QUERY Is SD capable of self-analysis?
Posted: Thu Jul 10, 2008 7:51 am
Posted by Jack Harich <register@thwink.org>
Imagine a patient walking into a psychiatrist's office and laying down on=
a
couch. The psychiatrist then proceeds to perform an analysis, using the
methods of psychoanalysis.
But is analysis possible if the patient and the analyst are the same pers=
on?
Probably not.
That may be the case here. Self-analysis is generally impossible, because=
you
cannot see yourself as others see you. There are blind spots that cannot =
be
detected and hence anything in those areas cannot be treated.
This paradox arose when Paul Holmstr=F6m and Jack Harich were discussing =
the
strategy thread. Jack asked: ""I wonder what other paradoxes/traps we are =
in
right now but don't know it?"" Paul replied with this insightful observati=
on:
""Another paradox that I think some are stuck in is that you can use a
particular methodology to analyze itself. I have read quite a lot of gene=
ral
systems theory and am trying to recall references to that such
self-referentiality is impossible. Some systems theorist also said that
'everything that is said is said by an observer, to another observer',
implying that one is always part of the system that one is observing. In
other words it is impossible to build a SD model that does not inherit ou=
r
own biases.""
Thomas Kuhn dealt with this phenomenon at length in ""The Structure of
Scientific Revolutions."" (SSR) Consider this quote from
http://en.wikipedia.org/wiki/The_Struct ... ions#Tran=
sition_period
- - - - - - - - - Begin Quote - - - - - - - - -
""The period of transition between established paradigms is neither smooth=
nor
rapid. Kuhn quotes Max Planck's observation (SSR, p. 151) that sometimes:
"" '...a new scientific truth does not triumph by convincing its opponents=
and
making them see the light, but rather because its opponents eventually di=
e,
and a new generation grows up that is familiar with it.'
""According to Kuhn, the scientific paradigms preceding and succeeding a
paradigm shift are so different that their theories are not comparable
(incommensurable). The paradigm shift does not merely involve the revisio=
n or
transformation of an individual theory, it changes the way terminology is
defined, how the scientists in that field view their subject, and, perhap=
s
most significantly, what questions are regarded as valid, and what rules =
are
used to determine the truth of a particular theory. Kuhn observes that th=
ey
are incommensurable =97 literally, lacking comparability, untranslatable.=
The
new theories were not, as the scientists had previously thought, just
extensions of old theories, but were radically new world views. Such
incommensurability exists not just before and after a paradigm shift, but=
in
the periods in between conflicting paradigms. It is simply not possible,
according to Kuhn, to construct an impartial language that can be used to
perform a neutral comparison between conflicting paradigms, because the v=
ery
terms used are integral to the respective paradigms, and therefore have
different connotations in each paradigm. The advocates of mutually exclus=
ive
paradigms are in an invidious position: 'Though each may hope to convert =
the
other to his way of seeing science and its problems, neither may hope to
prove his case. The competition between paradigms is not the sort of batt=
le
that can be resolved by proof.' (SSR, p. 148).
""Kuhn (SSR, section XII) states that the probabilistic tools used by
verificationists are inherently inadequate for the task of deciding betwe=
en
conflicting theories, since they belong to the very paradigms they seek t=
o
compare. Similarly, observations that are intended to falsify a statement
will fall under one of the paradigms they are supposed to help compare, a=
nd
will therefore also be inadequate for the task.""
- - - - - - - - - - End Quote - - - - - - - - -
To objectively self-analyze itself, SD practitioners would have to be abl=
e to
look at SD through several world views, as appropriate. This they cannot =
do,
because by habit they can only see the world (which includes themselves)
through the lens of SD.
When an alternative or improvement to SD is presented, it will be judged =
by
the theories and terminology of SD - Not the many competing theories, som=
e of
which probably offer a better, more productive world view.
(1) For example, suppose the field of SD is best represented as an
intelligent agent in a model. It mutates, makes decisions based on changi=
ng
rules, learns, etc. SD models have only stocks, rates and auxiliary
variables. Thus SD practitioners would tend to not represent SD (and its
competitors) optimally when modeling it to analyze the SD immaturity and =
too
slow evolution/learning rate problem. They would have a blind spot, one t=
hey
never knew was there.
(2) As another example, the more difficult and complex the problem, the m=
ore
important it is to be process driven. The field of project management has
accepted this principle. So has the field of systems engineering, as we
recently saw at: http://www.incose.org/practice/fellowsconsensus.aspx Thi=
s
page states that:
""Systems Engineering is an engineering discipline whose responsibility is
creating and executing an interdisciplinary process to ensure that the
customer and stakeholder's needs are satisfied in a high quality,
trustworthy, cost efficient and schedule compliant manner throughout a
system's entire life cycle. This process is usually comprised of the
following seven tasks: State the problem, Investigate alternatives, Model=
the
system, Integrate, Launch the system, Assess performance, and Re-evaluate.
These functions can be summarized with the acronym SIMILAR: State,
Investigate, Model, I ntegrate, Launch, Assess and Re-evaluate. This Syst=
ems
Engineering Process is shown in Figure 1.""
SD has never accepted this principle. It has no standard process, much le=
ss a
process that has been continually improved. SD practitioners mostly use a=
d
hoc approaches to solve problems. The implicit assumption is that SD is s=
o
inherently potent it doesn't need to be process driven and that the field
doesn't need to focus on a better process - just a better modeling tool. =
So
when a competing theory is presented that says SD needs to be process dri=
ven,
it is automatically rejected because it is incommensurable with the old
theory.
(3) As a third example, suppose someone proposed that SD analysts start
thinking in terms of root causes. The SD paradigm would automatically rej=
ect
this theory, because it sees system behavior as an emergent property of a
system's entire structure.
(4) Or suppose SD is best encapsulated and developed as a subset of syste=
ms
engineering. It's one of many modeling tools, and a relatively insignific=
ant,
less popular one at that. It achieves its potential best by working with =
a
suite of finely honed collaborating micro-tools wrapped in a single
application process. But the SD community cannot see this as an alternati=
ve
theory, because they are enamored with the potential of the tool, due to
reading the inspiring works of the field's founder and seeing the
sensational, international success of the Limits to Growth project.
(5) Finally, there is probably a vastly better and different way to repre=
sent
and model complex systems. This way would be so much better it could
accommodate many types of systems. It would be the Rosetta Stone of model=
ing.
But this possibility is not being actively considered, because SD is seen=
as
good enough. After all, it hit huge home runs with Urban Dynamics and Lim=
its
to Growth. All we need to do is step up to the plate again with the same
tool, wait for the right pitch, and we will hit another home run. So the
self-talk goes....
So, is SD capable of self-analysis? Is it capable of solving its own
immaturity problem?
Jack Harich and Paul Holmstr=F6m
Posted by Jack Harich <register@thwink.org>
posting date Wed, 09 Jul 2008 11:39:14 -0400
_______________________________________________
Imagine a patient walking into a psychiatrist's office and laying down on=
a
couch. The psychiatrist then proceeds to perform an analysis, using the
methods of psychoanalysis.
But is analysis possible if the patient and the analyst are the same pers=
on?
Probably not.
That may be the case here. Self-analysis is generally impossible, because=
you
cannot see yourself as others see you. There are blind spots that cannot =
be
detected and hence anything in those areas cannot be treated.
This paradox arose when Paul Holmstr=F6m and Jack Harich were discussing =
the
strategy thread. Jack asked: ""I wonder what other paradoxes/traps we are =
in
right now but don't know it?"" Paul replied with this insightful observati=
on:
""Another paradox that I think some are stuck in is that you can use a
particular methodology to analyze itself. I have read quite a lot of gene=
ral
systems theory and am trying to recall references to that such
self-referentiality is impossible. Some systems theorist also said that
'everything that is said is said by an observer, to another observer',
implying that one is always part of the system that one is observing. In
other words it is impossible to build a SD model that does not inherit ou=
r
own biases.""
Thomas Kuhn dealt with this phenomenon at length in ""The Structure of
Scientific Revolutions."" (SSR) Consider this quote from
http://en.wikipedia.org/wiki/The_Struct ... ions#Tran=
sition_period
- - - - - - - - - Begin Quote - - - - - - - - -
""The period of transition between established paradigms is neither smooth=
nor
rapid. Kuhn quotes Max Planck's observation (SSR, p. 151) that sometimes:
"" '...a new scientific truth does not triumph by convincing its opponents=
and
making them see the light, but rather because its opponents eventually di=
e,
and a new generation grows up that is familiar with it.'
""According to Kuhn, the scientific paradigms preceding and succeeding a
paradigm shift are so different that their theories are not comparable
(incommensurable). The paradigm shift does not merely involve the revisio=
n or
transformation of an individual theory, it changes the way terminology is
defined, how the scientists in that field view their subject, and, perhap=
s
most significantly, what questions are regarded as valid, and what rules =
are
used to determine the truth of a particular theory. Kuhn observes that th=
ey
are incommensurable =97 literally, lacking comparability, untranslatable.=
The
new theories were not, as the scientists had previously thought, just
extensions of old theories, but were radically new world views. Such
incommensurability exists not just before and after a paradigm shift, but=
in
the periods in between conflicting paradigms. It is simply not possible,
according to Kuhn, to construct an impartial language that can be used to
perform a neutral comparison between conflicting paradigms, because the v=
ery
terms used are integral to the respective paradigms, and therefore have
different connotations in each paradigm. The advocates of mutually exclus=
ive
paradigms are in an invidious position: 'Though each may hope to convert =
the
other to his way of seeing science and its problems, neither may hope to
prove his case. The competition between paradigms is not the sort of batt=
le
that can be resolved by proof.' (SSR, p. 148).
""Kuhn (SSR, section XII) states that the probabilistic tools used by
verificationists are inherently inadequate for the task of deciding betwe=
en
conflicting theories, since they belong to the very paradigms they seek t=
o
compare. Similarly, observations that are intended to falsify a statement
will fall under one of the paradigms they are supposed to help compare, a=
nd
will therefore also be inadequate for the task.""
- - - - - - - - - - End Quote - - - - - - - - -
To objectively self-analyze itself, SD practitioners would have to be abl=
e to
look at SD through several world views, as appropriate. This they cannot =
do,
because by habit they can only see the world (which includes themselves)
through the lens of SD.
When an alternative or improvement to SD is presented, it will be judged =
by
the theories and terminology of SD - Not the many competing theories, som=
e of
which probably offer a better, more productive world view.
(1) For example, suppose the field of SD is best represented as an
intelligent agent in a model. It mutates, makes decisions based on changi=
ng
rules, learns, etc. SD models have only stocks, rates and auxiliary
variables. Thus SD practitioners would tend to not represent SD (and its
competitors) optimally when modeling it to analyze the SD immaturity and =
too
slow evolution/learning rate problem. They would have a blind spot, one t=
hey
never knew was there.
(2) As another example, the more difficult and complex the problem, the m=
ore
important it is to be process driven. The field of project management has
accepted this principle. So has the field of systems engineering, as we
recently saw at: http://www.incose.org/practice/fellowsconsensus.aspx Thi=
s
page states that:
""Systems Engineering is an engineering discipline whose responsibility is
creating and executing an interdisciplinary process to ensure that the
customer and stakeholder's needs are satisfied in a high quality,
trustworthy, cost efficient and schedule compliant manner throughout a
system's entire life cycle. This process is usually comprised of the
following seven tasks: State the problem, Investigate alternatives, Model=
the
system, Integrate, Launch the system, Assess performance, and Re-evaluate.
These functions can be summarized with the acronym SIMILAR: State,
Investigate, Model, I ntegrate, Launch, Assess and Re-evaluate. This Syst=
ems
Engineering Process is shown in Figure 1.""
SD has never accepted this principle. It has no standard process, much le=
ss a
process that has been continually improved. SD practitioners mostly use a=
d
hoc approaches to solve problems. The implicit assumption is that SD is s=
o
inherently potent it doesn't need to be process driven and that the field
doesn't need to focus on a better process - just a better modeling tool. =
So
when a competing theory is presented that says SD needs to be process dri=
ven,
it is automatically rejected because it is incommensurable with the old
theory.
(3) As a third example, suppose someone proposed that SD analysts start
thinking in terms of root causes. The SD paradigm would automatically rej=
ect
this theory, because it sees system behavior as an emergent property of a
system's entire structure.
(4) Or suppose SD is best encapsulated and developed as a subset of syste=
ms
engineering. It's one of many modeling tools, and a relatively insignific=
ant,
less popular one at that. It achieves its potential best by working with =
a
suite of finely honed collaborating micro-tools wrapped in a single
application process. But the SD community cannot see this as an alternati=
ve
theory, because they are enamored with the potential of the tool, due to
reading the inspiring works of the field's founder and seeing the
sensational, international success of the Limits to Growth project.
(5) Finally, there is probably a vastly better and different way to repre=
sent
and model complex systems. This way would be so much better it could
accommodate many types of systems. It would be the Rosetta Stone of model=
ing.
But this possibility is not being actively considered, because SD is seen=
as
good enough. After all, it hit huge home runs with Urban Dynamics and Lim=
its
to Growth. All we need to do is step up to the plate again with the same
tool, wait for the right pitch, and we will hit another home run. So the
self-talk goes....
So, is SD capable of self-analysis? Is it capable of solving its own
immaturity problem?
Jack Harich and Paul Holmstr=F6m
Posted by Jack Harich <register@thwink.org>
posting date Wed, 09 Jul 2008 11:39:14 -0400
_______________________________________________