On the occasion of the current discussion about organizations on
the net, I would like to offer a few humble speculations on the
topic.
1. There is a difference between arguing that organizations
should be control systems, that a particular organization is a
control system and that all organizations are control systems.
The first is a normative statement. The second and third are
empirical questions. In my own field work in a particular
organization, I have not found it useful to consider it a control
system. If it is controlling something, I don't know what it is.
Therefore, I would have to say now that the final statement is
false.
2. It is extremely important to consider the level of analysis. In
my research I am interested in how living control systems
interact in an organization. I find viewing people as control
systems to be highly useful in this research. If I were
interested in understanding why people are control systems
and building better ones (a quite different and difficult task), I
would have to understand about what happens inside people's
bodies, e.g., about neurons and networks of neurons. I would
have to understand the principles of these components and
how they interact.
Likewise, if I were to consider how organizations interact with
each other in the environment (an important question,
although one I am not currently working on), I would have to
understand the properties of the organizations' behavior, e.g.
that they are control systems (which I doubt is true). If I were
interested in making these organizations operate more
efficiently, I would have to understand what happens inside
them: about their components and how these components
interact. These components are people.
Someone who understands how organizations behave does not
necessarily know more about why they behave that way (and
how to construct and re-construct them) than someone who
knows how people behave (a sociologist?) knows about neural
nets.
3. This also raises the question: Who are we helping control
their perceptions? Is building an organization which is a better
control system always a good idea if the "reference signals" of
the organization are set by people at the top of the
organizational chart? Maybe it would be better if the
organization wasn't a control system at all but more people in
the organization were controlling their perceptions relative to
the organization. Although these issues are related, they are
not the same, nor is there necessarily a positive correlation
between them.
4. At the present, the logic underlying my research is
something like this: People in my organization are control
systems. I want to find out some of the things they are
controlling for. I want to know how the environment interacts
with a person's behavior, which results in control or lack of
control. This environment consists of variables which are being
controlled for by the individual in question, variables which
are being controlled for by other individuals, these other
individuals and their behavior (both control and byproducts of
control), disturbances & obstacles, and tools & resources. Of
course, these different variables are systematically, not
randomly distributed, in the organization. For example, the
director has different obstacles to negotiate than a staff
member does.
There are a number of ways to organize research along these
lines. At the moment I am focusing on a particular class of
objects in the organization, the rules, and attempting to identify
their relationships with individuals in the organization. Rules
are both tools and disturbances in a number of interesting
ways, some of which are unintended by those who write them.
Rules can also be controlled variables (some people want to
write new ones or get rid of old ones). Rules are also suggested
reference signals. I might add that rules do nothing on their
own; someone has to enforce them (use them as a tool) and
different people use them in different ways.
5. This is my rough, evolving view of an organization (as
written on a particular Tuesday evening): An organization is a
site for interaction among living control systems, each one with
a unique set of reference signals. These individuals aren't able
to control all of their perceptions without help from other
people. The organization is a social construction created and re-
created by people to help them control some of their
perceptions. The organization is itself a high-level perception,
which is controlled for by most of the people involved in it
(although in different ways). There are also many other similar
and related reference signals which people must accept for the
organization to do what it does. What the organziation does is
the result of what all the people in and around it do. The
environment in and around the organization is filled with all
sorts of variables which can be controlled, including objects and
ideas, as well as all sorts of resources & tools, obstacles and
disturbances. Lots of important and interesting things happen
in organizations.
David Schweingruber
University of Illinois at Urbana-Champaign
dschwein@ux1.cso.uiuc.edu