The Army Capstone Concept: the Army wants your comments

[marct]

Hi Rob,

Constraints could be to the benefit (looking for a specific tolerance) or the detriment of the desired outcome (wrong tolerance and/or out of tolerance relative to the needs). The ability to affect these constraints (policies, biases, etc.) could be within a service's or organization's ability while others (in this case particularly those that are legislated and politically sensitive) may be beyond its ability to affect. They could also be hard to root out, but I think begins with developing an understanding of how the system really operates relative to how we desire it to operate, and what the effects are of our constraints. Then leadership can take better steps in modifying the system - hopefully adding some kind of measure to see if its now producing within the desired tolerances, and if in fact the new tolerance fits the actual requirement.

I'm sure that sounds mechanistic, but it is a "production system", and the bigger the system (be it PME, recruiting and assessment, universities, training programs,etc.), the more the output is going to look like a "product".

For a while now, I've been fascinated by how we, as a species, reify (aka "turn into as 'thing'") our interpersonal interactions and ascribe characteristics to that reification. What truly fascinates me is not that we do this but, rather, that as part of the process of doing it, we have to forget that we have done it. I would really love to see an MRI-based study on the neurological effects of becoming part of an organizational culture, since I suspect that there are actually reflections of that process in brain re-wiring.

That, BTW, is not a tangent - it is directly related to your comment about this sounding 'mechanistic". What I am getting at with it, is that whatever socio-technical systems (STSs) we create, they will have physical (and neurological) effects on the people who are part of them that will condition the probabilities of those systems being able to "produce" people capable of performing certain types of tasks. In effect, these STSs act as additional boundary criteria for natural selection.

The smaller the system (e.g. down at a company or BN level), the more control you have over the output, and the more it can be tailored to specific conditions - even to do as a mentor of mine recommends "putting round pegs in round holes".

Yup, and one of the reasons is actually tied in to the "span of control" problem, which is a sub-set of communications restrictions. Think about it this way: if I know the men under my command from having lived and worked with them, I have a much more nuanced understanding of them that if I only have their service records.

But, in any organization over a certain size (~1-2k), I have to use an exceedingly sparse and formalized system (service records) in order top "know" them. This "knowledge storage/transmission system", in turn, is designed to highlight minimum factors that, earlier in its life cycle, were deemed to be "important" (there's always a time lag, sometimes decades or even centuries long in this). Now this type of system is designed to abstract perceived "core information" about a person and use a standardized form to do so; it is a "mechanistic" system in the sense of interchangeable parts, and it really cannot guarantee "production" of anything more than satisficing behaviour (i.e. minimal standards; they haven't failed - yet).

But if we are talking about producing highly adaptive perceptions in people and rapid problem solving in messy problem space, that is not amenable to a system designed to produce satisficing behaviour in a "clean" problem space. If we want to do this, then we have to use a system that is differently designed, and that is where I was tagging into the strategies discussion of using "mentors". Now, a mentorship model is based on an apprenticeship system that derives out of para-kinship systems. It is much more flexible, generally has much higher minimal standards and, at the same time, tends to be much more labour intensive simply because a lot of time has to be spent working with an individual rather than an amorphous group. BTW, this ties back in to that span of control and communications problem thingy.

The question is how good do you want that product to be relative to the requirement? - again, in this case I think that means defining the tasks you want your people to be capable of executing to reasonable standard at a specific time (position and/or grade) - the product in this case is individual leader capability. If you cannot name and define the tasks then the output (capability) probably does not vary much from where you were - and you are probably limited on your ability to affect the constraints and provide rationale for change.

I'd have to say yes and no. Defining tasks and desired standards is a crucial capability, but that defines what is considered to be satisficing behaviour. In that sense, yes it is crucial, since it is defining the minimum acceptable level of accomplishment in a standardized form. Okay, that works well enough, but it can only capture the predictable, and about half of what the strategy implies it wants to develop is reactions to the unpredictable. See the problem ?

Now there certainly are ways of defining tasks, broadly construed, to capture the, hmmm, let's call it the "predictable unexpected"; that, after all, is one of the reasons for having wargames. But how do we "train" people for this? Standardized training only works with standardized problem spaces so, ideally, we use a melding of standardized training and highly individual education / experience. From some of the stuff I have done, I would estimate that about 80% of the required changes lie in the individual student's head; their perceptions and knowledge of themselves - sort of a "know thyself" on steroids.

In the strategy document, "mentoring" is used as a rhetorical silver bullet to get around this problem, which is why i can only comment in general terms on the document until I start to see some operational details.

Cheers,

Marc