Employees at Mid-Continent Refinery [ca. 1943 Tulsa, OK (LOC). By John Vachon]

Clearing Up My Misunderstanding of Requisite Organization (RO)

E. Forrest Christian requisite organization, Theory 4 Comments

In order, let’s go over what I think are the truths of Requisite Organization. Some of this comes from a result of reading Solaas’s article (see my other posts for a link) and some from fighting through my own questions.

And, yes, I know that TSD is validated and I already have the Craddock bibliography. That wasn’t quite my question but I admit that I’m not sure what I’m looking for. Probably controlled tests. Maybe I’ll design a few and see if anyone’s done them.

In order, my understanding of Elliott Jaques’s theory of Requisite Organization:

  1. RO describes how humans naturally want to order themselves in hierarchies. People want a boss to see out farther than he does but this difference is discontinuous, i.e., your boss’s time horizon has to be at the next quantum state for time horizons above your own for you to recognize him as being your boss. RO is the result of discovering this empirically and describes what appears to be a naturally occuring phenomena in human organizations across cultures.
  2. This is kind of but not quite like saying that my boss can’t just be a little better than me.
  3. Not every grouping of humans is a hierarchy or should be a hierarchy. RO only describes management accountability hierarchies. It does not describe the natural relationships within families, professional societies, or board rooms.
  4. You recognize your boss by intuitively knowing that he can see farther into the future than you can, and that this difference is enough to put him into this next quantum state for time horizons, which RO calls Strata.
  5. The time span of discretion of a job is about how far into the future are you responsible for a result that you are working towards today.
  6. This is problematical for technical expertise. Is the PhD chemical engineer in R&D who doesn’t want to manage anyone (or at least not more than a very small group of other PhD engineers not as advanced as he) supposed to be at the bottom of the management hierarchy? It would depend on the result that he is being held to, how far into the future that result is.
  7. All of this has to be understood in terms of accountability and responsibility.
  8. In IT, my experience (and reading of others’ experiences) has been that no one ultimately owns the long-term results of a system developed in-house for a particular purpose. This is stupid but often true. Which explains why no one knows how to measure anything.
  9. RO must be understood from the idea that humans are goal oriented creatures.
  10. RO is not a goal but a description of how management accountability hierarchies naturally form. So it must describe an emergent function of humans gathering together to accomplish a goal.
  11. Manager-Subordinate relationships that more closely resemble this natural alignment are more trust-full.
  12. Non-natural organizational arrangements (rationalized management layers, per Taylorism) produce distrust.
  13. The person must be able to fulfill the requirements of the role. Otherwise he cannot meet the goals for which he is accountable.
  14. To be adequate for a role, you must have the appropriate knowledge and skills; value the work that the role requires; and have a personal time-horizon that corresponds to the strata of the role’s longest-out goal.
  15. If you don’t value the work being done, it doesn’t matter whether or not you have the time horizon necessary since you won’t do it anyway. If the technical expert doesn’t value managing, don’t give him the job.
  16. These values may be innate, or at least set early in life. Ergo, you probably can’t do anything to change them.
  17. If you don’t have the knowledge and skills to do the work, you can be trained in them. It is probably better, given the choice between someone with the knowledge and skills but not the time horizon and someone without the knowledge but has the time horizon, to pick the latter. You can always teach knowledge and skills but you can’t teach time horizon.
  18. These relationships between boss and subordinate change as each grows.
  19. The boss or the subordinate may outgrow the job; i.e, he is time horizon matches the role above him. This leads to problems: if the boss and the subordinate are suddenly at the same level (sub grows up a stratum), the sub will not recognize the boss as his “true boss”. If the boss grows up a stratum, he may become bored in his job or not communicate adequately with the subordinate.
  20. Jaques later discovered that Complexity of Information Processing (CIP) corresponded with Level of Work stratum, but this is a different matter.

Is that right?

Image Credit: Employees at Mid-Continent Refinery [ca. 1943 Tulsa, OK]. Photograph by John Vachon. Via Library of Congress collection. ]

About the Author

Forrest Christian

Twitter Google+

E. Forrest Christian is a consultant, coach, author, trainer and speaker at The Manasclerk Company who helps managers and experts find insight and solutions to what seem like insolvable problems. Cited for his "unique ability and insight" by his clients, Forrest has worked with people from almost every background, from artists to programmers to executives to global consultants. Forrest lives and works plain view of North Carolina's Mount Baker.  [contact]

Comments 4

  1. Is the ability to gain knowledge and skills tied at all to CIP?

    My guess is that knowledge is a yes and skills is a no.

    I’m particularly interested in the disconnect between skills and CIP. A chess grandmaster doesn’t seem to need much TSoD, yet he’s skilled at making complex decisions.

    Though they may not be linked, is my ability to gain certain skills as genetically determined as my complexity mode? How big a factor is what you value?

    If I have never gotten better at skills requiring me to maintain and evaluate a number of factors in my head and this doesn’t have to do with CIP then how do I get these skills? For that matter, why the heck does what appears to be a debating skill (declarative/cumulative/parrallel arguments)correlate with time span of discretion? How is this different from card players who move from just trying to make their hand, to taking into account what cards are already in play and how that will affect their chances of making their hand, to evaluating the other players plays and how their own play will affect them?

    I’m getting that, “I thought I was close to getting it and now I’m thoroughly lost again” feeling.

  2. To take a quick stab at the last part of your inquiry. There are four “structures” – declarative, cumulative, serial and parallel. But there are also increasingly complex levels of information complexity. Infants start out with “pre-verbal”, then they move on to talking in “specifics”, i.e., this red car in my hand. Most adults function using “classes” of information which is the next level. i.e. red cars (as a class of items) are pulled over more frequently by police. Some adults mature into the ability to grasp and speak in terms of “conceptual abstract” which is a level beyond classes. It deals with using words to represent entire models.

    SIDEBAR: By the way, Jaques was changing his names for the various information levels until the day he died so any text you might read may have slightly different names for the information levels than the ones I’ve used here. Their definitions have pretty much stayed the same, it’s just the names he gave them that changed.

    So as maturation occurs, we progress through the increasingly complex 4 structures (declarative, etc.) within each increasingly complex level of information (preverbal, specifics, classes, abstract conceptual). Think of music. You have 8 notes, an octave, that repeat themselves in higher and higher pitches. We have four structures of processing that repeat themselves at higher and higher levels (more complex) of information.

    So, playing cards or chess may have a complex structure, but it falls within the realm of specifics. For example, if I move this rook (in my hand), then that knight (on the board) may move over here… Specifics is the category preceeding classes in terms of information complexity.

    Your card playing example sounds almost like you are describing someone playing using increasingly complex structures (declarative, cumulative, serial and then paralell processes) (when you start to consider the other players moves and how they impact your moves, that is parallel). However, again, you are using concrete specific data, a deck of cards with a set number of cards and suits, etc.

    I’m pressed for time, but to comment on your ability to acquire knowledge being affected by CIP. I used the example before about professors or ministers who have tons of knowledge regarding their subject, (memorized information and/or experience), but it’s their ability to organize, extrapolate and apply that knowledge that separates them. That ability is CIP.

    I do think the ability to gain knowledge and skills is highly affected by values. Do I care enough to put in the time to learn this?

  3. Words are classes, correct? They’re not the thing they represent? Magritte’s Treachery of Images makes this point with pictures.

    If I understand your octave metaphor correctly, the progression looks like this:

    PreVerbal – Declarative
    PreVerbal – Cumulative
    PreVerbal – Serial
    PreVerbal – Parallel
    Specifics – Declarative
    Specifics – Cumulative
    Specifics – Serial
    Specifics – Parallel
    Classes – Declarative — Strata I
    Classes – Cumulative — Strata II
    Classes – Serial — Strata III
    Classes – Parallel — Strata IV
    Abstract Conceptual – Declarative — Strata V
    Abstract Conceptual – Cumulative — Strata VI
    Abstract Conceptual – Serial — Strata VII
    Abstract Conceptual – Parallel — Strata VIII

    So anyone at Strata I has the requisite complexity (but not necessarily the knowledge, skills, or values) to play chess at Bobby Fisher’s level. I guess that goes along with the idea of not needing TSoD over a day (unless it’s a play-by-mail game). Of course, don’t ask that strata one person to tell you how they’d play at the upper levels: they can’t use words that way. N’est-ce pas?

  4. Again, remember, Time Span of Discretion does not measure the complexity of A TASK. It measures the complexity level of A ROLE which normally includes a basket full of tasks.

    To find the Time Span of Discretion of a role, you look at the LONGEST task associated with the role. Time Span of Discretion gets at one’s ability to decide (discretion) what I will work on today in order to deliver all my tasks on time. Some may have a deadline of tomorrow, while the longest may have a deadline of 5 years. If 5 years is my longest deliverable, it means when I make choices about what I am going to do TODAY, I need to be considering the next five years when I make that decision.

    You may have a 1 month planning TASK called for by a ROLE that has a Time span of discretion of 4 years (stratum 4). If you just looked at the TASK of planning that falls within that role and measured it by Time span of discretion (which again is NOT AN ACCURATE complexity measure for a TASK) and said, OK 1 month, we could get someone at stratum 1 to handle this, you would be disappointed with the planning.

    If the role is at stratum 4, the stratum 5 manager will expect a plan that takes into account all the variables that someone at 4 can see.

    If you hired a consultant to build you a website (one task) and you expected it to take 2 weeks to 1 month, again, if you just looked at the TASK through the Time span of discretion lens, you would construe that you could hire someone with Stratum 1 capability to do it.

    A stratum 1 consultant could very well do the concrete technical work associated with the job, but if you wanted the consultant to help you think through the designing of the site to suit your business from multiple perspectives (business goals, ease of use, sales generation, education, e-commerce, customer service, promotion, products and services, internet traffic generation), you would be disappointed with the work of a stratum 1 web consultant.

    Although the TASK will take under a month, you would need a consultant with stratum 4 capability or more to accomplish that TASK.

Tell Forrest how wrong he is: