We should start by returning to our three levels of military analysis: tactics, operations and strategy. We’ve dealt with tactics (how you fight) and operations (where you fight, and how you get there). Strategy is an often misunderstood term: most “strategy” games (especially real-time strategy) are actually focused almost entirely on tactics and operations; as a rule, if “don’t have a war” isn’t an option, you are not actually doing strategy. Likewise, a lot of basic planning in business is termed “strategy” when it really is tactics; not a question of goals, but of means to achieve those goals. Because strategy is the level of analysis that concerns why we fight – and thus also why we might not fight. Let’s unpack that.

(Attentive readers who know their Clausewitz (drink!) will recognize that I am being both broader and narrower than he in how I use the term strategy. Clausewitz terms strategy as “the employment of battles to gain the end of war” which is more nearly what we today mean as operations. In contrast, strategy as it is used today in a technical sense corresponds more nearly to what Clausewitz terms policy, the third element of his “marvelous trinity”. A full exegesis of Clausewitz’ trinity is beyond the scope of this essay, but I wanted to note the differing usages, because I’m going to quote Clausewitz below. And as always, every time Clausewitz gets quoted you must take a drink; it’s the eternal military history drinking game).

At the strategic level of analysis, the first question is “what are your policy objectives?” (although I should note that grand strategy is sometimes conceived as an analytical level above strategy, in which case policy objectives may go there). There’s a compelling argument common in realist international relations theory that the basic policy of nearly all states is to survive, with the goal of survival then suggesting a policy of maximizing security, which in turn suggests a policy of maximizing the military power of the state (which ironically leads to lower the security of other states who then must further increase their military power, a reaction known as the “security dilemma” or, more colorfully, the “Red Queen effect”). I think it is also possible for states to have policy goals beyond this: ideological projects, good and bad. But survival comes first.

From there, strategy concerns itself with the best way to achieve those policy objectives. Is peace and alliances the best way to achieve security (for a small state, the answer is often “yes”)? Would security be enhanced by, say, gaining a key chunk of territory that could be fortified to forestall invasion? Those, of course, are ends, but strategy also concerns itself with means: how do you acquire that defensible land? Buy it? Take it by force? And then – and only then, finally – do you come to the question of “what sort of war – and what sort of conduct in war – will achieve that objective?”

You may note that this is not the same kind of thinking that animates tactics or operations. Military theorists have noticed that for quite some time, often suggesting a sharp separation between the fellows who do operations and tactics (generals) and those who do strategy (typically kings or politicians). As Clausewitz says (drink!), “The political object is the goal, war is the means of reaching it, and means can never be considered in isolation from their purpose … war should never be thought of as something autonomous, but always as an instrument of policy [emphasis mine].” In short, Clausewitz stresses – and leaders have long ignored to their peril – that of all of the factors in war, policy ought to guide action (although no part of the trinity may be neglected).

This creates subordination between the three levels of analysis (to get technical, this is because operations and tactics are part of a side of the Clausewitzian trinity which ought to be subordinate to policy). Operations is subordinate to strategy; an operation which achieves something that isn’t a strategic goal accomplishes nothing. And tactics is likewise subordinate to operations. Thus the thinking pattern should always proceed from the highest questions of strategy down to the prioritization of ends (still strategy), to the means to accomplish those ends (still strategy); only then to the execution of those means (operations) and then to the on-the-ground details of that execution (tactics). Of course what this tripartite division is mean in part to signal is that all three of these stages are tremendously complex; just because tactics is the subordinate element does not mean it is simple!

Bret Devereaux, “Collections: The Battle of Helm’s Deep, Part VIII: The Mind of Saruman”, A Collection of Unmitigated Pedantry, 2020-06-19.