Dear Colleagues,

In a recent post to the DRS list, the Vice Chair of the Design
Research Society posted a "CFO" (Call for Objects) for a new strand
at the Common Ground Conference (Robertson: 2002). He also stated
that this strand might give rise to "a second volume of the
proceedings."

This strand takes the form of an exhibition of designed artifacts.

The idea of an exhibition of research-related artifacts is welcome.
Medical research in surgery is often reported with a demonstration of
surgical procedures. Advanced computerized simulations support
advanced research results in physical science. In the same way,
design research should embrace the use of exhibited artifacts in
reporting research.

What I do not understand is the notion stated in category c),
"stand-alone exhibits."

The call reads, "In some cases, researchers may feel that it is
appropriate to present their research in the form of an exhibition
which is self-explanatory and does not require a separate written
paper."

I do not understand how an artifact can be "self-explanatory" as a
research presentation. A research presentation presumes several kinds
of discourse that cannot be articulated in an artifact.

An artifact may be the outcome of research. It may demonstrate the
practical and applied research outcome. It cannot be self-explanatory
for all the criteria required of a RESEARCH RESULT.

These criteria include at least nine issues that no artifact can explain,

1. Statement of the research problem,
2. Discussion of knowledge in the field to date,
3. Discussion of past attempts to examine or solve the problem,
4. Discussion of methods and approach,
5. Comparison of possible alterative methods,
6. Discussion of problems encountered in the research,
7. Explanation of how the researcher addresses those problems,
8. Explicit contribution to the body of knowledge within the field,
9. Implications for future research.

It is time for someone to explain how a "self-explanatory" artifact
can address these issues.

If Ray Kurzweil (1999) is correct, advances in artificial
intelligence will make such self-explanatory artifacts possible
within the next two decades. They do not exist today.

There may be other ways for objects to explain themselves.

Perhaps, in a research universe long, long ago and far, far away,
designers with mystical powers knew how to elicit "self-explanation"
from mute artifacts. Dr. Leah Organa summarized this research
approach in the first part of the first Star Wars trilogy. This is
the "Help me, Obi-wan Kenobe!" method. And even without The Force,
little R2D2 would have been a self-explanatory object.

There is also the saga explanation. This is the tale of Wayland the
Smith and his Magic Knife. We tell it to our children on dark winter
nights.

Sit down by the fire, children . . . . Look into the flames. Can you
see the glow of Wayland's hearth? Listen to crackling wood! Can you
hear the roaring and the hissing of Wayland's coals? Yes! That's the
sound! Listen! Listen to the hammer on the anvil as Wayland turns the
knife. Listen! Listen to him pump the bellows to heat the coals.
Flames shoot up and roar as he thrusts the blade back into the
glowing fire.

Watch! The knife goes back into forge. Wayland stretches his left arm
out over the glowing metal. His right hand holds a tiny silver knife
covered with runes. He cuts his arm. Blood pours out onto the blade.
Red drops hit the glowing red blade. The drops leap into the air,
steaming and sizzling.

Listen to the Smith! He chants the runes as he walks around the forge
in a strange, halting cadence. He sings the magic words. "O-din!" he
cries out, "Ooooooooo-diiiiiiiiiiin! Give your rune-craft to this
blade. Let this knife speak!"

As we near the end of the tale, the Magic Knife learns to speak. The
Magic Knife attends the Common Ground Conference of the Design
Research Society. This is a self-explanatory knife, and it explains
itself very well indeed.

The knife gives a well-received presentation titled "Special
Tempering Qualities in High-Carbon Steel Subject to Astral Quantum
Spin-Effects of the Anglo-Norse Pantheon."

These three artifacts are self-explanatory. We will have them when we
have advanced artificial intelligence or great advances in mystical
power. Until then, I do not see how a designer can present "research
in the form of an exhibition which is self-explanatory and does not
require a separate written paper" (Robertson 2000: unpaged)

Let us teleport from the land of elves and wizards to another
research universe long, long ago and far, far away. Let us visit
England in the year 1993, when Nigel Cross wrote the first of two
editorials in Design Studies related to the theme of the
self-explanatory object.

In his first editorial, Cross (1993: 226-7) discusses questions
inherent in practice-based research. He raises several themes. He
points out the distinctions between practice and research and the
value of connecting research to teaching and to practice.

In his second editorial, Cross notes how little progress had been
made in practice-based research over the two years between 1993 and
1995. He writes that part of the problem involves the claim that
"works of design are also works of research" (Cross 1995: 2).

Cross (1995: 3) states that the best examples of design research are:
purposive, inquisitive, informed, methodical, and communicable. An
artifact can be the RESULT of research conducted according to these
criteria. No artifact can possibly ARTICULATE and COMMUNICATE how a
research project FULFILLS these criteria.

In the year 1999, the DRS list was the forum of a lengthy debate on
research methods in design. Nigel Cross once again addressed this
issue.

Looking back over the failed efforts of the past decade to produce
research in which the practical outcome was also the research result,
Cross (1999: unpaged) wrote, ". . . as I said in my Editorial in
1995, I still haven't seen much strong evidence of the output from
the 'research for and through design' quarters. Less of the special
pleading and more of the valid, demonstrable research output might
help."

In reviewing the debates and special pleadings of the past ten years,
I have noticed a major gap in the arguments for and against the
notion of the self-explanatory object. This missing argument explains
why we have not yet seen an example of a self-explanatory artifact.

The fact is that no artifact can perform adequately on the research
criteria delineated in Prof. Cross's list or mine. Since no artifact
can meet the criteria of an independent research result, no artifact
can be "self-explanatory" in the sense that a research report
explains itself.

In the decade since Cross wrote his 1993 editorial, no one has stated
clearly what such self-explanatory object might be and no one has
described how it would fulfill the criteria of a successful research
result.

There has been an endless round of debates, conference presentations,
list postings, and even a few journal articles to argue for the
notion of a self-explanatory object. Not one of these has moved
beyond articles of faith to the level of reasoned argument or direct
demonstration.

An artifact may constitute the PRACTICAL OUTCOME of successful
research. Penicillin, computer programs, and new kinds of engines are
the practical outcomes of successful research. These are not the
RESEARCH RESULT.

A research result contributes to a body of knowledge. It is of a
different nature than the practical outcome to which the research
result also gives rise. Practical outcomes are necessary and useful.
Research results move us beyond today's useful practical outcomes to
the improved practical outcomes that we will need tomorrow.

The RESEARCH RESULT is different than a practical outcome. In
confusing an artifact with a research result, the Vice Chair of the
Design Research Society thinks he is describing the same thing. If a
self-explanatory artifact is the same thing as a research result, the
"self-explanatory" artifact should be able to tell us in articulate,
conscious terms:

1. The research problem of which the artifact is the research result;

2. The knowledge in the field before the research project began;

3. Past attempts to examine or solve the problem;

4. The methods and approaches explored to examine or solve the
problem in the specific research project;

5. Possible alternate research methods that might have been used to
examine or solve the problem; the reason that any specific method was
accepted, rejected, partially accepted or modified; and why the
specific research method used was chosen;

6. Problems encountered in the research;

7. How the researcher addressed those problems; and whether he or she
managed to resolve all of the problems or only some of them;

8. The contribution this research result makes to the body of
knowledge within the field;

9. The implications this self-explanatory artifact holds for future
research; how the contribution to knowledge makes it possible for
future researchers to understand and account for research issues
central to the result; likely steps toward future research.

This cannot be done.

The notion of a "self-explanatory" artifact only makes sense to
someone who does not understand the nature of research. This
misunderstanding involves the failure to see design as a social
activity undertaken in networks. It is based on the notion that a
contribution to knowledge is an individual activity in which research
results are single artifacts set out for display like crafty objects
at an exhibition. This perspective fails to understand research as an
activity in which the contribution to knowledge is embedded in a
social discourse linked to earlier contributions and related to later
contributions. Moreover, it fails to respect the requirements of
explicit communication that enable others to understand the
contribution and - if necessary - to replicate or otherwise
conceptually trace each step of the research project leading to any
given result. This requires shared knowledge within a community.

Part of the research project involves sharing knowledge in a way that
enables the entire community to understand and work on a research
problem.

Explaining the conceptual confusion of a self-explanatory artifact is
related to the problem of tacit knowledge in research.

Tacit knowledge is an important form of knowledge. It plays many
roles in many forms of work, including socialization, skills
modeling, communication behavior, culture building and more.

Even so, tacit knowledge is unsuitable for describing research. This
is because tacit knowledge is unable to explicitly state and
distinguish among the kinds of problems that we must address to
complete and communicate research.

Tacit knowledge plays an important role in the research PROCESS.
Tacit knowledge of research skills, methods, methodics and
methodology; tacit knowledge of instrumentation, analysis, logic, and
rhetoric; and tacit knowledge of how to further a research inquiry
are all vital parts of any researcher's repertoire. A good researcher
develops tacit knowledge of all these issues.

Here arises a problem that many research novices do not understand.
This is the distinction between knowledge and information.

Tacit knowledge is a form of knowledge. It is not a form of
information or experience. Knowledge can never be shared directly
(except, of course, among telepaths and Jedi knights). Not even
explicit knowledge can be share directly. Knowledge is transmitted
and communicated through shared information and experience. Knowledge
is developed within an individual when he or she works to integrate
information and experience into a personal stock of knowledge.

Knowledge is always the property of a knowing entity. Any knowing
entity must by definition be a conscious being. A human being knows
and therefore a human being may have knowledge. An artifact without
self-conscious agency cannot know or have knowledge.

Some non-human animals probably know or have knowledge. (While I am
aware of reasonable opposing arguments, I argue that evidence
supports the likelihood that some animals know and have knowledge.
There are different reasonable views on the status of knowing and
knowledge among non-human animals.)

Many non-human animals are clearly the self-aware conscious centers
of their own worlds of perception and action. They learn from
experience. They remember, they demonstrate volition, and they act on
their desires. Some seem to make conscious plans and decisions. In
this sense, it is possible to argue that a non-human animal may know
and have knowledge.

In contrast, no one - other than Walt Disney or Wayland Smith -
argues that a teapot, a knife, a radio, or even a computer can know
anything as human beings know. There is no case to be made for the
idea that pair of scissors, an automobile, or even an advanced
telecom network system based on "intelligent agents" has knowledge in
any way.

Beyond this, it is vital to distinguish the fact that manufactured or
crafted objects may record, but they cannot experience.

Knowledge presupposes experience.

Although the process of learning and the nature of knowledge are not
completely understood, there is wide agreement that knowledge
creation requires experience. Kolb's (1984: 38) definition of
learning as "the process whereby knowledge is created through the
transformation of experience" offers a useful perspective.

Research is a knowledge creation process. A research result requires
reflection. A research result reports experience.

The practical outcome of a research result cannot report the
experience that leads to its creation. This is why the practical
outcome of a research process cannot, in itself, be the research
result.

The researcher is the one who experiences and reports the research.

Any kind of experience may, in principle, be transformed into
knowledge. Kolb emphasizes the relationship between experience and
knowledge as a dynamic process of continuous reproduction and
regeneration. It contradicts the static model of learning as
acquiring knowledge external to and independent of the learner.
Information and facts are external to and independent of the learner.
Knowledge inheres in human beings and the specific form of knowledge
depends on the learning process.

Because knowledge is human, developing knowledge requires thinking
and practice, mind and body both. Mindless recording will not
transform experience into knowledge. Learning requires human agency,
a concept synonymous with Heidegger's concept of care, the human
tendency for each person to care about his own existence (Heidegger
1993: 238). For Heidegger, both practical knowledge and theoretical
knowledge express human care in an intimate relationship between
action and knowledge.

When we speak of contributing to the knowledge of the field, we use a
metaphor. We do not actually contribute knowledge to the field. We
develop knowledge. We report it and transmit it as information. Other
human beings take in and process the information in different ways to
experience it directly or vicariously. It then becomes part of the
stock of knowledge of individuals.

Individuals collectively form a field. Their stock of knowledge
collectively forms the knowledge of the field. The whole field never
directly knows the knowledge of a field. The individuals who
constitute the community of the field know it. It must always be
transformed into communicable knowledge -- information -- for
transmission.

Only externalized knowledge in the form of information allows two or
more individuals to share knowledge. Knowledge must be reduced from
knowledge to information to be shared. It must be reconstituted
through an act of individual agency to become knowledge again.

The reason we can speak both metaphorically and truthfully of
contributing to "the knowledge of the field" is that the knowledge of
a field resides in the human beings who comprise the social community
of a field.

These issues are considered in such fields as learning theory,
organizational learning, and knowledge creation. I am not going to
discuss these issues more thoroughly here. I raise these issues to
explain why the notion of a self-explanatory object as a research
result is a contradiction in terms.

Kolb's definition of learning and Heidegger's concept of care form a
model of the ways in which human beings monitor and control
knowledge. This model rests on three human capacities.

These capacities are:

1) The ability to act,

2) The ability to apprehend action and the environment, within which
action takes place,

3) Critical comprehension.

Since an artifact lacks these central human capacities, the artifact
cannot by self-explanatory. The object is mute.

The mute object cannot communicate knowledge of any kind. A mute
artifact cannot even embody tacit knowledge. Tacit knowledge is, by
definition, embodied in a knowing agent. Unlike human beings,
artifacts cannot posses or transmit tacit knowledge. They may be the
result of tacit knowledge applied to problems, but they cannot
COMMUNICATE or EMBODY tacit knowledge.

An artifact cannot explain the issues and processes that constitute a
research result. An artifact cannot communicate or explain the
research result of which the artifact is the practical outcome.

Two issues will help to summarize and conclude this note. The first
involves the concept of a well-understood problem.

In every field, there is a notion that we understand and can address
certain problems before we understand their solution. We can define
the problems. We can state the criteria of a successful solution. We
can even know within reason why certain solution sets may or may not
contain possible solutions to those problems.

This does not mean that we have the answer. Rather, it means we have
defined a conceptual frame within which better understanding opens a
potential solution space that permits us to go about solving the
problem.

Mathematics offers many famous examples of well-understood problems.
One is the halting problem for a Turing machine. We know what the
problem is. No one has solved it yet, and most consider it unsolvable.

Another well-understood problem is the Hilbert consistency problem.
The problem was understood by the late 1800s. Hilbert stated it in
his list of problems in 1900. Mathematicians, logicians, and
philosophers struggled with it for decades. Finally, in 1931, Goedel
disposed of the problem with his incompleteness theorem. This
well-understood problem took decades to solve.

The second issue involves the issue of correspondence. Correspondence
involves the way in which new knowledge fits into the earlier
knowledge of a field.

This new knowledge may overturn earlier knowledge. If it does, it
will also show what worked and what did not in prior science. This is
what happened when Copernicus, Galileo, Kepler, and Newton overturned
and replaced Ptolemy.

New knowledge may revolutionize and build on earlier knowledge. If it
does, it will show how it enfolds and expands prior theory and
earlier results. Einstein's theory of gravitation embraces and
explains Newton's theory at a higher level. In contrast, the
development of string theory known as M-theory may (or may not)
embrace the two correct, well understood, and mutually contradictory
bodies of theory developed in large-scale theories of relativity and
gravitation and the small-scale theories of quantum physics. One
reason that so many scientists have great hope for M-Theory is the
profound correspondence it holds with prior knowledge, and the ways
in which it explains and integrates earlier theories on a more
fundamental level.

Physicist Jeremy Bernstein discusses why this happens -- and how --
in an article titled, "How Can We be sure that Albert Einstein was
Not a Crank?" (Bernstein 1993: 15-27).

Crank science is visible in the theories of researchers who insist
that they can produce a perpetual motion machine. All known physics
explains why this cannot be so. On an immediate scale, the second law
of thermodynamics explains why perpetual motion is impossible. So
far, no proposed principle of perpetual motion has been able to
escape entropy. The laws governing entropy fit neatly with all the
rest of physics, large and small.

Before I return from the world of quarks and the cosmos, let us read
Bernstein on crank research. "The vintage crank paper" he writes,
"has no correspondence and no predictiveness. This is why it, and its
author are, as a rule, so hard to deal with. It is as if the crank is
speaking in tongues." (Bernstein 1993: 27)

Bernstein notes that Einstein's astonishing work was rooted in "a
clear understanding of the physics that preceded his own. The typical
crank appears to regard all this apprenticeship as beneath his
intellectual dignity. He wants to go right to the head of the class.
No apprenticeship for HIM" (Italics in Bernstein 1993: 27)

Cross and others have repeatedly pointed out that no one has yet
produced a self-explanatory object.

I have explained why this cannot be done.

For a decade now, the field of design research has been subject to
endless debates confusing the valid relationship between practice and
research with an invalid notion. This invalid notion conflates a
practical or applied research outcome with a general or descriptive
research result.

In these debates, we have been told repeatedly that it SHOULD be
possible to produce a self-explanatory object. We have been told that
we SHOULD be willing to accept this self-explanatory object as a
valid research result. The problem here is that no one has yet
demonstrated that such an object is POSSIBLE. No one has even offered
a theoretical demonstration that it is possible IN PRINCIPLE.

Because this issue is important and interesting, many serious
scholars have been willing to engage in the debate. They have shown
patience for more than a decade with those who claim that they will
one day be able to produce the magical self-explanatory object.

When called on to offer a foundation for their belief, those who
argue for the "self-explanatory" object remain silent and point to
the concept of tacit knowledge. A general principle in research is
that a researcher should be able to explain his position. Those who
believe that objects can explain themselves are as silent on this
issue as their objects.

In the past year or two, most of us have begun to wonder where these
self-explanatory objects are. Now it is time to take the next step.

This step begins with a principle argument: I argue that given
today's technology, it is impossible to create a "self-explanatory"
artifact that also constitutes a full research report without the
need of an accompanying statement. I have demonstrated a theoretical
and practical foundation for this argument.

This will not be possible until an artifact can be said to possess
consciousness. Until we are ready to agree that an artifact can
explain itself in the way that a human being can do, there can be no
self-explanatory object.

Further, I argue that a self-explanatory artifact constituting a
complete research result is impossible even in theory.

Let us treat this as a well-understood problem.

IF it is possible in principle to produce a self-explanatory artifact
that constitutes a complete and proper research result, THEN it is
possible to state the nature of such an artifact and to explain the
basis on which it could, in theory, exist.

So far, every argument for the notion of a self-explanatory artifact
as a complete, valid research result has taken the form of personal
assertion with no foundation other than personal belief.

The proponents of the self-explanatory artifact seem to be saying,
"It can be done because I say it can!"

Despite the continual claims that a demonstration is around the
corner, no one has produced a demonstration. No one has shown an
object that meets the criteria of a complete research result.

Now, it is up to those who claim they can produce such an object to
demonstrate that this is possible.

In the spirit of Chris Rust's distinguished "Smarties Prize," I
propose a challenge. Let us call it the "Design Thinking Prize." To
the winner, I will give a copy of Henry Petroski's (1997) elegant
book, Invention by Design: How Engineers Get from Thought to Thing.

The challenge has four parts.

Any challenger who can meet the four parts of this challenge will
clearly have proven that a self-explanatory object is possible in
principle.

If it is impossible to prove in principle that such an object is
possible, I argue that we can end this debate and move on to
productive issues.

The four parts of the challenge involve defining a research result
and demonstrating that a self-explanatory artifact can meet the
definition of a research result.

To win the Design Thinking Prize, an entry must:


1) State the general criteria of a complete and valid research result.

2) Distinguish the concept of a research result as research from the
practical or applied outcome of the research.

3) State the basis on which a self-explanatory artifact would meet
the criteria of a complete and valid research result as distinct from
the practical or applied outcome of the research.

4) State criteria on which such a research result would be accepted
as valid and state the criteria that would invalidate such an effort.

That is the challenge.

It is not necessary to produce one of the self-explanatory artifacts
of which we have heard so much and seen so little during these past
ten years.

It is only necessary to define a research result, state clear
criteria, and articulate the basis on which such an artifact would be
acceptable if someone were able to produce one.

In offering this prize, I ask the Design Research Society to appoint
a board of expert jurors drawn from the conference committee of
Common Ground. I request the Design Research Society to select and
announce a jury by May 1, 2002. I will not serve on the jury, nor
will I be involved in the selection of a winner if any challenger is
able to claim the prize. The decision of the expert jury is final. If
the jury agrees that any challenger has properly met the challenge, I
will accept the decision of the jury as binding.

Entries may be posted to the DRS list or to the PhD-Design list at
any time before September 1, 2002. Any individual or group of
individuals may enter a challenge.

Nigel Cross recently reminded me of a slogan used at the MIT Media
Lab. The slogan is,

"Demo or die!"

After ten years of talking in circles, the time has come to
demonstrate that it is possible - even if only in theory - to produce
a self-explanatory artifact that also constitutes a full research
report without the need of an accompanying statement.

In my view, the notion is a fairy tale or science fiction. Fairy
tales are fine for long winter nights and science fiction makes good
reading. I believe in the value of fairy tales and the importance of
science fiction. I read both. They are also useful in research, at
least for stimulating debate and generating ideas.

Problems arise when researchers confuse fairy tales with research.
When the insistence on an idea begins to transform segments of a
research field into a religious cult, I say "Demo or die!"

For more than two thousand years, deluded alchemists and deliberate
frauds promised to deliver such miracles as the philosopher's stone,
the elixir of life, or a method for transforming lead into gold if
only their patrons would give them little more time and a lot more
money. This ended with rise of natural science, the development of
new philosophies and deeper historical studies, together with a
richer field of social and behavioral science.

We do not have the elixir of life. We do have penicillin,
non-invasive therapies, and a better understanding of how the natural
medicines of indigenous peoples work as contrasted with the false
claims of neo-scholastic alchemists.

We cannot transform lead into gold. We do have automobiles, Internet,
and nanotechnology.

We do not have the philosopher's stone. Instead, we understand
increasingly more about the relationships and forces that lie within
and beneath the world around us.

Some voices in the debates of the past decade insist that the
"self-explanatory" artifact is the robust design practitioner's
answer to the intangible theories of scholars and scientists.

I disagree.

The notion of the self-explanatory artifact is the direct descendent
of arcane medieval scholasticism. The debate over whether such an
artifact is good for design research resembles the debate over angels
dancing on the head of a pin. Since they do not exist, the issue is
meaningless.

This debate does not position the hearty guild master with his daily
pint against the effete bookworm with his dusty tomes.

Instead, the confusion of issues and claims for the self-explanatory
artifact resembles the assertions of scholastic medieval biologists.
These confused souls preferred reading Aristotle's mistaken claims in
an ancient manuscript to counting a spider's legs directly.
(Aristotle was not to blame for this nonsense. He was a great thinker
and biologist. He also made mistakes. The problem was never
Aristotle's writings. The problem was their misuse by researchers who
transformed Aristotle's writings from a philosophical source into the
iconic text of a religious cult.)

The physicist Wolfgang Pauli used to say that some papers were so bad
that they "were not EVEN wrong." When a debate moves from thought
experiment to religious cult, it takes on qualities that move in that
direction.

It is time to demo or die.

Is it possible in principle to create a "self-explanatory" artifact
that also constitutes a full research report without the need of an
accompanying statement?

If it is, prove it.


-- Ken Friedman



References

Bernstein, Jeremy. 1993. Cranks, Quarks, and the Cosmos. New York: Basic Books.

Cross. Nigel. 1993. Editorial. Design Studies. Vol. 14, No. 3, 1993,
pp. 226-227.

Cross. Nigel. 1995. Editorial. Design Studies. Vol. 16, No. 1, 1995, pp. 2-3.

Cross. Nigel. 1999. "Subject: Re: Research into, for and through
designs." DRS. Date: Mon, 13 Dec 1999 13:43:18 +0000.

Heidegger, Martin. 1993. Being and time. Oxford: Blackwell.

Kolb, David A. 1984. Experiential learning: experience as the source
of learning and development. Englewood Cliffs, N.J.: Prentice Hall.

Kurzweil, Ray. 1999. The Age of Intelligent Machines. How We Will
Love, Work, and Think in the New Age of Intelligent Machines. London:
Orion Business Books.

Petroski, Henry. 1997. Invention by Design. How Engineers Get from
Thought to Thing. Cambridge, Massachusetts: Harvard University Press.

Robertson, Alec. 2002. "Subject: SPECIAL EXHIBITION." Wed, 20 Mar
2002 08:17:21 -0000. DRS Digest. 11 Mar 2002 to 20 Mar 2002
(#2002-25).




--

Ken Friedman, Ph.D.
Associate Professor of Leadership and Strategic Design
Department of Leadership and Organization
Norwegian School of Management

Visiting Professor
Advanced Research Institute
School of Art and Design
Staffordshire University