Hi Klaus,
Thank you for your comments.
My feeling is you are assuming that the discussion is bounded by the duality between reductionism vs human creative action?
You suggest that prediction is an extrapolation of past pattern into the future. This is a very limited view of prediction. Prediction through understanding actions of causal forces doesn’t require the existence of a past pattern or the details of a 'starting point' .
You suggest that ‘ design brings forth something that cannot be predicted without reference to the human actions that realize it’
I agree. A difference between us is that from my perspective this is not a problem. There are large bodies of research findings and theories (and methods) about predicting the ‘apparently unpredictability’ of human actions. Myself and other design research colleagues have been working in this area of design research since the 70s and the available material we used in the 70s is much older.
My feeling is that we do not disagree on many basic aspects of design as an activity (though I suggest a less restrictive definition of design would avoid excluding much of the work of most of the designers reading this list.). I suggest the main difference between our understanding is that I’m assuming that it is straightforward to predict and analyse the functioning and outcomes of many of the situations, particularly involving human subjective action, that you regard as unpredictable. This includes the details of the process and outcomes of human social and political negotiations relating to design. This kind of prediction can be done and is done and theorised about - with its own body of 'design theory'. The problem is that most designers don’t do it and design researchers using, or reacting against, positivist thinking assume it cannot be done or theorised about.
One of my side interests in developing design theory has been in intervening in and managing ultra-complex socio-political-technical situations. To make the design theory challenge more interesting, I wondered whether it would be possible to develop theory that would suggest effective strategies for managing these situations for those who are not in possession of power. To clarify, we are looking at situations that involve large numbers of different people in different groups with dynamically changing relationships(power and otherwise) that have a variety of uses of different and changing technologies and legitimating authorities that shape the dynamically changing balances of power and control. We assume actions happen across the system boundaries and we assume that we cannot assume that the system remains the same system. We also assume that we cannot know in detail about individuals, groups, legitimation and technologies – or their dynamics. In other words we assume that reductionist and positivist perspectives do not apply. Also we wanted to see if we could find design theories that simultaneously apply at the individual level
An interest was to see if it is possible to develop design theories that fully take into account social dimensions of these kind of design situations and would predict or identify:
• areas of optimal design solution
• bounds on likely areas of design solutions
• changes in physical, social, political and informatic attributes of design solutions varying across the multiple dimensions of design contexts and solution space
• design principles, heuristics and guidelines
• theoretical approaches that provide design solution optimisation
So far, we have identified several design theories about social actions and interactions that fulfil these roles. We have published details of 6 of them.
I agree with you about seeing design in terms of interventions. For several years, this has been the underpinning theory approach we have been using We have found that when designs are seen as interventions in socio-technical systems, then much of the need for and categorisation of existing bodies of design theories disappears. Trist and the work at the Tavistock clinic went down this path many years ago (http://www.moderntimesworkplace.com/archives/archives.html )
Best wishes and thanks for your comments,
Terry
-----Original Message-----
From: PhD-Design - This list is for discussion of PhD studies and related research in Design [mailto:[log in to unmask]] On Behalf Of Klaus Krippendorff
Sent: Tuesday, 10 August 2010 1:48 AM
To: [log in to unmask]
Subject: Re: types of design research
dear ken, terry and others who contributed to this thread
i was abroad and followed the thread on my blackberry.
(1)
terry's conception of design research having to be predictive implies an epistemology that is antithetical to design -- unless terry means something altogether different than forecasting individual behaviors.
to me and in strict scientific terms, prediction is an extrapolation of past pattern into the future. it's success depends on the extent a previously identified pattern persists, continues unchanged by human observation and action.
predictive theories and evidence for the validity of predictions is the bread and butter of scientific research. but for designers to buy into this paradigm trivializes design. in my "semantic turn" i suggested that design brings forth something that cannot be predicted without reference to the human actions that realize it. so, design is inherently innovative, concerns itself with changing something that would not come about naturally, cannot be predicted from past observations.
to me, design is an inherent social activity, affecting others' lives. one target of empirical inquiries of interest to designers concerns the understanding of the stakeholders of a design, whether they are willing to commit themselves to realize a design or use it in the process of affecting others.
(2)
i agree with ken on the ambiguity of freyling's distinctions. in the "semantic turn" i adopted nigel cross' distinctions between:
* science(s) OF design, taking design as an object of research from various disciplinary perspectives, such as cultural history (margolin), psychology (norman), cognitive science, sociology, ...
* "design science, ... an explicitly organized, rational and wholly systematic approach to design"
from which i distinguish a
* a science FOR design, a systematic way of making the practice of design communicable: for designers to more efficiently work together; independent of each other to be able to examine design successes and failures and draw debatable lessons from them; to introduce design students into the design profession; to compellingly justify a design (proposal, suggestion) to needed stakeholders; and improve the design discourse and with it the design profession.
(3)
i am somewhat allergic to using the word "research" without much reflection on what is involved. to me, "re-search" (and i know that ken prefers the french interpretation of the word while i write in english) means repeatedly searching for generalizable patterns that underlie available data. if one takes the task of a science for design seriously, one would have to SEARCH for (a) what is changeable (not what persists), (b) who, which stakeholders, resist or support design interventions and what would need to be done to overcome the obstacles to a design; (c) what technological, material, individual, social and political resources are available or recombinable to realize a design. (d) the sole purpose of (a) through (c) is not to predict or understand for its own sake but to provide convincing arguments (justifications) for a design to be acceptable to interested stakeholders (so that they can take up their stake in it).
in light of criterion (d), one could say that all inquiries of interest to designers are to find ways to convince the community of stakeholders to back a design and eventually realize it. terry's prediction of behavior misses the essential feature of human commitment to make something real.
you may call these activities design research (and get trapped in positivist practices) or searches for compelling arguments or justifications (which i prefer). predictions focus on phenomena that are besides the point of design activity. in my conception, designers are innovators in the domain of (material) culture, intervene into ongoing social practices by encouraging novel interfaces with technology, and, by this definition, undo what is predictable.
instead of talking about "design research," we might well talk about "creating support for interventions."
klaus
rrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr
-----Original Message-----
From: PhD-Design - This list is for discussion of PhD studies and related research in Design [mailto:[log in to unmask]] On Behalf Of Ken Friedman
Sent: Sunday, August 08, 2010 7:42 PM
To: [log in to unmask]
Subject: Re: types of design research
Dear David and Terry,
There is are interesting distinction in this interchange that are worth
considering:
[1]
Design research may not be about improving outcomes, but design itself
does involve improving outcomes or creating preferred states against
current states.
Design research involves a wide range of questions, and only some of these
involve teleology or improving outcomes.
Design is teleological. We design to create outcomes.
I'm not disagreeing with anyone here, but teasing out a distinction, and
this includes an aspect of the distinction that distinguishes between design
research and design.
[2]
Terry argues that all design research involves "predicting behavioural outcomes
to improve them."
To me, that is a sweeping statement. I can conceive of many forms of design
research that involve other goals. There is also the need to clarify the term
"behavioral outcomes." If by behavior, Terry means the behavior and function
of metals or chemicals or artifacts as well as human behavior, that statement
has one range of meanings. If Terry means behavior and interaction between
artifacts and surrounding systems or end users, it has a different meaning.
Even though I will argue that many kinds of design research do not involve
prediction, I'd like to know what kinds of behavior Terry intends to cover
in this statement.
[3]
Without agreeing that all forms of design research involve prediction, I agree
that we can understand and predict far more than we understand and predict
successfully today. That is clearly one purpose of design research, and a
valuable purpose.
One reason I value Terry's work so highly is that he spends so much time
carefully and patiently working through the literature and practice of multiple
design fields, applying what he learns to the process and practice of design.
This has several consequences. One is a specific consequence of Terry's
background in engineering and computation. On the one hand, this means
that Terry seeks measurable and predictable outcomes. On the other, this
limits the ambiguous and interpretive. That simply bugs some of us -- and
I occasionally find it frustrating. The second consequence is general. Terry
approaches issues in a scientific manner. This means the rest of us must work
hard just to keep up with and understand Terry's work. Many of us find the
demands on our time difficult -- we can't follow Terry's work without wide
reading in fields where we do not often go. The third consequence follows
from the first two: Terry is sometimes wrong. That is what happens when
people actually work in a scientific manner. Some experiments fail, some
hypotheses prove wrong. Terry has a genuine ethos of scientific inquiry:
he wants to know whether his ideas prove out. In the grand style of Karl
Popper's philosophy of science, he proposes bold hypotheses and tests
them to build on what works while he cheerfully discards what doesn't.
Terry is engaged in a long-term, progressive research program in several
areas of design. In this respect, he is a model researcher. This is not the
only model of research, to be sure, but I am glad that Terry is one of us,
and I value his work.
The third paragraph -- below -- is a typical Terry Love statement, and Terry
really does work on the issues that he raises. What's so puzzling to me is
that we have too few people in our field doing the very necessary kind of
work that Terry does. I think that will change as more engineers, logicians,
mathematicians, and physicists become interested in design.
Warm wishes,
Ken
On Sun, 8 Aug 2010 23:08:07 +0800, Terence Love <[log in to unmask]> wrote:
[1]
>You say that ' Design research, such as my own, into the philosophy of
>design is not necessarily about improving outcomes, nor is critical research
>of the kind done by Victor Margolin.'
[2]
>I claimed that at heart all design research has an underlying focus on
>'predicting behavioural outcomes to improve them'. I suggest that this is
>true of both your research and Viktor's although some research may appear to
>be less directly connected. Point me to a research paper that you feel this
>is not true for and lets test it.
[3]
>You say ' There is a vast area of human activity which falls outside the
>category of things that are potentially predictable'. I suggest 1) that in
>the areas in which designers work, this is much rarer than designers claim,
>and 2) where designers design in areas where behavioural outcomes are truly
>unpredictable then they lay themselves open to legal action against them
>(on what basis would they justify that their designs were any good/optimal/
>satisfied the brief?). Again, the test is to look at some examples. It is
>true that one cannot exactly predict the behaviour of some indeterminate
>systems. One can, however, predict a lot about them, and, systems that are
>totally unknown in terms of their behavioural outcomes are usually not
>terribly useful. Please give examples and we can work through them to test
>them.
|