators and community
both involve human agency. Of course not
all human actors have the same level of agency; the community arguably has less
impact on the activity system than the collaborators per se.
In an important sense, CHAT can capture more of the ecological
framework that makes up the activity system.
After all, when we analyze the interview genre within the activity
system, we should recognize that the system entails more than just scholars
talking to other scholars. In a way, the
entire institution of the university is activated with all its moving parts,
from financial aid to student support services to plant services. Scholarship depends on the educational
enterprise that defines the scholar as professor and enables the scholar as thinker. The educational enterprise in turn depends on
a complex of supporting staff, facilities and equipment. The interview captures the professor as
scholar. Although the professor also writes
in an array of other genres: exams, syllabi, annotations on student papers,
emails, proposals, catalog descriptions, textbook adoption forms, expense
reports, and others, the interviewer does not ask the interviewee to comment on
these. The genres that are most highly valued
and frequently discussed are research articles and books. The interview may, however, mention
dissertations and conference presentations as well as genres of public
engagement such as blogs, TV interviews, newspaper columns, etc. By calling the larger ecology into play, CHAT
allows us to recall these other professional activities, whereas activity
theory’s first generation model largely narrows the focus to the interview itself
with a nod to the larger, though unelaborated, activity system.
Although in one way Prior et al’s
version of CHAT offers an even more powerful model than the two versions
mentioned so far, it also has some significant disadvantages. The “functional systems” part of the model
overlaps to a degree with CHAT, as it includes people, artifacts, practices, institutions, communities, and ecologies. In addition, the “literate activity” list,
includes production, representation,
distribution, reception, socialization, activity, and ecology. The purpose of the “literate
activity” list is to replace the five terms of the classical canons of rhetoric
with a more effective model. This is a
laudable goal that certainly extends the reach of the theory. The disadvantage is that it is hard to see
where literate activity fits into the larger model. If we were to adapt Spinuzzi’s diagram, for
example, where would the literate activity fit?
In the middle of the hexagon?
Should it be seen as an elaboration of activity in which each step
activates all parts of the hexagon? To
complicate matters, we have the concept of the “laminated chronotopes,” a term
that is not well-defined in the article.
Chronotopes are introduced as being “in the broadest context” and at the
same time, “embedded in semiotic artifacts” (p. 19). Does this perhaps mean that the application
of the theory is like the embedded Russian dolls where the artifact examined has
nested within it the whole universe of the activity system that created it,
within which is another artifact with its
activity system inside and so on down the line?
The biggest problem for me, however, is the “functional systems” list
which replaces Spinnuzi’s hexagon or Engeström’s embedded triangles with a
loose list of items that are not defined in the article. Prior et al claim that the “CHAT map points
to a complex set of interlocking systems within which rhetors are formed, act,
and navigate” (p. 22). While complex interlocking
systems and a “perspective that integrates communication, learning, and social
formation… as simultaneous, constant dimensions of any moment in life” (p. 23) are
intriguing, they may make the model too complex and too comprehensive to be meaningfully
applied to any single research project.
Movement and Change
To return to network structures
that we have established for AT and CHAT, we need to consider the network as a
dynamic system. What moves? How does it move? How does the dynamic nature
of the network allow the network to change or evolve? Within the AT triangle, as already stated,
there is a forward trajectory where the elements of the activity
system—subjects using tools to accomplish objectives—work together to move
towards outcomes. Considering that we
are applying the model to a genre of writing, the action here is rhetorical
action. What is moving through the
network is the actions taken in time to accomplish the goal and reach the
outcomes. As in a physical system, there
are forces driving results. The best candidate for these forces is the agency
of the subjects employing the genre as a tool.
Since we have already established that the basic system of AT remains
within CHAT, this dynamic also applies to the CHAT network. The difference is that the actions that move
through the system, the forces of agency, now have more internal structure,
since the knowledge or rules that apply, the larger community that may
influence the actors (“collaborators”, again to use Spinuzzi’s term), and the
structural system within the community or organization that does the acting
(“division of labor”) all leave their traces on the packaging of the action as
it moves forward. It is easy to see how
these elements shape the academic interview.
At times, as when an interviewee mentions scholarly influences or describes
the development of ideas within a discipline, these elements may emerge within
the text of the interview. In both
activity theory models, I see the network as operating like a machine that is
activated to package and produce a rhetorical artifact or event. If we persist in seeing the genre as the tool
or mediational means, then we may see the text as the distributional outcome of
the work of the network. In this
conceptualization, the content or the meaning comes to exist when it is package
and moves forward as a stable artifact.
What happens as the text emerges from the activity system, as it moves
into “outcomes,” is less certain.
Presumably it enters a new activity system to enter the community and be
taken up as domain knowledge, or even, depending on its content, to argue into place
new instruments, changes to community, adjustments to divisions of labor and so
forth. If this understanding is correct,
we have to assume that the activity system lies within a larger universe of
activity systems. Perhaps this is what
Prior et al are trying to achieve with the separation of ecology (the single activity system that packages a given text?)
and ecologies (an interlocking
network of activity systems?) Incidentally,
Prior et al suggest that what travels through the network is not only action,
but cognition. “Mediated activity means
that action and cognition are distributed over time and space” (p. 17-18). Particularly in terms of rhetorical activity,
we might argue that what moves here is cultural capital and knowledge, to be
repackaged by human agents. As Prior et
al point out, “It’s about attending to semiosis in whatever materials at
whatever point in the activity” (p. 23).
In each model discussed, action
theory argues a dynamic system, one that allows recurrence as well as
change. Although human actors tend to
repeat things that work, circumstances vary, allowing minor shifts to occur
even when a genre, for example, is redeployed.
What would drive larger changes to the network would be changes within
the ecology. As time passes and as each
element of the system evolves or renews itself—either as a result of what
happens within the activity system under discussion or as a result of other
activity systems in the society—the system will shift slightly in new
directions. If we use the first
generation activity theory as a model, we can imagine an interviewer and
interviewee coming together to produce an interview. The interview is published, and it inspires a
graduate student to apply the ideas to another object of study. By the time the study is published to
significant acclaim, more journals have shifted to online formats. When the graduate student has become a noted
scholar and is interviewed, the network has changed. The subjects are different people, the online
nature of the interview has changed the genre as a mediational means, and,
although here the object is largely the same as in the earlier interview, the
outcomes may well be different. What
remains stable is how things tend to work within activity systems. Human beings still do accomplish things using
mediating tools.
An Evaluation
It is no accident that activity
theories have been attractive to genre theorists. Spinuzzi’s 2003 study, of course, traced
genres through time, exploring in depth how they functioned within a moderately
stable activity system. The main goal
for Nowacek’s (2011) study was understanding transfer in student writing, but
she is concerned with questions of genre as well, arguing that genre can serve
as exigence for or obstacle to transfer.
In a 1997 article David Russell
surveyed studies that used CHAT as an approach for exploring genre and writing
in higher education or workplaces. Unfortunately,
I was only able to get my hands on the abstract, but it does demonstrate the
fact that activity theories have been productive for genre researchers. Within activity theory, as already noted,
genres are mediational tools or instruments.
In this pivotal position, it is easy to track the importance of the
genre within the system, as indeed Spinuzzi does. Even though my own discussion here has only
proposed how the theory might work rather than actually applying it to specific
texts, it is easy to see how it could work.
If I were in fact to apply the framework to actual examples of academic
interviews, I would use Engeström or Spinuzzi’s model because these
versions of CHAT capture more of the ecological framework. On the other hand, the AT model has the
advantage of focus and simplicity. I
might want to reflect the generational development by making that the core of
my analysis. I would not use the version
proposed by Prior et al. For one thing,
I struggle with what I see as the incompatibility of terms. Things are not
quite “flattened,” as Latour (2005) tries to do, and yet there are not levels
either. Process elements are mixed with
spatial elements. Furthermore, some
elements can be conflated or exchanged: Socialization
arguably occurs through activity, ecology is the spatialization of socialization and activity, distribution is
a type of activity, and so are production and reception.
One final thing that I found
interesting about activity theory in general is that in a sense both AT and
CHAT are future oriented. It is true
that CHAT refers to cultural-historical
activity theory, but the system is not examined in terms of how it operated
through history. Rather the system is an
artifact of historical processes, an idea that is captured by the laminated
chronotopes of Prior et al’s version. In
other words, instead of focusing on what has happened historically, the
analysis is focused on the deployment of the activity system to reach not yet
realized goals. This is neither a
strength nor a weakness, but it is, nevertheless, an interesting feature of AT
and CHAT.
Arnold, W.
(2012). The secret subject: Michel Foucault, Death and the Labyrinth, and the interview as genre. Criticism, 54(4),
567-581.References
Bourdieu, P.,
& Wacquant, L. J. D. (1992). An
invitation to reflexive sociology. Chicago: University of Chicago Press.
Bury, S. and
Scott, H. (2000). The artist speaks: The interview as documentation. Art Libraries Journal, 25(1), pp. 4-9.
Chakraborty, M.
N. (2010). Everybody’s Afraid of Gayatri Chakravorty Spivak: Reading interviews
with the public intellectual and postcolonial critic. Signs: Journal of
Women in Culture and Society, 35(3), 621-645.
Crawford, T. H.,
Smout, K. D., & Bazerman, C. (1995). An Interview with Charles
Bazerman. Composition Studies, 23(1),
21–36. Retrieved from http://www.jstor.org/stable/43501314
Eldred, M., &
Bazerman, C. (1995). "Writing Is Motivated Participation": An interview
with Charles Bazerman. Writing on
the Edge, 6(2), 7–20. Retrieved from http://www.jstor.org/stable/43156966
Engeström, Y.
(1987). Learning by expanding: An
activity-theoretical approach to developmental research. Helsinki:
Orienta-Konsultit Oy.
Giddens, A.,
& Pierson, C. (1998). Conversations
with Anthony Giddens: Making sense of modernity. Stanford, Calif: Stanford
University Press.
Latour, B.
(2005). Reassembling the social: An
introduction to actor-network-theory. Oxford: Oxford University Press.
McNely, B.
(2013). “That lightbulb feeling”: An Interview with Clay Spinuzzi .Present
Tense: A Journal of Rhetoric in Society 3(1). Retrieved from
http://www.presenttensejournal.org/volume-3/that-light-bulb-feeling-an-interview-with-clay-spinuzzi/
http://www.presenttensejournal.org/volume-3/that-light-bulb-feeling-an-interview-with-clay-spinuzzi/
Nowacek, R. S.
(2011). Agents of integration:
Understanding transfer as a rhetorical act. Carbondale: Southern Illinois
University Press.
Prior, P., & Shipka, J. (2003).
Chronotopic lamination: Tracing the contours of literate activity. In C.
Bazerman & D.d Russell (Eds.),Writing selves, writing societies:
Research from activity perspectives (pp.180-238). Fort Collins: The
WAC Clearinghouse and Mind, Culture, and Activity. Retrieved from http://wac.colostate.edu/books/selves_societies/prior/
Prior, P.,
Solberg, J., Berry, P., Bellwoar, H., Chewning, B., Lunsford, K.J., . . .
Walker, J.R. (2007). Re-situating and re-mediating the canons: A
Cultural-Historical remapping of rhetorical activity. Kairos, 11(3).
Retrieved from http://technorhetoric.net/11.3/binder.html?topoi/prior-et-al/index.html
Russell, D.R.
(1995). Activity theory and its
implications for writing instruction. In
J. Petraglia (Ed.), Reconceiving writing,
rethinking writing instruction, (51-78). Hillsdale, NJ: Erlbaum.
Russell,
D.R. (1997). Writing and genre in higher education and workplaces: A review of
studies that use cultural--historical activity theory. Mind, Culture,
and Activity, 4(4), 224-237.
Spinuzzi, C.
(2003). Tracing genres through
organizations: A sociocultural approach to information design. Cambridge,
Mass: MIT Press.
Tachino, T.
(2012). Theorizing uptake and knowledge mobilization: a case for intermediary
genre. Written Communication, 29(4), 455-476.
Van Oers, B.
(2010). Learning and learning theory from a cultural-historical point of
view. In B. Van Oers, W. Wardekker, E. Elbers, & R. Van Der Veer (Eds.), The transformation of learning: Advances in
cultural-historical activity theory (pp. 3-12). Cambridge: Cambridge
University Press.
Yamagata-Lynch,
L. C. (2010). Activity Systems Analysis Methods: Understanding Complex
Learning Environments. Berlin: Springer US.
No comments:
Post a Comment