This seems an appropriate context in which to pursue some recent xmca
questions about tools and contexts, production and use, activities and
timescales.
The Kellers' article is part ethnographic account of the making of
decorative and functional ironwork objects, part theoretical argument for a
more unified account of the role of signs and tools, visual mediations and
verbal mediations in activity.
I generally agree with them that tools also function as signs, and that all
signs also have material affordances as tools, especially if we analyze
them in the context of their actual functions in complex activity, and not
as if they functioned as isolated artifacts. They are also trying to make
their point by shifting attention away from the paradigm case of language
as sign system (where the materiality of the sign seems least significant)
to the case of visual imagery. In some respects their account is a bit more
aligned with traditional cognitive theory than I would wish to be, e.g. in
emphasizing visual images and schemas "held in the mind" during activity
and assigning a key role to mental comparisons of imaged goals with the
current state of the work. Despite this, and some of the biases often
associated with it (e.g. emphasizing static imagery and printed diagrams
more than dynamical process-habits, imagined action-flow, etc.), their
account arrives at an admirably dynamical view of emergent trajectories of
activity in which visual-kinesthetic tools and models interact with
verbalizations to produce improvisation-within-tradition.
I was fascinated first of all with their photograph of "visible storage of
tools" in a modern smithy, and the accompanying account of how the tools
visibly remind the artist-smith of possible actions and past actions in
working the heated, plastic iron. The parallel to my own office with its
visible display of books (organized by topic or theme or sometimes task),
piles of papers (organized by task or activity), and on screen on my
computer, the rows of icons for programs and worktasks and program-actions.
I know how much I am often influenced by these visible reminders, evokers,
stimuli as I work. These material-semiotic artifacts (i.e. things that also
mean) are for me in sharp contrast to the more cognitivist emphasis on
visual _mental_ images. Mental imagery, if there really is such a thing,
seems to me very fluid and protean, or else schematic and built more of
semiotic elements than of sensory features. When you really see a vise or
hammer, or a specialized implement (there are dozens along these walls),
you CAN see a great deal of detail that is banished from the mental image
of a tool because it is not part of the semiotic scheme for the tool (just
as we do not usually visualize random pencil marks on a visualized page of
text, or every freckle on a friend's face). In a rich visual OBJECT, like a
painting, there is always much more than we ever remember or visualize, and
sometimes, perhaps esp. when we improvise, it is the unrepresented detail
that matters most. (See below on choreographic improvisation.)
In their account of the visual stock of knowledge put to use in smithing,
the Kellers show us the examples of worked iron and the diagrams of the
steps in forming certain standard shapes, but necessarily do so, and to
some degree speak as if these static images were themselves the directly
mediating artifacts (or as if mental images of them are), though later in
the article it seems clear that something much more dynamical is involved.
It is easy to talk about a static image being "held in the mind", but this
does not seem so natural if we are talking instead about a dynamical visual
imagining, like a movie or video, which has to "run" or develop in time.
Moreso, I think, if we add in, as the Kellers certainly do, the critical
importance of the _kinesthetic_ aspect of the sensorimotor construction of
meaningful activity (done or imagined). A first-person video of feeling as
well as seeing ourselves in action, doing and moving, in a continuous
temporal development, does not seem to me the sort of thing one "holds in
mind". It is rather the stuff of which mind, or consciousness itself is
made. It _is_ consciousness, just with a characteristic meta-twist of being
aware of our running this little movie (as playback or as imagination),
overlaying consciousness of being conscious itself (the typical
neurological "re-entrant connectivity" view of self-awareness, as in
Edelman). And the kinesthetic-visual fusion seems very important (as in
virtual reality phenomena like telepresence) here: we rehearse
ourselves-doing, sometimes with a lot of vivid detail, more often in
semiotic schematic and abstracted outline.
I once did a brief observational study of a choreographer inventing a new
dance piece 'on the dancers'. He began from the music, imagined movements
and sequences of movements, and then directed the dancers to try what he
wanted. Both his imagination and his communication were limited in a sense
by the semiotic resources he was using: classical ballet movement
"vocabulary" and "syntax", and the verbal register and speech genres of
this sort of communication. He also had his own considerable visual memory
of dances seen, as well as his kinesthetic memory (he was himself a dancer)
of movements performed. His dynamical imagining was often in first-person,
as if he were performing the movements rather than observing someone else
do so. He often demonstrated the movement. But the limitations of the
schematized semiotic were apparent when he "got stuck", usually because
there was some small detail about how the dancers bodies (two at a time,
particularly) could not fluidly and anatomically get from one imagined
movement to the next. From these transitions that felt or looked awkward on
or for the dancers, he would improvise next moves or revisions that would
in turn often inspire unplanned further sequences and continuations. There
was a lot that was emergent, a lot that was improvised. But a key
"constraint" that was also enabling was that schematic semiotic imagining
always misses the non-distinctive details; the little material facts of
continuous reality that do not matter to the distinctions that a semiotic
system constructs for us. Choreographic semiotics is both typological and
topological: it is concerned with this-vs-that as types of movement, but
also with fluidity and grace and style in the continuous topology of a body
in space. A starker contrast occurs in the case of language, where if we
imagine new combinations of sounds to make new words, we normally stay
within the phonemic repertoire of our language and do not make significant
distinctions based on the continuous spectrum of different frequencies (or
positions of tongue, lips, etc.) that in principle is possible. In some
sense we always imagine phonemes, but we always have to utter "phones",
that is a distinctive sound of the language PLUS those extra little details
that identify an "accent" or our personal voice, but don't count as making
a completely different word.
If an experienced smith looks at a picture in a book of a sequence of steps
in the working of iron into a particular complex shape, or if a dancer
views the analogous forms of their art, it seems to me much more important
that they produce a dynamical feeling-visioning of the doing (one that can
be "re-played" with some fidelity) than that they hold in mind the static
visual images they have seen.
Likewise, in the improvisation of activity in real-time, it would seem to
me that it is the interplay between the dynamical action-imagining,
unfolding in time, and the felt and seen actual doing that is most relevant
to how past experience or planning influences real-time action. And in this
case, unlike that of the static mental image, it also seems less neat to
imagine back-and-forth comparisons of the planned result and the current
state. Dynamical imagination is trickier to control the rate of; it has its
own intrinsic pacing, or at least relative pacing, and we cannot easily
just re-start it like a videotape at any point we wish. There are key
issues of relative timing and timescales between the dynamic imagining and
the actual doing, which I think preclude a simple model of
goalstate-nowstate comparison, even as they enable a more dynamical picture
of the emergence of unplanned improvisation.
But back to tools. What enables us to use the smithing tools to work the
redhot metal is most basically that they are more stable in time under
these conditions than the working metal. Their constancy of form and
material affordance, perhaps even more than that of an imagined static goal
state, helps us to control the fluidity of conscious imagining, as it does
the plasticity of the heated metal. The tool does not change on the
timescale of our use of it (or only in minor ways), yet as a key
participant in the activity of smithing, it can stabilize the entire
action-dynamic; it provides the constant element needed for improvisation
to avoid lapsing into unpredictability and chaos. It is important I think
to analyze the timescales of tools and tool-use. On what timescales do the
tools themselves change? the minor and major ways of using the tools?
I won't take the time here to discuss this for the case of language, or
language and dance semiotics in choreography, but I think one can see that
the syntax of such tool-resources changes quite slowly compared to rates of
real-time use, vocabulary a little faster (but still slow on the focal
timescale of speech), and combinations of words, the deployment of
vocabulary and syntax, as fast as speech itself (for it _is_ speech,
linguistically) for individual sentences and clauses, but as one goes to
higher scales of organization (rhetorical patterns like 'problem/solution',
or genre structures like that of a folktale), over longer times of
speaking, we again make use of more stable patterns.
At the end of the article, having argued that language plays no "essential"
role in smithing (but is functionally replaced by the visual semiotics of
smithing imagery, both as resource and as cultural repertoire), the Kellers
do raise some questions about what the role of language is. I would tend to
be more skeptical that language does not have an essential role. It may
well be that language is not the primary semiotic medium for creation in
smithing, as it may not be for musical composers or many other visual
artists. But it does normally seem to be an essential component of the
multimedia mix by which we act meaningfully. My own view is that the
semiotic resources systems do not ever function independently or
autonomously, not in ontogenetic development nor in any normal cultural
activity. You can't make meaning with just one (I have argued this
elsewhere in more detail). They are all specializations and
differentiations with a common root, and even once separated functionally,
they have co-evolved phylogenetically and historically to work together.
The Kellers suggest some of the cooperative roles of language in smithing.
Their argument may not need to depend as much as it does on opposing visual
semiotics to language, esp. if we take a more dynamical view of language as
well. Modern linguistics, which too often really studies writing rather
than speech (or at least only those aspects of speech that are also found
in written text), has perhaps contributed to a static view of language as
an array of signs, rather than as a mode of action. It has also unduly
restricted the scope of "language" by demanding semiotic-structural
relationships and excluded aspects of meaningful vocal action (laughing,
pausing, shouting, hmm-ing, etc.) that are tied more dynamically to the
speech and action stream. Do smiths work in dead silence? no humming, no
little vocalizations or subvocalizations to pace and self-guide their
action? Vocalization of one sort or another is a very pervasive aspect of
human action-flow, and it is always meaningful and functional, though
perhaps not in the limited semantic sense recognized by much linguistic
theory. This is only the most basic level of connection of speech with
other forms of motor activity; the semantic resources of language also
certainly play a role in learning how to do smithing and in communicating
about its activities and forms to others.
I have deliberately emphasized the material level of speech/vocalization
here, because that is also where it is most clearly dynamical, and my
argument here is for the dynamical integration of semiotic mediating
practices into ongoing activity, with attention to the relative timescales
of characteristic processes and practices. In this view, language operates
as the material reality of speech, dynamically integrated into ongoing
action, and material tools (whether hammers or books) play their roles in
relation to long timescales of tool- design and making as well as shorter
timescales of tool-use, linking the two, and ultimately, linking
longstanding cultural patterns and slow historical change with short-term
activity and improvisation.
JAY.
---------------------------
JAY L. LEMKE
PROFESSOR OF EDUCATION
CITY UNIVERSITY OF NEW YORK
JLLBC who-is-at CUNYVM.CUNY.EDU
<http://academic.brooklyn.cuny.edu/education/jlemke/index.htm>
---------------------------