Water shapes its flow by the ground

Reading: The way of an army is like water. Water avoids the heights and runs to the low ground; an army avoids the enemy’s strength and strikes at his weakness. Water shapes its flow by the ground; an army shapes its victory by the enemy. So an army has no constant formation, as water has no constant shape; one who can win by changing with the enemy may be called divine.

This passage is often quoted in defence of “analysing each concrete problem on its own terms”, and it does have something in common with the real options of Chapter 2 and the tracking model of Chapter 4: in a moving environment, a fixed practice is bound to fall behind.

But one fundamental difference must be noted: Sunzi is discussing strategy against an enemy. The first chapter of the Art of War, “Laying Plans”, says that “warfare is the way of deception” and “when able, appear unable; when active, appear inactive” (孙武 撰，杨丙安 校理 1999). In an adversarial game, being unpredictable to your opponent is an advantage; in a cooperative game, being unpredictable to your partner is a cost. The decoding model of Chapter 1 shows exactly this: a partner who cannot read your changes revises you towards the opportunist type. Carrying “no constant formation” unchanged over to a partner is treating the partner as an opponent.

Other passages of Sunzi bear more on cooperation, because they concern delegation within one’s own side:

Reading: He wins whose general is capable and not interfered with by the sovereign. There are roads not to be taken, armies not to be attacked, cities not to be besieged, ground not to be contested, and commands of the sovereign not to be obeyed.

The idea fits the delay model of Chapter 4: conditions at the front change fast, and if every decision must wait for a distant sovereign’s approval, the delay from signal to action is too long. Delegating to the front is how to shorten the delay. In modern terms it hands part of the residual control rights to the party closest to the information (Chapter 3). Yet such delegation presupposes that the general is “capable”: that the sovereign trusts his ability and loyalty. In a cross-border venture this brings us back to incentives and trust (Chapter 6).

When a course is exhausted, it changes

Reading: When the Changes reach an end, they transform; transforming, they get through; getting through, they endure. Hence “Heaven helps him; good fortune, nothing that does not further.” The firm and the yielding establish the foundation; change and continuity follow the times.

The often-quoted line “when exhausted, change; changing, get through; getting through, endure” is preceded in the original text of the “Appended Statements” by the word yi, “the Changes”; it speaks of the way of change itself. More noteworthy here is another sentence of the same text: “The firm and the yielding establish the foundation; change and continuity follow the times.” It distinguishes two levels: some things establish the foundation and do not change with circumstances; others follow the times and change with them.

This has a structural likeness to the hybrid contract of Chapter 3: what does not change are the rules and principles (the foundation); what changes are the particular adjustments under those rules (following the times). A good state-contingent contract writes down “when and how we will change” as a constant agreement. Change then becomes the keeping of the agreement, not a departure from it.

“Words must be kept” and “only where righteousness lies”

Reading: Zigong asked, “What must one be like to be called a shi (a gentleman-scholar)?” The Master said, “One who has a sense of shame in his conduct, and who, sent on missions in all directions, does not disgrace his lord’s commission, may be called a shi.” “May I ask about the next rank?” “One whose clan praises him as filial and whose village praises him as respectful to his elders.” “And the next?” “One whose word must be kept and whose deed must be carried through: a stubborn, petty man! Still, he may be counted the next rank after that.”

This passage is often simplified into Confucius disparaging “keeping one’s word and carrying deeds through”. The original is subtler: Confucius ranks it third among the grades of shi and calls it “stubborn” (kengkeng, rigid and narrow), yet still counts it as a shi. It is not the highest state, but neither is it a quality to discard.

Reading: Mencius said: “The great man need not keep every word he speaks, nor carry through every deed he begins; he goes only where righteousness lies.”

Mencius’ words are often used to defend flexibility: so long as it accords with yi (righteousness), what was said earlier need not stand. But the sentence presupposes “only where righteousness lies”: a change must come from a just reason and not from private gain. The trouble is that what one side thinks righteous the other may not; in the model of Chapter 1, a partner who cannot see your reasons cannot tell whether your change goes “where righteousness lies” or where profit lies.

Another saying in the Analects perhaps answers this tension most directly:

Reading: Master You said: “When one’s promises are close to what is right, one’s words can be made good.” The Master said: “A man who does not keep faith: I do not know how he can get on. A large cart without its yoke-pin, a small cart without its collar-pin: how can they go?”

One reading of “when one’s promises are close to what is right, one’s words can be made good” is that a promise should accord with what is right at the moment it is made, so that it can be kept. Rather than make an absolute promise that may later have to be broken, and then break it on the ground of “only where righteousness lies”, it is better to make, from the start, a promise that contains its justified exceptions. In this book’s terms: rather than promise “we will invest for three years” and leave in month three because the market is poor, promise on day one “we will invest for a 90-day exploration period; if such-and-such a metric stays below such-and-such a threshold for two consecutive months, we will leave in such-and-such a way and compensate you for such-and-such investments”. The second promise is far easier to make good.

Covenant and changed circumstances

The same tension exists in the Western tradition. In the Hebrew Bible, a covenant (berit) is a binding relationship established by oath; scholars have compared its form with ancient Near Eastern suzerainty treaties (Mendenhall 1954). Hobbes wrote in Leviathan: “Covenants, without the Sword, are but Words, and of no strength to secure a man at all.” (Hobbes 1651) The principle pacta sunt servanda, “agreements must be kept”, which took shape in medieval canon law and early-modern natural-law thought, is written today in Article 26 of the 1969 Vienna Convention on the Law of Treaties: “Every treaty in force is binding upon the parties to it and must be performed by them in good faith.” (United Nations 1969)

Yet Article 62 of the same Convention also recognises a “fundamental change of circumstances” (rebus sic stantibus) as a ground for terminating or withdrawing from a treaty under very narrow conditions (United Nations 1969). English contract law has the doctrine of frustration, traced to Taylor v Caldwell in 1863: the music hall the parties had agreed to use burned down before the performances, and the court held both parties discharged (Court of Queen’s Bench 1863). The UNIDROIT Principles of International Commercial Contracts provide for hardship: when events fundamentally alter the equilibrium of the contract, the disadvantaged party is entitled to request renegotiation (UNIDROIT 2016).

The limits of this reading

First, the Art of War is about war, the Analects and Mencius about the virtues of the gentleman, and the Book of Changes about the way of heaven and human affairs; none of them is a text on commercial contracts. This chapter borrows their ideas, and does not claim that they contain modern contract theory. Second, “the Chinese tradition values adaptation, the Western tradition values keeping agreements” is too neat a contrast: the Chinese tradition also has “a man who does not keep faith: I do not know how he can get on”, and the Western tradition also has changed circumstances and frustration. Third, the interpretation of these classical texts has a long history of debate; this chapter follows common standard readings, and readers may consult the commentaries to form their own view.