Template:M intro design protestant and catholic: Difference between revisions

Tags: Mobile edit Mobile web edit
Tags: Mobile edit Mobile web edit
 
(6 intermediate revisions by the same user not shown)
Line 3: Line 3:
{{form and substance capsule}}
{{form and substance capsule}}


These are divergent philosophies when organising a complex system.
In any case these are quite divergent philosophies when managing [[risk]] in a [[complex system]]. Form has but one advantage over substance: ''unit cost''.  You don’t need expensive [[expert]]s, who have invested the time and resources in understanding the territory, to follow a [[playbook]]: a [[school-leaver from Bucharest]], with a suitable command of English and a sullen teenage disinterest in asking precocious questions, will do. Better, in fact, since [[expert]]s ''do'' tend to ask precocious questions, and that really isn’t in the spirit of things in a formal world:
Form has one advantage over substance, and one only: ''unit cost''.  You don’t need expensive experts who have invested the time and resources in understanding the territory: a [[school-leaver from Bucharest]], with a suitable command of English, and a sullen teenage disinterest in asking precocious questions will do. Better, in fact, since [[expert]]s ''docc tend to ask precocious questions, and that really isn’t in the spirit of things.


Here is how things roll in these opposing modes of operation:
====Formal organisation====
====Formal organisation====
So we implement process ''A'', to deal with malign contingency ''X'', but processes being only simplified models — derivatives — of the worlds they represent,<ref>We take it as axiomatic that, the “real world” being analogue, fractal and complex, a process ''cannot'' perfectly map to a target contingency: to believe it might is to mistake [[the map and the territory|a map for the territory]].</ref> process ''A''’s shadow inevitably falls across ''benign'' contingencies ''Y'' and ''Z'': circumstances not needing process A, but which “[[If in doubt, stick it in|it won’t hurt]]” to subject to Process ''A'' anyway.
Top down, the organisation implements process ''A'', to deal with malign contingency ''X''. Process A will be formulated and managed by an ''[[administrator]]'', who need not, and usually won’t, be a [[subject matter expert]].  


(The alternative would be to implement a Process ''A''', drawn wholly ''inside'' the boundary of malign contingency ''X'', and whose shadow therefore didn’t fall across ''any'' benign contingencies, but which also did not quite cover all instances of contingency ''X''. Such a process, which fails to address [[tail risk]]<nowiki/>s, is a ''bad'' process).<ref>It is also the principle upon which almost [[Black-Scholes option pricing model|all modern risk management is based]], but that is another story.</ref>
Now, processes being based upon simplified models of the worlds they represent,<ref>We take it as axiomatic that, the “real world” being analogue, fractal, complex and unbounded, and a valid process being short, digital, algorithmic and complete, a process ''cannot'' perfectly map to a target contingency: to believe it can is to mistake [[the map and the territory|a map for the territory]].</ref> process ''A''’s shadow inevitably will fall across ''benign'' contingencies ''Y'' and ''Z'': circumstances not needing process A, but which [[If in doubt, stick it in|it won’t hurt]]” to subject to Process ''A'' anyway.


We should expect process ''A'' to get in the way every now and then, when a contingency ''Y'' or ''Z'' comes about.  
(The alternative would be to implement a Process ''A''', drawn wholly ''inside'' the boundary of malign contingency ''X'', and whose shadow therefore didn’t fall across ''any'' benign contingencies, but which also did not quite cover all aspects of contingency ''X''. Such a process, which fails to address “[[tail risk]]s”, is a ''bad'' process).<ref>It is also the principle upon which almost [[Black-Scholes option pricing model|all modern risk management is based]], but that is another story.</ref>


The options are:
We should expect process ''A'' to get in the way every now and then, when a contingency ''Y'' or ''Z'' comes about. When it does, the alternatives are to run process ''A'' anyway, even though everyone knows it isn’t needed, or to ''waive'' process ''A'', by invoking process ''B'' (the “''process A waiver''” process).


(i) to run process ''A'' anyway, even though everyone knows it isn’t needed, or
Either alternative has a cost, the first being marginally preferable because it is already costed in. To justify the second one must show that the cost of obtaining the waiver will be less than the cost of just running process ''A,'' and so will result in a saving. This will, in turn, will trigger process ''C'' (the “''justifying the cost of a new initiatives''” process) which will, of course, increase the cost of process ''B'', making process ''C'', and therefore process ''B'' more likely to fail.


(ii) to waive process ''A'', invoking process ''B'' (the “''process A waiver''” process).  
There is another way of doing things, of course: engage [[subject matter expert]]s — which we define as “one who understands the territory and therefore the map’s limitations” — who can apprehend that this is a benign contingency ''Y'' and make the substantive judgment that, Process ''A'' is not substantively relevant, and thereby ignore process ''A''.


Either option has a cost: option (i) being marginally preferable because it is already costed in. Justifying option (ii) involves demonstrating that the cost of obtaining the waiver will be less than the cost of just running process ''A,'' and so will result in a saving. This will trigger process ''C'' (the “''justifying the cost of a new initiatives''” process) which will, of course, increase the cost of process ''B'', making process ''C'', and therefore process ''B'' more likely to fail.
This will upset two categories of people: [[administrator]]s — who we define as “that class of people who are not [[Subject matter expert|subject matter experts]], don’t understand the territory, fetishise the map, and therefore are aggrieved, or fearful, when the map is disregarded”; and [[Rent-seeker|rent-extractors]] — those who stand to be gain by rigid application of the map, some of whom will be [[administrator]]s, but many of whom will be external [[professional advisers]]. (Often the waiver process might require, for example, a [[legal opinion]]: being a certificate from a specially-engaged [[subject matter expert]] that there is nothing to worry about.)


====Substantive organisation====
====Protestant and catholic modes of organisation====
There is another way of doing things, of course: a [[subject matter expert]] — which we define as “one who understands the territory and therefore the map’s limitations” — can apprehend that what she beholds is benign contingency ''Y'' and make the substantive judgment that, while it is formally applicable, Process ''A'' is not substantively relevant, and thereby ignore process ''A''.
This leads the [[Jolly Contrarian|JC]] to offer two modes of operation: the run-of-the-mill ''protestant'' mode by whose principles nearly all significant organisations are run — in which rules are rules, to suffer is divine, and rewards are presumed to be in the next life, since they clearly don’t apply in this one — and the much-talked-about-seldom-seen ''catholic'' mode —  in which people who know what they are doing act immediately and. if need be, ask for permission, or forgiveness, later.  


This will upset two categories of people: [[administrator]]s — which we define as “that class of people who are not [[Subject matter expert|subject matter experts]], don’t understand the territory, fetishise the map, and therefore are aggrieved when the map is disregarded”; and [[Rent-seeker|rent-extractors]] — those who stand to be gain by rigid application of the map, many of whom will be, of course, [[administrator]]s.
That always seemed more fun, open-minded and apt to uncover opportunities, and expose lurking risks, one might not find when following the same, mandated path home every day. The [[map]] may be linear, after all, but the [[territory]] is definitely not.
====The protestant and the catholic modes or organisation====
This leads the [[Jolly Contrarian|JC]] to offer two modes of operation: the run-of-the-mill ''protestant'' mode — being the first one, in which rules are rules, to suffer is divine, and rewards are presumed to be in the next life, since they clearly don’t apply in this one — and the much-talked-about-seldom-seen ''catholic'' mode — being the second one, in which you act now and ask for permission, or forgiveness, later. That always seemed more fun, open-minded and apt to uncover opportunities, and expose lurking risks, one might not find when following the same, mandated path home every day. The [[map]] may be linear, after all, but the [[territory]] is definitely not.


To be catholic is to walk a high wire: there is no room for bluffers, fakers or know-nothings — they may ''ask'' for permission, but few will get it. But nor should you hire bluffers or fakers, and if you have hired no-nothings in your protestant strategy, you will be a long time finding a catholic path to redemption.
To live the catholic life is a risky bet, with no room on the broom for bluffers, fakers or know-nothings — they may ''ask'' for permission, but few will get it. But nor should you want bluffers or fakers on your broom, and if you have hired no-nothings as part of your protestant strategy — which is what it recommends — good luck changing faiths: you will be a long time finding a catholic path to redemption.
 
====Adjacent possible====
The history of innovation, of commercial revolution, is one of inspired connections, lucky hunches and creative routes out of apparent mishap. But to make a serendipitous connection you must put yourself in the way of them, and that means skiing [[off-piste|off the groomed trails]]. The skiing there is harder, requires technical chops, rewards ''care'', but the return is correspondingly greater, the possibility for discovery — if opportunity as well as imminent disaster — more immense.
 
If you “run on rails” you systematically deny yourself that chance of serendipity. All the more so should you work remotely — whether on a different floor, at home or in a call centre in Bucharest—, where there can be no “water cooler moments” — those mythical connections when institutional sparks fly, so beloved of CEOs when imploring work-shy staff back to the office — for remote work, in self-imposed physical exile, communicating only digitally, in pre-arranged multi-party meetings with a standing agenda, is about as railish as one could possibly be (hence the JC’s theory that [[COVID]], and [[bring your own premises]], brought to its logical conclusion/reductio ad absurdum the 40-year deprecation of [[subject matter expert]]ise — from “ineffable genius, deserving of an office, a company car, a personal secretary, an expense account and private club membership” through to “battery hen with a probabilistic entitled to space somewhere along a row of Formica trading desks” to “sort out your own accomodation, here is a network account which you can access via two-factor authentication on your personal device, and we've blocked YouTube, Twitter and Reddit BUT NOT [[LinkedIn|LINKEDIN]]” in the manner of a battery hen —that has been the management modernists’ intellectual project.) There is no chance to bump into anyone, no raised eyebrow at the resting state absurdity of it all, no off-colour wisecracks to bring some humanity to the situation (“AND JUST AS WELL”, intones [[human resources]], tersely) — no opportunity to reimagine the way things are currently done.
 
If we can only evolve by reference to the [[adjacent possible]] — doors to rooms off this particular corridor, with their own doors to potential places elsewhere, and ''one is not permitted to open any of those doors'', then one cannot expect to end up at a place other than the one to which management thinks you should be going. The voyage of discovery never made where management says it would.
 
Of course expecting creative sparks to fly from rocks as damp and browbeaten as most modern legal, compliance, risk and HR personnel is wishful, but the principle is surely right — only opportunities for unscheduled interactions, spontaneous, communication , an ''absence of an agenda'', and unstructured play — designed, or at least  permitted, [[system redundancy]] — gives the organisation any chance of serendipity.
 
Do we even ''want'' operational grunts having bright ideas? An 80/20 analysis says, stick to your knitting, kids. Follow the playbook and almost everything will be alright.
 
If it isn’t simple enough to reliably leave to an ''actual'' rule-following machine — if there are enough unpredictable scenarios, leaving open outcomes ugly enough to require the intervention of an imaginative human mind to stave off calamity, then do you really want that intervening hand to belong to an ingenue [[school leaver in Bucharest]]?
 
Here the modernist programme contradicts itself. Either
 
The modernist agenda, though sees business as fundamentally soluble by data. There may be a dim appreciation that new business requires innovation, but managing the bau surely does not, beyond further refining, digitising and simplifying the process. Risk management is a matter of operational excellence, and operational excellence is a matter of following the rules as quickly and quietly as possible.
 
The JC has said it enough already on this site that it hardly needs being said again, but ''nothing for be further from the truth''. The [[roll of honour]] refers. You don’t die under the wheels of a brand new opportunity. You die when something you thought was boring blows up. People riding on rails will never be able to stop boring things blowing up. They are doomed to be surprised by them.
 
Risk management as protestant, new business as catholic