Resource structure of collaboration and learning organizatio

[Michael Fletcher]

Greetings:

I been investigating the resource structure that is required to foster
collaboration. As it turns out, it looks like what I thought the
intangible resource structure of Learning Organizations probably looks
like. This was not a big surprise since to my thinking collaboration is
mostly the natural outgrowth of healthy social systems moving towards
some shared goal, and Learning Organizations are healthy systems. The
structure of the two will therefore have some similarity.

Looking through the usual sources, I wasn't able to find anything those
spoke specifically to these issues, but since this cuts across a couple
of fields the potential literature is very large. I anyone aware of
some publication or document that addresses these issues in this manner?

Thanks,

Michael Fletcher
mefletcher@speakeasy.net

[Jean-Jacques Laublé]

Hi Mike

Collaboration leads the world. I can see two different kinds of
collaboration:

collaboration between people with or not with hierarchical links.

In my business where collaboration is critical, I have found that the best

way to foster collaboration is to share the problems of the people by
working closely

with them and being able to do their job with them. This fosters mutual
esteem

and helps the hierarchy to design realistic policies that increases
participation.



You can find in the open course ware http://ocw.mit.edu two interesting

courses with plenty of links from the Sloan school of management called

'power and negotiation' and 'negotiation and conflict management'.

Negotiation is a close brother of participation and to my opinion
necessitates

the same kind of surrounding.



I think that the fact that in many organisations, people at the head do not
know

the work of the people underneath them by doing it, is the principal cause
of non participation.



This point has to do with S.D. because I suspect that S.D. models

that are constructed with or by people that do not know the basic job of
their business

( intellectual versus realistic people)

are prone to be unrealistic and generate no usefulness.



Regards



J.J. Laublé, Allocar, rent a car

Strasbourg France

JEAN-JACQUES.LAUBLE@WANADOO.FR
From: =?iso-8859-1?Q?Jean-Jacques_Laubl=E9?= <JEAN-JACQUES.LAUBLE@WANADOO.FR>

[Ray on EV1]

I have just started reading a book called ""Linked"" by Albert-Laszlo
Barabasi. So fay it is providing a history of linear graph theory. There
are many comments about sociological uses of graph theory.

Raymond T. Joseph, PE
RTJoseph@ev1.net

[John Gunkler]

I hope your query isn't just supplying me an excuse to pontificate, but I've
been wanting to write about ""learning organizations"" for a long time.
Without a sensible notion of what that animal is, any attempt to study or
model mechanisms within that creature is doomed to frustration.

I'd like to suggest we avoid the sophisticated bickering and agree to start
with as simple a definition of a ""learning organization"" as will prove to be
useful. I will supply a candidate definition in a moment.

The first difficulty is that learning and educational psychology has had
trouble defining what ""learning"" is. After much wrangling, most now agree
to an operational definition that's something like:

Learning is a change in behavior due to experience.

Or, a bit more precisely:

The evidence that learning has occurred is an observable change in
behavior due to experience.

Now, as system dynamicists, we know there are dangers in the ""due to"" phrase
-- causality, we know, is not often simple and one-way. The reason for
including a reference to ""experience"" in the definition is that there are
other kinds of behavior change that few would want to ascribe to ""learning""
and what differentiates ""learning"" from other kinds of behavior change is
that learning requires a set of events that affect the person over time.

So, let's try again:

""The operational evidence that learning has occurred is an
observable change in behavior accompanying a series of one or more
interactions between the organism and its environment.""

I'm fairly content with this. Now, to take it into the organizational
context.

Change ""organism"" to ""organization."" That's simple enough. And we
certainly model interactions between organizations and their environments,
so that shouldn't present too many difficulties. Which leaves: ""observable
change in behavior"" -- what does this mean in a system dynamics view of an
organization?

We might be tempted to think of ""behavior"" as ""behavior modes"" -- those
""behavior over time graphs"" (BOTG's) we like to draw. After all, we already
use the word ""behavior"" in that way. And, I think, in a loose way (much the
way we sometimes talk loosely about people's behavior) it's all right to use
behavior modes as a way to describe the behavior of an organization. We
say, ""She bought a watch"" as a shorthand description of someone's behavior
-- but, actually, it doesn't describe ""behavior"" does it? It describes a
resultant of the person's behavior. One of the more useful tricks of
psychology is to get past the shorthand ways we have of talking about what
people do and get to the level of true descriptions of behavior. This is to
psychology what operational definitions are in SD modeling.

So, while BOTG's are a useful shorthand for organizational behavior in SD, I
think the real ""behaviors"" of an organization are its ""policies"" (in the
precise sense Forrester uses that term) -- where a policy is nothing more
nor less than what is captured by a rate equation (and any auxiliaries.)

But is that the only ""behavior"" of an organization or system? Probably.
Are rate equations the only thing that can change and be called ""learning?""
Maybe not. I can see at least one other candidate: the dynamic structure
of the system or a part of the system. As a result of policy changes,
organizations can add or eliminate stocks/accumulations and flows (of
materials or information.) This, too, should count as organizational
learning.

Where are we, then? With this picture of what organizational learning is:

THE EVIDENCE FOR ORGANIZATIONAL LEARNING IS AN OBSERVABLE CHANGE IN
POLICY (the nature of the flows of materials or information) OR DYNAMIC
STRUCTURE (the existence of stocks and flows and their connections with each
other) ACCOMPANYING A SERIES OF ONE OR MORE INTERACTIONS BETWEEN THE
ORGANIZATION AND ITS ENVIRONMENT.

From: ""John Gunkler"" <jgunkler@sprintmail.com>

[Philip Keogh]

Hi All

Might I suggest you look at the Learning Organisation List
(http://learning-org.com). It might save re-inventing the
wheel and I have found it a very useful community.

[Host's note: Thanks for this direction Philip. I recommend
the learning organization list for those who wish to pursue
this further.

I am today posting pending messages on this subject but will not
post any more that are not strongly related to system dynamics.]


Regards

Philip Keogh
From: ""Philip Keogh"" <Philip.Keogh@leedsth.nhs.uk>

Pathology Information Officer
(see our website at www.leedsteachinghospitals.com)

[Bill Braun]

John, I propose a slight reformulation of the definition, which
(to my way of thinking) is more robust reflection of your own
thinking.

Yours:

The evidence for organizational learning is an observable
change in policy (the nature of the flows of materials or
information) or dynamics structure (the existence of stocks and
flows and their connections with each other) accompanying a
series of one or more interactions between the organization and
its environment.

Revised:

The evidence for organizational learning is an observable
change in results/outputs as a function of changes in policy
(the nature of the flows of materials or information) or
dynamic structure (the existence of stocks and flows and their
connections with each other), both of which are [re]designed in
response to changes in the relationship between an organization
and its environment.

My premise is that problem symptoms are indicators of gaps or
misalignments in structure, which lead to inquiries into the
nature of the relationship between organization and
environment, which lead to insights into the actual gap, which
lead to the redesign of structure which ameliorates the
problematic observable symptoms. (This is inclusive of looking
at systems of symptoms, not each one taken separately.)

If an organization's environment does not change, once the
nature of the environment is known (I assume for this
discussion that it can be completely known, eventually), and
the organization has adapted to the environment, in theory no
additional learning is required to maintain peak performance.
If the environment then changes, a deterioration of peak
performance (the symptoms) would be the first indicator that a
gap has opened. The symptom is an early indicator that learning
is needed - then the inquiry, then insight, then redesign.

This definitional reformulation appears to encompass your
reference to the addition or deletion of stocks/accumulations
and flows.

Is the reformulation consistent with your thinking from which
the definition you presented evolved?

Bill Braun
From: ""Bill Braun"" <bbraun@hlthsys.com>

[Finn Jackson]

John Gunkler makes an excellent introduction to the topic of Learning
Organizations, and I look forward to reading further responses.

The point I want to make is that the analysis he has applied to ""what do we
mean by 'Learning'?"" also needs to apply to ""Organizations"".

If we want to be precise, which I think we do, then we have to realise that
it is difficult (if not impossible) to make a distinction between an
""Organism/Organization"" and its ""Environment"".

I know the two words *seem* to be very clear. But they are not.

When we think about systems that change, learn, or ""evolve"" together, we
have to realise that the one changes the other, and it is *both together*
that make the 'system'.

For example, when the height of available food encouraged giraffes
(""organism"") to grow longer necks, that encouraged trees (""environment"") to
grow taller to escape them (I recognise that this is a simplification). It
was the ""organism"" (giraffe) AND its ""environment"" (tree) TOGETHER that
evolved (or ""learned""). And similarly with dung beetles, hummingbirds,
mangroves, ....


It is the organism-AND-environment together that make up the 'System' that
evolves.
[For more on this see ""Steps Toward an Ecology of Mind"", Gregory Bateson.]

Or, better, there is no such thing as ""environment"" there are only ""other
organisms"".


This may seem picky, because ""we all know what we mean by a Learning
Organisation, don't we"".

But I think when we apply it to ""organizations"" instead of ""organisms"" then
it becomes even more difficult. Does ""the system"" only include people on the
payroll of the ""organization""? What about people who only work part-time?
What about organizations without a payroll? What about consultants and
strategic partners, both of which can have a major impact on the abilty of
the organization to learn and to act on its learning, and yet both of which
appear to be ""outside"" the organization.


The more I think about it, the less I see a separation between
""Organisation"" and ""Environment"".

I know this doesn't help address the problem. But I do think it is real.

And the conclusion I come to is that the reason the study of Learning
Organizations has proved so difficult is that we are thinking about the
problem in the wrong way, with the wrong mental models. We are asking the
wrong question.

I look forward to the upcoming posts.


Finn Jackson
From: ""Finn Jackson"" <Finn.Jackson@Tangley.Com>

[Jean-Jacques Laublé]

Hi John



You say that Learning is a change in behaviour due to experience.


But Why use the word 'Learning' ? For that definition I would prefer the

word adaptation.




The word 'learning' conveys a positive and normative idea that is not always
associated

with change in behaviour.


A child with 'bad' parents adapts his behaviour towards the external word
according

to his 'bad' experience. Is he learning? Or more precisely is this a
positive experience

that will help him in the rest of his life?




A child educated in a sect, may 'learn' things that do not correspond to the
current norms.

He is obviously not learning in the sense of traditional moral.



One solution would be to add to the 'change in behaviour due to experience'
'motivated

by the accomplishment of an objective.'

This way the learning can be bad or not useful if the objectives are not
well settled.



In a completely different register, there is a strict mathematical
definition of learning

based on statistical concepts exposed by instance in the MIT open course
ware department

of brain and cognitive sciences, course Statistical Learning Theory and
Applications.

http://ocw.mit.edu.

It is a very elegant theory, but I do not know if it has practical
applications. It has some

similarity with S.D. theory.

There are other interesting courses in this department like

Probability and Causality in Human Cognition which is partially based on

Bayesian statistics.



Regards.
From: =?iso-8859-1?Q?Jean-Jacques_Laubl=E9?= <JEAN-JACQUES.LAUBLE@wanadoo.fr>



J.J. Laublé Allocar rent a car company

Strasbourg France

JEAN-JACQUES.LAUBLE@WANADOO.FR

[Stephen Wehrenberg]

John recognizes that most disagreement has its roots in semantic
differences ... particularly unstated differences ... so trying to nail
down a common meaning for organizational learning seems like a necessary
first step.

I read, reread, and rereread both John's logic and Jay Forrester's
definition(s) of policy, and can't help but feel there is something
missing in John's definition below. I assume that John refers to Jay's
third level description of policy wherein some rational, logical, basis
exists that describes the conversion or ""transfer function"" that is the
decision maker's causal model (implicit or explicit).

In the definition below, John, do you mean ""accompanying"" to assert more
than just coincidence? I can imagine an interaction between an
organization and its environment (event) that is concurrent with an
unrelated change in policy or dynamic structure. In other words, Jay's
first and second level description of decisions as policy ... decisions
NOT based on any notion that ""if I do A, perhaps it will cause B and
create a different (event),"" void of a theory of causation.

When I think of learning in the Kolb cycle sense, it has a component of
intent -- intending a different result (event), intentionally setting
out through some action to create that result (experimenting),
monitoring the result to see if the intention was achieved, then using
the difference between intended and actual results to modify a causal
theory and subsequent attempts to create the result.

I'm a behaviorist (actually, social behaviorist), so I can see the
concept of operant conditioning as learning, the foundation of your
description below. But I also see a non-trivial difference between
Skinner's learning (association), and Kolb's learning
(experimentation). I'd like to separate superstitions from intentional
learning in your definition by adding one word within, and a clause to
the end:

THE EVIDENCE FOR ORGANIZATIONAL LEARNING IS AN OBSERVABLE, intentional, CHANGE IN
POLICY (the nature of the flows of materials or information) OR DYNAMIC
STRUCTURE (the existence of stocks and flows and their connections with each
other) ACCOMPANYING A SERIES OF ONE OR MORE INTERACTIONS BETWEEN THE
ORGANIZATION AND ITS ENVIRONMENT, where each interaction informs the nature of
the change in policy.

That's a little awkward, but I need the loop, I guess. Thanks for
stimulating my incoherence!

Steve
From: Stephen Wehrenberg <stephen.wehrenberg@verizon.net>

Stephen B. Wehrenberg, Ph.D.
Human Resources Capability Development and
Director, Future Force, US Coast Guard
Organizational Sciences, The George Washington University
stephen.wehrenberg@verizon.net

[Michael Fletcher]

Thanks to everyone who has responded so far. Although investigating what
the potential resource structure of the Learning Organization is very
interesting, it wasn't my primary purpose. I was more interested in how
to develop strategies for enabling collaboration & data sharing. My
basic premise was that most attempts to ""engineer"" collaboration are
counterproductive; the only way to do it effectively is in essence to
build a healthy organization where collaboration will naturally occur as
needed. (That is, pushing out memo's or directives to a dysfunctional
organization ordering them to ""collaborate or else!"" is a sure way to
ensure policy resistance and continued dysfunction. Unfortunately, I've
seen attempts at exactly these type of strategies.)

I simply used the learning organization (or one possible view of its
resource structure) as one possible example of what a healthy
organization should look like. From the above, it follows that the
healthy organization collaborates as needed, when needed. Since the
paper was for Dr. Warren's course I used his Strategy Dynamic framework
to suggest a way to go about building the ability to collaborate
sensibly (and from the right direction.)

Some additional work and research I've done since completing the paper
points in some rather interesting directions, and potential future
efforts. There are some hints that collaboration isn't the end-all
panacea that many organizations think it is. Although I didn't go into
it in any length I argued that collaboration is goal focused and isn't a
continual condition of any organization. Even organizations which
collaborate well will disband those efforts when the goal is reached.
There are also some indications that there are valid knowledge reasons
why organizations collaborate during fast moving crisis but then disband
those efforts once the crisis is past. A small model of
collaborative/non collaborative learning vs. a dynamic data set appears
to show that collaboration wasn't all that much more beneficial against
slowly changing data sets, but if people didn't collaborate under
conditions of rapid change they were at risk of knowing nothing! Like I
said, potential future work. I'd certainly be interested to hear what
others think or have done in this area.

Mike Fletcher
From: Michael Fletcher <mefletcher@speakeasy.net>

[Sheldon Friedman]

I have been observing with interest the thread on collaboration. Juhn Gunkler describes behaviors as the results of policy. In modeling we try to capture those policies. However, some behavior is not the result of clear policy, but the results of long term social interactions, we call it culture. These behaviors trap not only unwritten rules, but emotional tones that people and groups have invested in. It is this area that often does not get captured in the modeling that we do. Yet it is often this area that can mask the written policy of the organization. Maybe this is why capturing true collaboration and enacted beliefs is as important as the offical line we often see or hear about.

Shelly Friedman
From: ""Sheldon Friedman"" <sheldon.friedman@comcast.net>