Interpreting Sun Tzu: The Art of Failure? by John F. Sullivan

If you now wish to inquire into the Way of [the ancient sages], may I suggest that one can hardly be certain of it? To be certain of it without evidence is foolishness, to appeal to it though unable to be certain of it is fraud.
—Hanfeizi (3rd century BCE)

“Translation,” an American poet and translator of Dante’s Inferno opined, “is the art of failure.”[2]  In Don Quixote, the eponymous character notes that distortion is often a natural byproduct of the effort: “translation from one language into another…is like looking at Flemish tapestries on the wrong side; for though the figures are visible, they are full of threads that make them indistinct, and they do not show with the smoothness and brightness of the right side.”[3] The reverse tapestry is an apt metaphor for reading any ancient Chinese text, particularly The Art of War. While the use of logographs to express complex thoughts has been a constant feature throughout China’s recorded history, the written language of thousands of years ago differs significantly from its modern variant. While the original Art of War consists of approximately 6,ooo characters, a modern Chinese version requires more than double that number to convey the same approximate meaning.[4] Even most native Chinese speakers, therefore, read a translation of the original.

A copy of The Art of War by Sun Tzu. (Wikimedia)
A copy of The Art of War by Sun Tzu. (Wikimedia)

While The Art of War is surprisingly short and compact, much remains ambiguous in its received message. As a result, our contemporary interpretations require constant skepticism, debate, and revision. While Sun Tzu’s text is arguably the oldest within the core strategic canon, it has been studied for the least amount of time by Western military theorists, in comparison with Thucydides and Clausewitz, for example. First translated into English only in the early twentieth century, strategists largely ignored The Art of War until the Vietnam War renewed interest in Asian military thinking.


Despite the limited scholarly focus on the text, in his foreword to the 1963 Griffith translation B.H. Liddell Hart confidently proclaimed that The Art of War “has never been surpassed in comprehensiveness and depth of understanding…Sun Tzu has clearer vision, more profound insight, and eternal freshness.” The certitude, though, with which we purport to understand The Art of War’s “clear vision” and “eternal freshness” remains inversely proportional to the collective effort we have put into researching its historical context or subjecting it to harsh philological analysis and extended debate.[5]  Unlike Thucydides’ work, which has been well-served by the commentarial traditions of A.W. Gomme and Simon Hornblower, nothing remotely similar exists in English for contextualizing this Chinese classic. While translations of Sun Tzu vastly outnumber those of Clausewitz, reliable secondary-source references on the latter theorist and his milieu abound, while those on the former remain conspicuously absent.

Given the scarcity of authoritative writings or clarifying analyses on Sun Tzu’s text, how confident should we be that we have correctly grasped “the Way” of this ancient sage? Of particular importance, one of the core ideas we almost universally believe serves as a bedrock to Sun Tzu’s overall military philosophy—that his ideal strategic objective is “to take the enemy whole and intact”—rests on a problematic and potentially untenable textual foundation.[6] Instead, a stronger case favors an interpretation of Sun Tzu prioritizing self-preservation. Whether or not one’s adversary is destroyed or taken non-violently remains a distant secondary concern.



The idea of “taking the enemy whole and intact” comes from the first verse of the third chapter. Lionel Giles’ 1910 English translation proposed the following rendition:

In the practical art of war, the best thing of all is to take the enemy’s country whole and intact; to shatter and destroy it is not so good. So, too, it is better to recapture an army entire than to destroy it, to capture a regiment, a detachment or a company entire than to destroy them.[7]

Since Giles, almost every subsequent translator of the text produced a similar interpretation. Before comparing with the original Chinese, though, it is helpful to also consider Ralph Sawyer’s version, since his work is more consistent and literal than either the Giles or the Griffith translations:

Preserving the [enemy’s] state capital is best, destroying their state capital is second-best.
Preserving their army is best, destroying their army is second-best.
Preserving their battalions is best, destroying their battalions is second-best.
Preserving their companies is best, destroying their companies is second-best.
Preserving their squads is best, destroying their squads is second-best.[8]

Now looking at its original written form, even without any knowledge of Chinese characters, it is clear that the verse in question is structured as a nearly identical repeating pattern. The only variation in each line is the specific organizational unit being referenced, arranged from largest to smallest entity:


This means we only need to translate a single line to unlock the key for how all remaining lines in the verse should best be interpreted. Below is a rough direct translation of each individual character in the first line:


To further simplify our analysis, we can limit ourselves to only the first two characters within each individual line:


Unlike in English, where a declarative statement generally requires an explicit grammatical subject, in classical Chinese the subject is often left unexpressed, as it is here.[9] What Sun Tzu presents the reader with is simply a predicate, composed of a verb (全) and a noun (國), with the noun serving as the direct object of the verb. It is from these two characters alone that Giles derived his initial translation of: “to take the enemy’s country whole and intact.” But one will notice in its original form there is no additional character or any other indicator clarifying which state is being directed to be preserved—the enemy’s state or one’s own? This is common in classical Chinese, a notoriously terse system of writing, that relies on the overall context to flesh out specific meanings.[10]

Giles’ decision to interpret the noun as clearly referencing the enemy state is a choice, not an unambiguous direct translation. Sawyer’s version elucidates this by placing the term “enemy” in brackets, indicating that although this is not derived directly from the original wording of the text, his own interpretation suggests adding the modifier is a logical conclusion. But from a translation standpoint, arguing that Sun Tzu is singling out one’s own state for preservation is an equally valid option. Why did Giles—and his subsequent imitators—decide the entity directed to be preserved could only be the enemy’s?

In fact, later in his own translation, Giles reconsidered his initial choice. In the final verse of the twelfth chapter, Giles identified a second juxtaposition of the characters “preserve” (全) and “army” (軍): “Hence the enlightened ruler is heedful, and the good general full of caution. This is the way to keep a country at peace and an army intact [全軍].”[11] But here Giles recognized the verse had an unambiguous meaning conveying the need to preserve one’s own army, not the enemy’s. In a footnote to this verse, Giles expressed doubt of his initial interpretation’s correctness: 

It is odd that 全軍 [preserve army] should not have the same meaning here as in [the first verse in chapter three]. This has led me to consider whether it might not be possible to take the earlier passage thus: “to preserve your own army (country, regiment, etc.) intact is better than to destroy the enemy’s.”[12]

Giles, though, never modified his earlier passage, and when Samuel Griffith completed the next major translation of the text half a century later, he followed Giles’ lead: “the best policy is to take a state intact.”[13] In response to Griffith’s translation, the renowned sinologist D.C. Lau, a native Chinese speaker and prolific translator of other ancient Chinese texts, including Sun Bin’s Art of War, critiqued Griffith’s effort. In his 1965 article, “Some Notes on the Sun Tzu,” Lau took aim at Griffith’s rendering of the first verse of the third chapter, noting that the received wisdom that it was the enemy’s entire state and military forces directed to be “taken intact” was based on an incompatible rendering of the ancient Chinese verb “to preserve” (quan 全):

In taking the term quan as referring to the capture of the enemy intact, Griffith is following the Chinese commentators who, practically without exception, all take the term that way.[14] But the word quan normally means “to preserve intact,” and to make it mean “to capture the enemy intact” is to stretch its meaning…there seems, thus, to be a case for arguing that quan refers to “preserving oneself intact.”[15] (emphasis added)

The conventional interpretation, therefore, pushes a reading of “preserve” (全) well beyond its definitional limits by implying that even the enemy’s five-man squads—his lowest level tactical formations—are ultimately meant to be spared from any adverse effects during conflict.[16] Had Sun Tzu meant to convey a meaning of “capturing the enemy’s state and forces intact,” he could have used the much clearer verbs “to take/to annex” (取) or “to capture” (擒).[17] In other places within the text, Sun Tzu uses these same alternate characters to convey these specific meanings when referencing the enemy’s territory and military units.[18]

Moreover, other portions of the text appear to contradict the conventional view that one should prioritize the enemy state’s preservation. Just several verses later, Sun Tzu emphasizes that “one who excels at employing the military…destroys other people’s states without prolonged fighting.”[19] In the eleventh chapter he notes that “if you pursue your own program, and bring your prestige and influence to bear on the enemy, you can take his walled cities and lay waste to his state.”[20] Expunged, it would seem, are any sacrosanct appeals to preserve the enemy state intact. While the text certainly preaches moderation, prudence, and caution in warfare, it never places the adversary’s preservation above one’s own desire to achieve military victory.


These arguments suggest that sufficient textual evidence exists to approach the conventional interpretation skeptically. A better reading is as follows:

Preserving one’s own state (army/battalions/companies/squads) is a commander’s highest duty; destroying the enemy’s state (army/battalions/companies/squads) is a secondary objective.[21]

Through this revision, Sun Tzu may be reminding his adherents that the continued preservation of one’s own state and potential combat power should always be their foremost priority, superseding even a ruler’s unquenchable desire to annihilate his rivals. As the Syracusan general Hermocrates cautioned in a speech documented by Thucydides, “I am not inclined to ruin myself for the sake of hurting my enemies.”[22]

Conversely, under the conventional wisdom, the commander should always prioritize the least violent means available, even if this means limiting one’s own strategic objectives, since the ultimate goal is to preserve the enemy state and its forces after victory as much as possible. Viewed through the revised reading, violent methods always remain a viable, and potentially even preferable option, so long as the commander first feels secure enough in his own strategic position.

A movie poster depicting the Battle of Changping at the end of the Warring States period (Heritage Auctions)
A movie poster depicting the Battle of Changping at the end of the Warring States period (Heritage Auctions)

This updated interpretation, though, would not preclude the use of less direct or even non-violent means to achieve strategic objectives, provided these methods do not unduly imperil one’s own security. Thus, China might decide open military conflict with the United States incurs too much uncertainty and risk, preferring diplomatic, economic, informational, and psychological efforts to achieve its aims. Simultaneously, it might order a violently executed, full-scale military invasion of Taiwan if it determines that seizing the island is vital to the regime’s survival and the relative balance of forces skews in its favor. The key point is that one’s own preservation is the unerring guiding principle, not ephemeral commitments to “preserve” one’s hypothetically subdued adversary.

A revised interpretation of this passage will also assist in making the overall text more logically consistent. Take the contradiction inherent in the dual insubordinations authorized by Sun Tzu in the tenth chapter:

If the situation is one of victory but the sovereign has issued orders not to engage, the general may decide to fight. If the situation is such that he cannot win, but the sovereign has issued orders to engage, he need not do so.[23]

If the highest priority is to take the enemy whole and intact, the latter insubordination is understandable, since purposefully fighting a lost battle will preserve neither the enemy nor oneself. But, by using this same logic, Sun Tzu would have no good justification for advocating the first insubordination. Here, the sovereign simply rejects a method at odds with the supposedly preferred ideal of non-violent submission. There is no rational basis, then, for Sun Tzu’s unwillingness to find a more suitable method to subdue the enemy short of fighting. But if we recognize that Sun Tzu’s highest objective is to safeguard his own state and forces, not the enemy’s, both insubordinations now appear logically consistent. In many other cases throughout the text, these similar logical inconsistencies instantly vanish once one accepts that “preserving oneself” is the highest priority.[24]



In a review of multiple problematic English translations of another popular ancient text, Chinese philosophy professor Paul Goldin notes that we are prone to seek incompatible modern interpretations, and too often discount the controversial and uncomfortable ideas these ancient texts espouse:

The Daode jing is old; it is alien; it is Chinese; and it is difficult. These are the recalcitrant facts that too many readers seem disinclined to accept…Not much research is necessary to discover that there is more to Daoism than “letting events take their course” and that the scary political overtones cannot be disregarded as the detritus of imaginary interlopers. Like any profound work of philosophy, the Daode jing is dangerous. We do it no justice by pretending that it is easy to swallow.[25]

Similarly, The Art of War  is much more than “winning without fighting,” and we do a disservice by promoting the view that it eschews violence or exists as a convenient strategic antipode to bellicose Western military thinking. As Robin McNeal notes in Conquer and Govern, the idea that military treatises of the aptly labelled Warring States era rejected the efficacy of force does not reflect the thoughts found in the texts themselves, but instead “grows out of a vision of early Chinese intellectual history that has been overdetermined by the late imperial ideology and a particular strain of Confucianism.”[26] We have long projected our own hopes and prejudices onto this enigmatic text. None of this means that we cannot or should not teach The Art of War in our professional military education institutions, but it would be helpful to periodically remind ourselves which side of the tapestry we are truly observing it from.

John F. Sullivan is a former U.S. Army China Foreign Area Officer. He is currently a J.D. candidate at the University of Hawaii’s William S. Richardson School of Law. This article is the final in a series seeking to revise the conventional interpretation of the first three verses of the third chapter in Sun Tzu’s Art of War. The second verse was analyzed in “Sun Tzu’s Fighting Words.” The third verse was considered in “Who Was Sun Tzu’s Napoleon?

Published THE STRATEGY BRIDGE.ORG,  July 13, 2021

Header Image: One of the terra cotta warriors of Emperor Qin Shi Huang (


[1] Hanfei, Hanfeizi, trans. W.K. Liao, vol.2 (London: Arthur Probsthain, 1959), 299. (Chapter 50).

[2] John Ciardi, “Translation–The Art of Failure,” The Saturday Review, October 7, 1961, 17-19.

[3] Miguel de Cervantes, Don Quixote, trans. John Ormsby (Woodstock, Ontario: Devoted Publishing, 2016), 448.

[4] For instance, in a modern Chinese language edition produced by Cui Guozheng (崔国政), he needs to use 12,413 characters to express the same meaning of the 6,000 character original.

[5] B.H. Liddell Hart, “Foreword” in Sun Tzu: The Art of War, trans. Samuel Griffith (Oxford: Oxford University Press, 1963), vi.

[6] Sun Tzu, The Art of War: The Oldest Military Treatise in the World, trans. Lionel Giles (London: Luzac & Co., 1910), 17.

[7] Ibid. Giles’ version is not technically the first English translation. In 1905, a British army officer stationed in Japan, E.F. Calthrop, published a version of the text in English, followed by an updated edition in 1908. However, Calthrop based his translation on a Japanese language version of the text, so as a translation of a translation, it is unsuitable for textual comparison with the original Chinese version.

[8] Sun Tzu, The Art of War, trans. Ralph D. Sawyer (New York: Basic Books, 1994), 177. Note also that Sawyer is one of the only translators to render the character 國 as the “state capital” rather than as a “state” or “country.” Sawyer is probably correct that at the time it was composed, this character more accurately referred to a territory’s main walled city, from which its power and authority flowed to outlying regions, rather than any defined boundaries that represented the territorial limits of a larger political unit.

[9] Edwin G. Pulleyblank, Outline of Classical Chinese Grammar (Vancouver: UBC Press, 1995), 13.

[10] As a textbook on classical Chinese notes: “Literary Chinese texts are often over-determined; that is, there may be several perfectly grammatical ways to explain the syntax of a sentence. Skill in reading, then, lies in deciding which alternative is most likely rather than simply whether the alternatives are grammatically possible.” Michael A. Fuller, An Introduction to Literary Chinese (Cambridge, MA: Harvard University Asia Center, 2004), 2. (emphasis in original).

[11] Giles, 159

[12] Ibid. (Footnote to verse 12.2).

[13] Sun Tzu, The Art of War, trans. Samuel B. Griffith (Oxford: Oxford University Press, 1963), 77.

[14] Although it is true that several of the historical commentators interpret Sun Tzu’s advice as applying to the enemy’s state and forces, their notion of “preservation” might conflict with a modern understanding of this concept. For example, Cao Cao believes this advice means that one should first invade the enemy state and push one’s army deep into its interior, cutting off lines of communication and supply and thereby hope that the enemy submits without resistance (the superior option). If this aspirational goal is not achieved, though, one will then be forced to attack the enemy’s army while still deep inside their home territory (the inferior option). Several of the commentators see parallels between Sun Tzu’s advice and the historical example of the Han dynasty general Han Xin’s conquest of the states of Zhao and Yan in 204 BCE. In this case, Han Xin invaded and brutally exterminated the state of Zhao, but then sent an emissary to Yan requesting immediate submission, which was granted. Thus Yan was “taken whole and intact.” It is clear from this historical analogy, though, that but for the example of the violent annihilation of Zhao, Yan most likely would have resisted Han Xin’s demand for immediate submission.

[15] D.C. Lau, “Some Notes on the Sun Tzu,” Bulletin of the School of Oriental and African Studies, University of London, Vol. 28, No. 2 (1965), 335. Edited to reflect the more modern pinyin transliteration of the Chinese character 全 (quan). Lau shows no awareness in his article that Lionel Giles had also previously considered the same conclusion in a footnote to his earlier translation.

[16] This logic holds even if we consider all of the uses of the character 全 found in the other works of the Seven Military Classics of ancient China. For example, in the Wei Liaozi, the text describes the “third military teaching” as “preserving the army” (全軍), in which the meaning is clearly referring to one’s own army, not the enemy’s. The original Chinese reads: 三曰全軍,謂甲首相附,三五相同,以結其聯也

[17] For example, in the Zuozhuan, China’s oldest historical narrative, an exegetical passage notes that when the ancient Chinese texts use the verb “to take” (取) it refers to militarily annexing an opposing state with little to no resistance: “In summer, Shi fell into disorder and was partitioned into three sections. Our troops went to Shi’s aid and went on to take it [取]. In all cases when the text has ‘took’ [取], it is to say that it was easy; when a great army was employed, it says ‘extinguished’; when territories were not retained, it says ‘entered.’” Zuo Tradition: Zuozhuan, trans. Stephen Durrant, Wai-yee Li, and David Schaberg (Seattle: University of Washington Press, 2016), 999. (Lord Xiang, Year 13).

[18] “When one’s opponents are in chaos, seize them [取].” Sun Tzu, The Art of War, trans. Victor H. Mair (New York: Columbia University Press, 2007), 79. “Therefore, he who stubbornly persists though his forces are fewer will be captured [擒] by an enemy whose forces are more numerous.” Ibid, 86.

[19] Sun Tzu, trans. Sawyer, 177 (emphasis added).

[20] Sun Tzu, The Art of Warfare, trans. Roger Ames (New York: Ballantine Books, 1993), 161 (emphasis added).

[21] Note that Roger Ames is the only other translator who follows D.C. Lau’s suggestion: “It is best to keep one’s own state intact; to crush the enemy’s state is only a second best. It is best to keep one’s own army, battalion, company, or five-man squad intact; to crush the enemy’s army, battalion, company, or five-man squad is only a second best.” Sun Tzu, trans. Ames, 111. It is not surprising that Ames takes this view, given that D.C. Lau was previously his graduate school advisor.

[22] Thucydides, The Landmark Thucydides, ed. Robert B. Strassler (New York: Touchstone, 1998), 257. (Book IV.64).

[23] Sun Tzu, trans. Griffith, 128.

[24] For instance, even before the reader leaves the third chapter, Sun Tzu discusses force ratios in which he advises that one attack the enemy if one holds a 5:1 numerical advantage. If one truly holds a fivefold numerical advantage and the overall goal is to preserve the enemy intact, why would one need to resort to an attack at all? Conversely, if one is focused on preserving oneself but not necessarily the enemy, a 5:1 advantage would certainly make an armed attack appear as a less risky path to certain victory—even if one’s ruler gives an order not to initiate a fight. As I discussed in a previous article, even Sun Tzu’s admonition against conducting sieges does not derive from a concern about the enemy’s welfare, but instead derives from a desire to protect one’s own forces and combat power. See John F. Sullivan, “Reconsidering Sun Tzu” Parameters, 49 (1–2) Spring–Summer 2019, 72-74.

[25] Paul R. Goldin, “Those Who Don’t Know Speak: Translations of Laozi by People Who Do Not Know Chinese” in After Confucius: Studies in Early Chinese Philosophy (Honolulu, HI: University of Hawaii Press, 2005), 133. Although it should be acknowledged that Professor Goldin himself advocates for the conventional interpretation of the first verse of the third chapter in The Art of War. See Goldin, The Art of Chinese Philosophy: Eight Classical Texts and How to Read Them (Princeton: Princeton University Press, 2020), 155.

[26] Robin McNeal, Conquer and Govern: Early Chinese Military Texts from the Yi Zhou shu (Honolulu: University of Hawaii Press, 2012), 1.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s