[Moderator's note: This thread led to an extensive discussion of a theme that was discussed face-to-face in Crete.]

Topic:	Consider: Change via inquiry vs. social networks (1 of 11)
Conf:	SIG: Sys Appl in Business & Industry
From:	David Ing
Date:	Sunday, November 10, 2002 12:25 PM

In discussions we've had on the Webboard (and and offline telephone call to Bela A. Banathy), it appears that some of the systems community isn't familiar with the literature on inquiring systems. If you're one of these people, a key figure in this area is C. West Churchman, who is featured as one of the luminaries of the ISSS, at http://www.isss.org/lumCWC.htm .

I propose that a review of the literature on inquiring systems isn't sufficient, but that we should challenge the idea of making social change (or more specifically, organizational change in businesses) through inquiring systems. The key question to consider could be: is an inquiring systems approach to social change too rational? I use the concepts of inquiring systems in personal practice, but I wonder if anyone can actually use them properly on a large scale.

The foundation work (which you'd have to get from a library, because it's out of print) is C. West Churchman, "The Design of Inquiring Systems: Basic Concepts of Systems and Organizations", Basic Books, 1971. A wonderful substitute (which I and others have found to be more readable) is Ian I. Mitroff & Harold A. Linstone, "The Unbounded Mind: Breaking the Chains of Traditional Business Thinking", Oxford University Press, 1993. Mitroff & Linstone provide a definition in their Chapter 1:

"An Inquiry System, or IS for short, is a system of interrelated components for producing knowledge on a problem or issue of importance. Since the IS's which we describe in the following chapters differ radically in almost everyone of their critical components, we want to understand why by comparing them systematically. Further, since the term "system" is by now so overworked that it is meaningless, we want to show that it still has important and valid applications."

They have a nice diagram with four elements: inputs; operator; outputs; and the guarantor. The guarantor is the key element that defines the type of inquiring system. There are then five "ways of knowing" related to various philosophers: inductive-consensual [Locke]; deductive-analytic [Leibniz]; multiple realities [Kant]; conflict (and dialectic) [Hegel]; and multiple perspectives (along a vector of progress) [Singer].

Mitroff & Linstone is an thin, inexpensive paperback worth reading, but if you want to read a web page, surf over to http://www.cba.uh.edu/~parks/fis/inqorg.htm to get a feeling around "Inquiring Organizations" by Courtney, Croasdell and Paradice (1998).

I've also cited that Singerian inquiring systems have been applied (to varying degrees of success) in business. The easiest references to find are Vincent P. Barabba and Gerald Zaltman, "Hearing the Voice of the Market: Competitive Advantage through Creative Use of Market Information", Harvard Business School Press, 1991, and Vincent P. Barabba, "Meeting of the Minds: Creating the Market-Based Enterprise", Harvard Business School Press, 1995.

The challenge I've discovered (inside IBM working with some consultants who were prior associates with Barabba) is that it's really difficult to maintain the purity of a Singerian inquiring system. In a cynical view, I could say that the systems concepts are really correct, but people just won't behave in appropriate ways.

Another, more populist view is that we should be looking at social change through social networks, and the ideas of social contagion. A popular bestseller is Malcolm Gladwell, The "Tipping Point: How Little Things Can Make a Big Difference", Little Brown, 2000. If you want to do some web surfing, you could get a feel for Gladwell's style at http://www.gladwell.com/archive.html , reading the June 1996 article on "The Tipping Point", and the March 1997 article on "The Coolhunt".

In digging into the research foundations, a key source for Gladwell is Thomas J. Allen, whose web page can be found at http://sloancf.mit.edu/vpf/facstaff.cfm?ID=267&ProfType=F&sortorder=name . His study of technology gatekeepers led to Gladwell's depiction of "mavens", "connectors" and "salesmen". I entered this body of work through UCLA cultural anthropologist Karen Stephenson, whose publications are available at http://www.netform.com/html/knowledge_management.html . I just noted a link between Gladwell and Stephenson (and Jane Jacobs!) in Gladwell's Dec. 2000 "Designs for Working" article, available both on the Netform site, and the Gladwell archives. (Karen Stephenson describes "hubs", "gatekeepers" and "pulse-takers").

So, instead of working on social change from the perspective of an inquiring system -- which I'll portray as "how we know" -- should we instead by focused on social change from the perspective of social networks -- as "who we know"?

Within this frame, do you think you have some ideas that are worth discussing online, and/or worth structuring into a paper? Click on the "reply" button to express yourself. Opinions, clarifications, and expressions of outrage are welcomed.

The notion of inquiring systems is a very interesting point of connection between more traditional systems work and issues in business and industry. While West Churchman's work is fairly analytical (i.e., ways of defining and describing systems of interest), most in the systems community will tell you that he was quite an advocate for meaningful human change -- not just theorizing.

What I personally see as the most relevant topic for discussion, in this realm, is the growing understanding that knowledge is collective, not individual. One of the articles by Malcolm Gladwell which David Ing linked (The Televisionary) introduces this nicely. But an even more pointed article by the same author, found under the same link, is The Talent Myth. This topic then opens a host of other issues around participation, communication & understanding, organizational knowledge, motivation, meaning and work, etc.

Gary Metcalf

I guess my challenge is that I'm a big inquiring systems fan, but I'm not convinced that they really work.

(1) I observe that the ISSS doesn't really use them -- so far, I observe primarily inductive-consensual behaviour, with some analytic-deductive when inductive-consensual isn't working. If the ISSS doesn't use inquiring systems, should we expect anyone else to?

(2) Churchman wrote his books before the PC revolution and Internet, and it's difficult to have online debates without people misunderstanding each other. (For the record, I have met Gary in person, and like him. We may debate, but I expect after this, I'll still like him!) There hasn't been a revival of Horst Rittel IBIS (Issues-Based Information Systems) structures. In fact, communications are becoming much more informal, e.g. instant messaging, with emoticons. (My wife and I are berating our children for using those dreadful Internet contractions, rather than full words, but we know that those protocols are expected in his age group).

If we find that inquiring systems don't work in the real world, then we should do everyone a favour, and say so. If, on the contrary, we do believe that they should work, then we should practice what we preach, and use them more often -- particularly within the ISSS community.

I think there are a couple of ways to approach this -- but I think the most important question is whether we're asking the right questions.

Churchman proposes his typology of inquiring systems based on the concepts of the five philosophers listed in the table (Courtney, et al article). One approach is to use that typology as a way to structure processes of learning within an organization (or to identify how an organization functions in regards to inquiry or learning.) Based on that approach, the desired results might or might not be achieved. For instance, if it's decided that a Singerian inquiry system is the broadest and most encompassing, or the most ideal for whatever reason, then Churchman's description could be used to guide how inquiry happens within an organization.

A different approach is to step back and ask, first, whether this typology truly represents what we understand to be occurring within organizations. It would seem that there are people within most organizations who are involved in learning. Some new knowledge becomes institutional knowledge (retained beyond the level of individuals) within many organizations. But just how closely those processes, in any given organization, fit Churchman's typology, I'm not certain. So I'm personally not certain whether we should focus work on applying the model to improving organizations, or work further on the model itself. (I do question very deeply the assumption that learning directly and necessarily results in new and better action -- a problem I think that gets lost in most of this strain of work.)

At the risk of over-generalizing, I observe that North American businesses are most comfortable with inductive-consensual (way #1, Locke) and analytic-deductive (way #2, Leibniz) approaches to reaching decisions. I agree that even if this "right way" is known, that there's still the issue of whether the social group will follow it.

My concern is that the other ways of knowing -- particularly Hegelian debate -- are often not understood for the benefit that it can create. Conflict can be a good thing, in the right situation. And conflict in pursuit of deeper understanding can be separating thing from the action moving forward. (Have you ever disagreed with a group decision, but followed it without resistance anyway, once the group has declared a decision).

You're right about whether we should view the inquiring systems approach as descriptive versus prescriptive. As a practitioner, I like prescriptive frameworks, because they at least point us in some direction (right or wrong).

One of the concerns that I have about viewing social networks by themselves, as opposed to with an inquiring system, is the spirit of direction emerging.

I could argue that extreme believers of "free markets" would argue that whatever direction most people take (i.e. whatever catches on in the social network) must be right. Obviously, this isn't true, as we values outside of free markets. (In Jane Jacobs "Systems of Survival", it's the difference between commercial syndrome and guardian syndrome).

I like the Singerian guarantor of "progress", because it suggests that we're moving towards some particular direction. On the other side, I'm worried about too much rationalism, as a group of people "wants" to move in a particular direction, but never seems to get there.

Two of the articles by Gladwell, The Coolhunt and The Tipping Point, provide very nice "surface level" examples of concepts that I (personally) find to be of extreme importance to current systems work. Once we move beyond the Mechanical Science of reducing human behavior to current mathematical models of some sort, we're generally left with metaphorical models. But that's where most discovery begins.

The notion of large-scale change as an epidemic may be very useful. This same general concept has been explored as "memes" (elements of social structure, equivalent to "genes" in biology) in work by Richard Dawkins, and later expanded by Susan Blackmore. The same general concept was applied to marketing by Seth Godin as the Idea Virus. Most of this work has remained at the "popular" level, but does seem to have at least intuitive value.

Another, but separate, field of work attempting to make sense of large-scale human behavior is in economics. The value here is that it more easily lends itself to numbers and formulas, which brings easier credibility in some arenas.

Focusing discussion at this level will be more difficult for a group, but has the potential for groundbreaking thought and research (in my opinion).

Gary Metcalf

I've run into issues of using systems metaphors versus systems models. The advantage of a metaphor is that it usually has a "good story", so people "get it". The disadvantage of a metaphor is that people start dreaming up properties that don't really exist, and thus I spend time undoing the damage that I created by presenting the metaphor. I find that models work better people who can tolerate them.

In the reading I've done about memes, I've never really liked them. Maybe it has to do with the vagueness, and metaphor-like way they've been presented.

The social network reading that I've been doing is much more model-like. I think that social networks may have become ready for prime time, because there's recently been extensive coverage in Strategy+Business, 4Q2002 -- and this is the publication from the consultants, Booz Allen Hamilton!

There's an extensive article on Karen Stephenson's work (actually the best summary I've seen!) at http://www.strategy-business.com/press/article/?art=9056282&pg=0 (free registration required). Even more intriguing is that fact that Michael Schrage (from the MIT Media Lab, and a former Fortune columnist) writes (or was permitted to write) a review of "Best Business Books in 2002" on Networks! This is available at http://www.strategy-business.com/press/article/?art=9966476&pg=0 .

I like to think about contagion in social settings, much like infections: you're either sick (and maybe a carrier), or else you're not. When it comes to social change, you're either in the "new world", or part of the "old world".

I'm not sure that the same things have been written about memes. I may have missed something.

I don't disagree at all with how vague and ill-defined the concept of memes is. Richard Dawkins proposed it as side-line notion in his book about biological evolution -- and he's so hard-line about "genes control everything" that he's questioned even in that realm, I think. Susan Blackmore latched onto the popularity, but did nothing to further the concepts. I have to wonder, though, in light of The Tipping Point, why this notion caught on so furiously? Maybe that it holds some intuitive value, even without better specificity?

Which brings us to metaphors. My current thoughts about models and metaphors comes generally from work by Robert Rosen, to which I was introduced through the 1999 ISSS meeting. (He was a mathematician and biologist who died just prior to the conference -- otherwise would have been there.) As I understand his description, models, as developed through scientific work, ideally map one to one with thing in the world being modeled or described. If something can be described mathematically, we assume greater validity of the description. So, ideally, we use accurately measured variables of the world to build the models in our minds through which we see and understand the world. (This generally assumes, of course, that the variables can be isolated and measured, which means disconnecting the variables from all other phenomena above, below, and around them -- but that's another issue.)

The fact is, though, that we as humans work from within incomplete and not fully accurate mental models continually. Metaphors can be seen as just "trial models", using what we know as a template for understanding what we don't. Instead of starting from an accurate reality in the world to build our mental models, we usually start from rather vague maps in our heads and use them to navigate the best we can -- and hopefully to learn more.

Please understand, this a different use of the idea of metaphors from the way it typically gets abused in pop-business literature -- for which I think there is no good excuse. Most of that is as useful as Forest Gump's mother's analogy that "life is like a box of chocolates..."

What this leads to is consideration of a more Pragmatist approach to modeling social systems, philosophically, aimed at learning competence and mastery of performance more than absolute accuracy of modeling. (I don't think that one approach rules out the other, though.)

I was also at the 1999 talks around Robert Rosen. I agree that for the average person working his or her way around the world, an incomplete or poor metaphor (in the absence of a full model) may be better than no understanding at all. I think that the risk comes in with arrogance (one of David Hawk's favourite research areas), because a person needs to be sufficiently humble so as to abandon a bad metaphor when it's not working.

This thread appears to be related to discussion on dichotomies versus messes. Modelling a mess can be a huge task. A simple metaphor may be simpler to understand, making it easier to "sell" an idea, even if the full idea isn't really correct.

I need to look at the Blackmore work -- your posting doesn't enforce any more optimism in me -- but memes strike me much more as metaphorical, and potentially useful in a community that is biologically oriented. Taking that metaphor to a group of sociologists, however, is likely to run into some resistance.

Apart from social networks, I'm also interested in Pierre Bourdieu's work on social reproduction of practice. Where this differs a bit from Karen Stephenson's work is that she's much more focused on communication (i.e. who talks to who) rather than how much individuals converge in practice (i.e. one person does it because other people do it).

This isn't because Karen Stephenson hasn't read Bourdieu. (I taught in some classes with her, so we got to discuss foundations of cultural anthropology). More practically, however, it's easier to measure how much one person communicates to another than how alike they are in their practices. (In Karen's consulting work, all you have to do is trace phone calls, conversations and e-mails).

I guess that I'm implicitly taking Bourdieu's idea of practice, and applying a social network view on top of it. This probably isn't an earth-shaking combination by sociologists who have read both sets of literature, but it's a slightly different bent on the world.