It's hard to learn to drive a car, but in time the many laws that govern the
road, habits that govern the car, and social forms that shape the cultures
of driving settle into a subsidiary awareness, and our mind can focus on
other things until something unusual happens to recall the focal awareness
of conscious attention to the problem of driving. It is also hard to learn
to read a book, but while the interpretation of printed signs eventually
settles into subsidiary awareness, our focal awareness often enters into a
kind of dialogue with an author, and the printed page becomes a medium for
thinking. What kind of tool is a computer, and in what ways do we engage
with computers that we expect them to be a medium of educational change?
Perhaps the distinction between driving a car and reading a book is
arbitrary, for there are many ways we might engage a tool, whatever its
design or function, whether smart or inanimate. A recent car accident might
make a driver painfully aware of the rules of the road, and a mechanic might
listen to a car engine with a diagnostic mind that in some sense is engaging
in a dialogical relation with a machine. A proofreader or typist might see
only signs on the printed page, not ideas, or a bored student might daydream
while reading. While we are always free to daydream, it is when we engage
with tools in dialogical thought, as when we read, that they might have the
power to bring about the intellectual and social changes that are implied in
phrases like "the computer revolution."
And it is possible that computers might have the power to change us even
when we engage with them unconsciously, as when we relate to a tool through
the performance of a skill like driving or typing. Although skills are tacit
and embodied forms of knowledge, they are still forms of knowledge; as
Michael Polanyi defines skill, "the aim of a skillful performance is
achieved by the observance of a set of rules which are not known as such to
the person following them" (Personal Knowledge, University of Chicago Press,
1958). Although skills are ordinarily tacit forms of knowledge, we become
aware of them as knowledge when we are learning to use tools, or when tools
are broken. This is true in a trivial way with a tool like a typewriter
because a typist rarely fixes a typewriter, but fixing a car requires a
diagnostic sense of the machine and a mode of action that resembles a
performance in its marriage of thought and action.
Skills as knowledge
Michael Polanyi describes swimmers as skillfully yet unconsciously keeping
themselves afloat by regulating their breathing-not emptying their lungs
when breathing out and inflating them more than usual when breathing in; one
expert swimmer tried to discover what made him swim and was puzzled to
discover that "whatever he tried to do in the water, he always kept afloat."
Technical skills are a kind of performance requiring that theoretical
principles be transformed into embodied practices, so that mindful attention
can be directed toward the task at hand, not toward the tool. Thus technical
skills are accomplished by practicing rules that are not known as such to
the person following them.
To the non-expert, tools are either invisible or broken; in neither case are
they objects to think with. Tools disappear from our attention when we are
working with them, because we focus on the task at hand (as the eye fixes
upon the nail, not the hammer) and the tool tacitly becomes an extension of
the body. After the use of the tool has become habitual it becomes an object
of attention only when it doesn't work. I interviewed one novice who said,
"The computer is like a car. I don't want to pay any attention to it; if it
doesn't work I'll take it to someone who can fix it." Yet the tool that the
novice defines as "broken" is defined by an expert as "a problem," an object
of analytic attention.
To the expert, the computer is never "broken," rather, it presents problems
to be solved. Problem solving is essentially a process of communication with
an "other," although one that "interacts" rather than speaking. Seymour
Papert describes tools as "objects to think with," not just to use. As he
describes playing with gears in his childhood in an attempt to understand
the mathematical ratios that governed them (Mindstorms, Basic Books, 1980),
"Before I was two years old I had developed an intense involvement with
automobiles. The names of car parts made up a substantial portion of my
vocabulary: I was particularly proud of knowing about the parts of the
transmission system, the gearbox, and most especially the differential . .
.playing with gears became a favorite pastime. I loved rotating circular
objects against one another in gearlike motions . . . I became adept at
turning wheels in my head and at making chains of cause and effect: "this
one turns this way so that must turn that way, so . . . . ."
This kind of play consists of an attentive search for clues that illuminate
the laws governing the mechanism, and is the essence of the inductive
reasoning process which engineers call "problem solving." Inductive thinking
is the essence of problem solving, giving it a kinship with playing music or
games in that it is a kind of performance which unifies discipline and
spontaneity. And, in fact, this is the pedagogical core of problem solving,
for it defines knowledge operationally, in terms of the procedures necessary
to produce knowledge, not theoretically or in terms of its content. To an
expert, using a computer is like playing a musical instrument; the
"human-machine relationship" exists only for novices.
Why Humanists Drive Cars, While Artists Read Books
While a machine has a purpose built into its mechanism, a tool requires the
novice to acquire skill to realize a goal. Thus a computer is a field of
play only to a skilled programmer. A programmer told me, "Humanists will use
it as a tool, not get into programming, that is, as a plaything. They will
need low-level data communications, local networks, electronic mail to
exchange papers with colleagues-in sum, use the computer as a tool, not as a
source of pleasure. Programming gives one access to the computer as a
plaything, as a source of gratification. The pleasure is getting something
to work."
However, there is an exception to the observation that novices tend to use
machines as conceptual models and to approach the computer deductively. The
novices most successful in learning the computer were performing
artists-creative writers, poets, painters, musicians, dancers. Their
conceptual models were acts of performance rather than objects produced. All
thought of themselves as craftsmen, not scholars, and had learned their
craft by example in workshops. One said, "The computer is threatening to
scholars because they want to conserve the traditional forms of knowledge;
as artists we want to change them."
Yet there is a sense of craft in all forms of scholarship that resemble the
engagement between Papert and gears, as in the words of this English
professor: "I love script, manuscript, and handwriting. I write letters
longhand when I really do want to communicate with someone. I love the
physical process: I like to write, physically. I like shaping my paragraphs
and sentences on the page. I revise and insert endlessly on the page and I
get a pleasure, a sort of sculptural pleasure, out of inserting and revising
with my hands. I love the words that I use, and like to have immediate
contact with them. I love the English language and love the words as they
are, letter by letter, and the shape of them, and how I make them when I
write them. And I don't want to lose touch with that." There are obvious
differences between a mathematical engagement with gears and a linguistic
engagement with words, which parallel the differences between scientific and
humanistic cultures. For one thing, language is not only sculptural, a love
that is disrupted by the mechanistic discipline of the keyboard, it is also
potentially spoken and thus potentially human communication.
The Play's the Thing
Experts and children both often approach learning the computer by playing.
Children often focus upon the computer program as a field for play,
exploring in a random way, but forming a picture inductively through
experimentation. Problem-solving children learn to beat computer games by
losing, discovering when and where the program is "dumb" or predictable.
Given the information that in WordStar Control-K will "kill" a word, one
child tried out the whole alphabet as commands to see which worked and what
they did, building an inductive picture of the program (including its
undocumented features). Play, then, is an inductive mode of learning, while
work, or school, tends to be deductive.
The physical and mental engagement produced by skill and attention to clues
produces a feeling of integration that experts call play, and which they
compare to playing music or playing with games. This feeling of "flow" is an
experience of intense concentration and a distinct slowing of the sense of
time, produced by "a feeling of control and pursuit of a non-contradictory
goal," (Mihaly Csikszentmihayli, "The Concept of Flow," from Play and
Learning, edited by Brian Sutton-Smith, Gardner Press, 1979).
This kind of engagement is not play in the sense of spontaneous activity, it
is a feeling of control that is possible when playing games in which the
player's attention is focused within an environment with a known set of
rules which have been mastered and become tacit. Problem solving is a kind
of game to experts, a "field" of rule-governed play which is a contest for
mastery between their skill and the machine.
Several of the computer programmers I studied were also musicians, and their
music making served as a powerful metaphor for their relationship to the
computer. In performance, the musical instrument and player interact in a
manner that cannot accurately be described as a human-tool relation. David
Sudnow comments, "If I watch my hands on a typewriter I don't recognize
their movements...but the sight of my piano- playing hands is familiar" (The
Ways of the Hand, Simon & Schuster, 1984). Problem solving is performance,
whether it is a matter of fixing the program or creating a program, and in
performance the tool never entirely becomes tacit.
Central to all of these dimensions of skill is the operational definition of
knowledge as performance, without a feeling of separation between performer
and tool, or content and procedure. Computer knowledge is operational
knowledge in that the content of knowledge cannot be separated from the
activity by which it is produced.
The concept of performance suggests why there is a kinship between computer
technology and the arts, and why computers and the arts are estranged from
the humanistic disciplines. Indeed, Papert summed it all up when he
commented that in some respects "scientific knowledge is more similar to
knowing a person than similar to knowing a fact or having a skill."(1980).
(Adapted from a longer paper in Work and Technology in Higher Education,
Laurence Erlbaum Assoc., 1995)
© 1995 Educom.