April 14, 2004

We need a "Web" not "chain" of command

"You ARE the weakest link: Good-bye."

   If only organizations were that simple. Some middle link in the chain of command is impeding (my) progress. I say the words and poof the chain is mended -- and we're all better off for it. It happens: just not for me. The real problem is the organizational model's "chain" linearity. The structure condemns the organization to be as dumb as it can be rather than as smart as it could be.
Let's consider this shortcoming of the modern, industrial organization structure within the context of non-linearity and network theory, specifically in the 21st-century. The industrial form, which is essential for achieving commercial goals where a core need is consistent, repetitive, time-critical processing of masses of not-especially-unique items has its place: just not dealing with knowledge.

A little background is in order. I work for a VERY industrial organization. Its firm class has been around since the beginning of the industrial age. Today it is not especially different than the industrial-military organizational model that created and perpetuated it through the 19th and 20th centuries. It is prototypical of the industrial form. This despite the fact that it has been forced to create pockets of knowledge employees to align it with the demands of this so-called "new" economy.
The effect of linearity on management is that people are "chained" in a relatively rigid line-of-command hierarchy. Each person in the vertical chain represents a "link" between the mind of the organization (the senior decision-maker) and the rank-and-file. This is not new. Nor is it a novel observation that people are not necessarily intelligent and capable in descending order of ability from top to bottom of the chain. We hope that the best and the brightest are moved upward so that intelligence and competence prevails. But we regularly see people promoted, as Peter has it, to their level of incompetence. Even merit promotions are made for subjective reasons including relationship to the organization and supervisors, (perception of) specific achievement, compliance with organizational norms, political artistry, eloquent incompetence, and sometimes just plain luck. So, chances are that there are some fairly dumb – which is not the same as unskilled – people in the middle strata of an organization.
I've observed in enterprise organizations that the hierarchy readily substitutes power position for intelligence. Again, not new. Particularly regarding the movement of ideas, each link is given a chance to influence communication going up the chain through him/her. That right appears to be jealously guarded more or less in directly inverse correlation to the overall intellectual capability of the individual. These links recast ideas being moved through them in accordance with their own perception of what those above are able to understand. For instance, a junior vice president of lesser abstract intelligence will repackage (if (s)he forewent killing it altogether) an idea from a subordinate into something that "the big guy will understand" based on his/her own ability to understand the idea. In other words, a middle link will – typically, and very humanly – project him/herself onto those above and below. Consistent with the mathematical principle of reduction to the lowest common denominator, in linear chain-based hierarchies the organization (or at least that particular chain) is thereby as dumb as its dumbest link.
Even where links on either side operate at a higher-level, the dumb link will eliminate abstraction – losing much in the process. The dumb link is uncomfortable representing that which (s)he doesn’t understand – and won’t allow it to happen. Thus only ideas comfortable for the dumbest link are transmitted. That's bad despite accepted wisdom that good ideas are simple and understandable by the masses. It's bad because, unfortunately, before an idea can become simple it has to be complex. Gravity and motion are today relatively simple concepts. But the original idea and its proof were, at the time of initial exploration, very complicated. By making ideas dumb before they're ready to be over-simplified does an injustice to the organization and everyone in it – and to those who might have been served by the idea. On the other hand, dumb links ensure that the organization is always well within its comfort zone and skill set for higher probability of achievement.
Now consider the organization as a network. The network has much different characteristics than does the chain: It is non-linear. The "links" are "nodes" which can be limitlessly connected to other nodes. The obvious example is the Internet. The beauty of the network structure is that although some nodes are more important than others – based on their value to the network as determined by the network – no single node inhibits the workings of the overall system. The failure of a "middle" node (though there is no real "middle" in a distributed network) is imperceptible because of the multitude of alternate routes between nodes.
The ideal for a knowledge environment then, might be to have a network organization that connects the most valuable nodes dynamically. Just like with the Internet, the nodes that provide the greatest value (i.e., knowledge, ideas, decision-making, etc.) will be linked directly, irrespective of structural hierarchy. It is a meritocracy on the dimensions of intellect and actionability. This ideal is relatively democratic and the number of links to, would determine the value of the node. Those nodes not linked so much – i.e., those not intellectually up to par or unable to make a value decision – would be systematically neutralized without prejudice. So, at least in respect of communicating ideas, a network-based organization is at least as smart as its smartest node.
This Utopian notion is, of course, unrealistic because an organization is more than its IQ – even a post-modern knowledge organization. Getting things done, for one thing, is an area where intellect and ability do not always correlate (except in the inverse, usually). Also, I've presumed that the weakness is with a middle link of the chain. The highest level positions are, for the sake of this exposition, intellectually advanced abstract thinkers. Many are not. Now what? Well, I'll take the optimistic and charitable view that those at the very highest ranks of an organization do not last there unless they have a few things going for them, one of which is the ability know when they don’t know something important, and demand either to be made to understand or trust confidantes who do.
Still, a "network" organization could and should exist side-by-side with the formal authority "chain" hierarchy, each exerting influence and power as need warrants. That’s a place I’d work at.

Posted by Grayson at April 14, 2004 09:00 PM

Hmm very interesting - a flexible and adaptive organization with the connections needed to allow for ideas, decisions and activity when and where it is needed. Sounds like the type of organization that would be able to embrace and move disruptive markets and technologies. Although isn't this like giving the work to the person who is busy cause they will get it done - even though others are not otherwise occupied?

Posted by: entropy at April 21, 2004 04:59 PM

"Still, a "network" organization could and should exist side-by-side with the formal authority "chain" hierarchy, each exerting influence and power as need warrants. That?s a place I'd work at."

Er, Grayson...This is what escalation is for in an organization. That is, we expect that if a link in the chain is broken, you route to the next higher link (i.e. you have your manager talk to the other guy's manager).

Chain of Command shouldn't be thought of as a linear node-system. Rather it is a "Prefered Routing" system. That is, I normally expect my information to go one path. When that breaks, I have secondary and tertiary nodes to try.

Many companies do try an Internet-style system, and it can cause some problems. You basically have information overload, and dropped balls because you cannot predict the path that information will take getting from point a to point b.

Posted by: Henry Bramlet at April 21, 2004 06:56 PM

Entropy: maybe the busy people will get more work despite others not being otherwise busy. That makes perfect sense. Ultimately that person's efficiency and capability will be recognized (as will the opposite in his/her less capable colleagues) and the rational organization will do more of what it's doing right and less of what it's doing wrong. Perhaps that "node" will be given more resources to do more. If the organization acts irrationally (huh? here? no!), that person will be overworked and underappreciated while those others carry on status quo. The situation can not hold: that star will turn into a black hole either by leaving or burning out. Either way, everyone loses.

Mr. Bramlet: You make a good point. No argument that the linear form (Sorry. From my vantage point, while it may be a "preferred routing system," the industrial form is nonetheless linear.) has its strong points, one of which is structure that provides information filtering. That creates a smaller chance of information overload. Whether balls are inhibited from dropping because of this structure is, however, debatable. But you've suggested that "escalation" is the solution to the problem I raised in the post: have your manager talk to the other guy's manager. Here's my question to you: What if the problem is your manager coming between you and his boss, who needs the information to make the decision? Who would you go to then?

In your last paragraph you describe an Internet-style system typified by unpredictable routing of information among nodes from "a" to "b". I guess it wasn't clear that I'm not suggesting a "packeting system" that bounces information all over the place until it ultimately finds the desired destination. That, IMO, you are right about: it is untenable. My point, rather, was that the connections would be direct rather than intermediated by superiors or subordinates.

My issue is that the linear structure, for all its benefits and comfort in the industrial structure, inhibits by filtration the communication of ideas and knowledge. The work-around (those secondary and tertiary routes) is therefore politics. Taking my example of the boss being the problem, you couldn't go around your boss to his boss without incurring his wrath. Thus, you have to seed your message through "tertiary" routes and hope that the message gets through OR spend an excess of time lobbying your boss to represent your position.

Posted by: Grayson at April 21, 2004 10:27 PM