Tit for Tat

Friday

---

Several times in this forum, and most recently in A flare from Cold War embers, I have mentioned the game-theoretic idea of tit-for-tat, and tit-for-tat with forgiveness. An example of tit-for-tat strategy from Wikipedia shows its simplicity and straight-forward character:

1. Unless provoked, the agent will always cooperate
2. If provoked, the agent will retaliate
3. The agent is quick to forgive
4. The agent must have a good chance of competing against the opponent more than once.

(As an aside I will mention that tit-for-tat developed as a strategy for the prisoner’s dilemma was pioneered by Anatol Rapoport, whose work I have elsewhere made use of in relation to conceptions of history. On this cf. More on Clausewitz, Species of War and Peace, and Three Conceptions of History. In both Rapoport’s analysis of war and his tit-for-tat strategy, the idea of agency is central.)

I derived the tripartite division of conceptions of history from Anatol Rapoport's discussion of Clausewitz.

While it can be shown that tit-for-tat is almost always an optimal game strategy, we know that it fails, and that it fails spectacularly at times. Most often, tit-for-tat fails because it is transformed into a spiral of escalation. In the simplest version of the game, tit for tat with two players: “A one-time, single-bit error in either player’s interpretation of events can lead to an unending ‘death spiral’.” (Wikipedia) The dyad here can be understood as two individuals or as two nation-states, or any other political entity you care to substitute for a nation-state. This is, of course, a very big exception to the rule of optimality, and it is easy to see how such games escalate catastrophically. In a purely theoretical context, we lack the additional information of a real world scenario, and this additional information can transform our understanding of the situation, justifying escalation.

The dyad of conflict can involve any two political entities as agents.

The escalation of the Hatfield-McCoy feud provides many interesting examples of real-world application of tit-for-tat strategy, and how it goes wrong. The first murder in the Hatfield-McCoy feud took place during the Civil War on 07 January 1865, when Asa Harmon McCoy was murdered by a local Confederate death squad after his return from service in the Union army. In the Tug Fork region, service in the Union army was seen as a betrayal. Therefore there was local sympathy for the murder. Also, 13 years elapsed between this murder and the next murder in the feud, which was the murder of Bill Staton, a relative of both families. For this murder, Sam and Paris McCoy were tried and acquitted on the ground of self-defense.

The Hatfield Clan

We can immediately see how the murky circumstances (call it the fog of feud, if you will) surrounding these widely separated murders could lead each side in the feud to interpret matters differently. The first murder was popularly sanctioned. The second murder was later judged to be self-defense, so that the perpetrators walked. Later in the feud, Hatfields deputized as constables arrested some McCoy brothers, and again it is easy to see how even if this action has the force of law, it would not be considered legitimate (by the McCoys) because of the circumstances.

The McCoy Clan

Escalation replaces a successful tit-for-tat strategy when the two agents involved in the game differently interpret matters of fact appropriate to their circumstances. To put it simply, what one side identifies as a “tit” the other side does not so identify, and vice versa. It is important to understand here that both sides understand themselves to be playing the same game — tit-for-tat — but that they disagree over how this abstract strategy is to be interpreted in the context of real world events. This is why, in A flare from Cold War embers, I wrote that, “In philosophical parlance, we have distinct ontologies even if we share an axiology.” That is to say, even if we have values in common, we may not apply these values to the same world.

ROKS Cheonan (PCC-772) before being sunk, likely by a torpedo from a DPRK submarine. The ship has since been recovered from the bottom.

In the case of a tit-for-tat game, we share the principle of tit-for-tat, but we differently identify events as constituting a “tit” or a “tat.” This is why we say that the two agents playing the game have different ontologies. Because each views the world differently, and each identifies objects in the world differently (as being a “tit” or not being a “tit”), each pursues the game on the basis of the same principles but applying those principles to what is, essentially, a different world. In yet another philosophical formulation (this time from Kuhn), we could say that the strategies are commensurable while the worlds to which the strategies are applied are incommensurable. This complexity of overlapping commensurabilities and incommensurabilities is very interesting, but I will not develop this idea at present.

They know how to goosestep in the DPRK.

In the recent clashes between North Korea and South Korea we can see these differing identifications of events in action. In what is sometimes called the Battle of Daecheong, a DPRK patrol boat, despite being warned by ROK Navy vessels, crossed the NLL (Northern Limit Line) on 10 November 2009. An engagement followed in which the DPRK patrol boat was severely damaged and at least one DPRK sailor was killed (and possibly as many as ten). It has been widely circulated that the sinking of the ROKS Cheonan (PCC-772) — also identified as the Chon-An in some media sources — was undertaken in retaliation for DPRK losses in the Battle of Daecheong. Thus, for the DPRK, the Battle of Daecheong was a “tit” that demanded a response. Moreover, since the DPRK does not recognize the legitimacy of the NLL, but claims the disputed islands between the NLL and the coast as its own, the actions of the ROK navy vessels in warning away their patrol boat were illegitimate and unjustifiable.

For South Korea, the sinking of the ROKS Cheonan was an unprovoked attack that itself demands a response. While the South Koreans would be fully aware that the DPRK understands the sinking of the Cheonan as “payback” for the Battle of Daecheong, the length of time that has elapsed between the incidents, as well as the striking disproportion of the incidents, suggests that the two incidents are not related. If the relationship between the two incidents is problematic, it is just as legitimate to interpret it as an independent action as to interpret it as a response. Moreover, for the South Koreans, who recognize the NLL, all of their actions were justifiable and legitimate within their recognized legal framework, whereas the actions of the DPRK were neither justifiable or legitimate.

Once we recognize that divergent ontologies lie at the base of escalation, we can immediately see that we as yet possess no ontology adequate to the purposes of strategic and military thought. Why do I make this claim? Because strategic and tactical thinking requires that we identify objects in the world in a particular way. A mass of men marching toward us, guns raised, is not merely a mass of men, it is an opposing force that means to disable us, if not kill us. We must identify it as such (as “hostiles”) if we are to respond appropriately.

Friendlies or hostiles? We need a strategico-tactical ontology to determine the place of soldiers within the geopolitical world.

A strategico-tactical ontology demands a system of categories specific to strategic and tactical engagement. While this is of course familiar to those on the front lines (whether metaphorically or literally), it has not been assimilated into strategic thought at its most abstract and general level. Strategic thinking requires a strategic ontology, and I do not know of any systematic exposition of a conception of the world based on strategic imperatives, except that ontology that is implicit in classic works of strategy like Clausewitz’s On War. This absence of a formal strategic ontology is a cause of the weakness of strategic thought, but it is also an opportunity to make good the deficit.

. . . . .

. . . . .