In my first post on strategic concepts, I started with what I considered the most basic questions: preparation, equivalence, efficiency, anticipation, dominance, and the difference between staving off a loss and clinching a victory. Maybe it was a mistake to start with the most abstract concepts. They are probably the most difficult to appreciate. But I tackled this topic to be able to reference concepts that shape the way I think, so it was natural to attack logically-prior concepts I might need to invoke in later posts first.
The strategic concepts I am calling “Machiavellian” are simpler to explain, and their value easier to intuit. In part this is because you, the reader, are probably familiar with the historical context which inspired the questions Machiavelli sought to answer. Furthermore, their premier domain of application is military strategy, which makes it easier to imagine examples and test-cases that illustrate the concepts.
Machiavelli was active as a politician, diplomat, writer and sometime-conspirator at the turn of the sixteenth century, the tail end of Italy’s glittering quattrocento. In Machiavelli’s lifetime feudalism was in its death throes; he lived and politicked in a world still dominated by polities, institutions, and practices which had been perfected at the apex of feudalism. Machiavelli understood that these institutions were relics and understood that they were being replaced, and why. He hoped to use these insights for the glory of the Florentine Republic, or perhaps for the glory of Italia as a whole. These ambitions largely came to end with the fall of Piero Soderini, but continued to dominate his thoughts during his premature retirement.
Roughly speaking, Machiavelli’s great insight was into the nature of the modern army that was emerging out of the ramshackle chaos of failing feudal states. Feudal wars were structurally very similar to the conflicts of primitive tribes. Each duchy or commune had traditional ties of fealty to some political units and suzerainty over others; its leader or leaders had ties of blood to the leaders of yet a third set of polities, which could be a source of cooperation but also of conflict (think: dynastic succession). Their regimes were transient. A duke’s own vassals or a consul’s own colleagues had interests opposed to his, and could rise up against him in support of an invading army. Wars were fought not between states, nor even coalitions cobbled together out of a patchwork of states, but (often enough) coalitions of factions and parties which formed within (and across) various notionally sovereign regimes.
This ambiguity of organization and loyalty was repeated in the structure of the armies such coalitions assembled. Each powerbroker would summon his feudal retainers (if any), call on his kinsmen, appeal to volunteers, beg sympathetic foreign powers for spare troops, hire mercenaries, and present this motley crew to his partners in the coalition as it prepared for battle. Obviously, feudal armies faced the same problem as feudal states; their constituent parts had different interests and would, under stress, become insubordinate. Or switch sides — or rebel!
In any strategic conflict between two teams of people, if the team is relying on some of its members to act in a certain way (because that was agreed beforehand, because they were ordered to do so, whatever) the team must have full confidence in those members. This is the fundamental problem of collective strategy; solving it is prior to solving every other strategic problem. Even an objectively weak organization, if it is truly united and if its officers can have absolute faith that their orders will be executed, can accomplish surprising things. An organization whose members are collectively powerful (in theory) but which cannot coordinate its actions is a disaster which unfolds in slow motion.
(The powerful-but-disorganized group can also do extremely surprising things: but not pleasant surprises.)
There is a garbage-in-garbage-out problem in organizational strategy. If you cannot predict that members of the organization will follow orders, you cannot predict anything else about it. It could perform beautifully; it could fall into bloody strife without ever having been challenged by a competitor. There’s no way to know.
And if an outside observer cannot predict how well an insubordinate organization will do, an inside observer cannot predict what it will do, which is far worse. Strategy requires the preparation of whole trees of plans for certain contingencies — and for contingencies-within-contingencies. The first principle of strategic interaction is that you cannot know what your enemy will do, but you can know what you will do, and in particular what you will do if your enemy does X, Y, or Z. From here, you work your way backwards from winning outcomes to present choices: that is strategic thinking.
But if “you” refers not to your side but to you, an agent in an organization, and you can only know how you communicate with other agents in your organization but not how they will respond to what you tell them, your position is hopeless. You cannot prepare plans for how your team will react to a threat: you can prepare plans for what orders you will give, of course, but that doesn’t tell you what your subordinates will do. And if you have no idea what your team’s response might be, you certainly cannot anticipate how your enemy will react to it!
So before trying to win a competition with outsiders, establish your strategic dominance over insiders. (That is to say, make sure someone else on your team has; not everyone can be #1.) Once you have prepared a plan for translating your orders into action, and you can accurately anticipate your own teammates’ responses to events, then you can start trying to anticipate how other teams will respond to your team.
Note that while the overriding importance of the concept of subordination is to institutional design, the logic applies to all modular strategic analysis, even where there are no principal-agent problems per se. For example, if your strategic preparations make use of a certain machine, or a certain technique, or predictions based on a certain calculation, all of your strategic analysis is for naught if your original predictions were wrong. Bungled calculations, incompetent technique, poorly-maintained machines: you can get loyal followers, you can even carry out the plan yourself, but the net result will be the same if your confidence in each “module” was misplaced.
Part of a series on Strategic Concepts:
- Basic Strategic Concepts
- Machiavellian Strategic Concepts I: Subordination < You are here
- Machiavellian Strategic Concepts II: Centrality
- Machiavellian Strategic Concepts III: Drilling & Disengagement
- Machiavellian Strategic Concepts IV: Hierarchy