Sun Tzu on Espionage or: How I Learned to Stop Worrying and Love the Double Agent
John Sullivan
It is no state secret that intelligence professionals tend to favorably view Sun Tzu’s The Art of War. Whereas the West’s most famous philosopher of war, Clausewitz, appears to disparage the importance of military intelligence, the Chinese sage revels in the clandestine arts.[i] Well known verses such as “He who knows the enemy and himself will never in a hundred battles be at risk,” reflect the raison d'être of modern intelligence agencies.[ii] When Clausewitz argues that the use of cunning is a waste of valuable military resources, Sun Tzu boldly declares that all warfare is based on deception.
Beyond a few aphorisms scattered throughout the text, Sun Tzu also expands his thinking on intelligence in his thirteenth and final chapter, “Using Spies.” In the introduction to his 1963 translation of the text, Samuel Griffith notes that “Sun Tzu’s chapter on secret operations, [is] as pertinent today as when he composed it [and] requires little elaboration.”[iii] We should take a fresh look at this claim. A careful analysis of the chapter reveals an outdated and problematic espionage system that should be viewed today more as a historical curiosity than a ready guide to contemporary spy-craft. Despite popular sentiment, Sun Tzu’s The Art of War is ultimately a product of the period and place in which it was composed, not a timeless recipe for universally applicable strategic wisdom.
The final chapter’s status within the overall text is not without controversy. In the penultimate chapter, “Fire Attacks,” Sun Tzu quickly dispenses advice on incendiary methods before summing up the importance of understanding war, in general, to the safety and security of the state. This final section serves almost as a bookend to the opening verses of the first chapter, leading some scholars to view Chapter 12 as the original conclusion to The Art of War.[iv] This view is buttressed by the fact that the oldest extant copy of the text includes a reference to Su Qin in the thirteenth chapter, a well-known strategist from the late Warring States period.[v] His inclusion dates at least this version of the final chapter to no earlier than 284 BC, more than two centuries after the original text is traditionally thought to be written.[vi]
Even if future archaeological finds determine Su Qin’s inclusion to be a later edit to an already completed chapter, the fact remains that much of the information found in the espionage section is poorly integrated within the first twelve chapters of the text. In Chapter One, there is no reference to one’s espionage efforts in deriving the initial assessments that Sun Tzu notes is the key to preparing for war. In Chapter Two, Sun Tzu lists in detail all of the personnel, equipment and funding needed before engaging in war without mentioning the staffing or budget of the nation’s espionage system. While the thirteenth chapter discusses the intelligence, one needs to successfully conduct a siege, the third chapter argues that siege warfare should be avoided at all costs.
In both Chapters 7 and 11, Sun Tzu repeats an admonishment that “local guides” (yin dao) need to be recruited from the enemy population to provide information on the adversary’s terrain in preparation for an invasion. From Chapter 13, it seems clear that these guides would fall under the first group of Sun Tzu’s fivefold classification of all agents—“local spies” (yin jian). The fact that the text shifts from using the term local “guides” (dao) in earlier chapters to local “spies” (jian) in the last, lends further credence to the theory that the thirteenth chapter is most likely a later supplement to an original twelve-chapter core text.
Controversy on dating this portion aside, the chapter on espionage opens with an interesting argument. If one is willing to spend vast sums of money on equipping, mobilizing and deploying an army to conduct warfare, baulking at the expense of obtaining information on the enemy to support these operations is the height of folly. Thrice in this comparatively short chapter Sun Tzu will emphasize the need to spare no expense in compensating one’s spies – and in particular one type of agent – a revealing point that we shall return to shortly.
Sun Tzu next argues that information about the enemy should come primarily through foreknowledge, which he defines as information coming from men who know the enemy’s situation directly. Modern commentators occasionally deploy this verse to criticize the U.S. for its perceived failure to adequately emphasize human intelligence (HUMINT), particularly in its ongoing counter-terror and counterinsurgency operations. This is somewhat of a specious argument, as Sun Tzu was obviously not emphasizing HUMINT over the more technical collection methods available today. Instead, he was arguing that obtaining information directly from living people was more useful than relying on inquiries to spirits or divinations, an all too prevalent method during the period in which The Art of War was composed.
Nothing from the text supports a conclusion that Sun Tzu would necessarily be less prone to over-reliance on technical intelligence collection methods if afforded the opportunity. Given the book’s overarching emphasis on using terrain to gain decisive advantage over the enemy, one can easily imagine a modern day Sun Tzu micromanaging the deployment of aerial imagery assets and poring over advanced terrain models at the expense of other collection priorities.
The heart of Sun Tzu’s system, though, is his classification of intelligence agents into five groupings and the use of these classes of spies for espionage. The text identifies them as follows:
Local spies: citizens from the enemy state that are employed by us
Inside agents: enemy officials that are employed by us
Double agents: enemy agents originally intending to exploit us, who we then recruit to our side
Expendable spies: those deliberately fed false information to be spread amongst the enemy (presumably executed once the spy’s true purpose is discovered)
Unexpendable spies: those sent to gather information on the enemy and then return to report
As taxonomy, this classification system is somewhat problematic. Sun Tzu is not entirely consistent with the criteria used to distinguish the various classes of spies. While local spies, inside agents, and double agents are all assigned based on their citizenship (of the enemy state), expendable and unexpendable spies are designated based solely on how they are to be used. Is Sun Tzu’s intent that a local spy, inside agent or double agent should never be utilized in the same capacity as either an expendable or unexpendable spy? If this is somehow not recommended, the reasoning for this prohibition is not quite clear.
It should also be noted that Sun Tzu’s definition of spies diverges from our contemporary understanding of the term. We tend to define spies today as part of a professional organization composed primarily of one’s own citizens, who are then tasked with collecting information on potential adversaries and rivals. In modern terms, what Sun Tzu is describing would be more akin to intelligence assets – that is, those people used by our intelligence services to gain access to the information we desire, or else manipulated as a conduit to pass messages or false information to the target nation.
This is disconcerting, as a professional layer of intelligence management is completely absent from Sun Tzu’s espionage system. According to the text, the commander (or sovereign) interacts directly with the intelligence assets without any intervening trusted filters in place to help prioritize collection requirements or else verify and critically analyze the information being supplied by these foreign agents. For simple operations involving a very small cadre of assets, this might be a workable solution. For the complex issues that modern states and their supporting intelligence agencies are tasked to deal with, Sun Tzu’s concept is far too primitive to seriously consider as a useful template.
Another area of concern is the fact that within Sun Tzu’s model, not all classes of spies are considered equal. Counterintuitively, the double agent is unequivocally placed at the apex of the classification strata and is the sine qua non of the entire intelligence apparatus. The text makes clear that it is only from the double agent that one can target and recruit the necessary local spies and inside agents, and only from the double agent’s direction can we provide the correct information to both feed the expendable spy and task the unexpendable spy to collect. For this service, Sun Tzu once again highlights the need to lavishly compensate the double agent.
That Sun Tzu would place so much trust and confidence in someone who, by definition, has already once betrayed their former employers, is not easy to comprehend without further context. In reality, his idea of a double agent is markedly different than who we generally think of when we use the term today. As the text elaborates:
... since the key to all intelligence is the double agent, this operative must be treated with the utmost generosity. Of old the rise of the Yin (Shang) dynasty was because of Yi Yin who served the house of Hsia; the rise of the Chou dynasty was because of Lü Ya who served in the house of Shang.[vii]
Through his use of Yi Yin and Lü Ya as exemplary double agents, it becomes clear that Sun Tzu is not simply referring to mid-level bureaucrats who choose, for various reasons, to betray their native or adoptive country. While these agents might possess sensitive and potentially damaging intelligence, they are also unlikely to be privy to the spectrum of information and in-depth knowledge at the highest levels of decision-making that would significantly alter the balance of power between two rivals.
In contrast, Yi Yin was one of the most powerful counsellors to the ruler of the Hsia (Xia) before betraying his sovereign and lending assistance to the founders of the Shang dynasty in usurping the throne. Lü Ya was the most experienced military commander of the Shang dynasty before offering to fight against his former comrades in service of the rebel clan that went on to establish the Chou (Zhou) dynasty.
To put this into a contemporary, comparable hypothetical, this would be the equivalent of General H.R. McMaster defecting to Beijing and then being provided a position as Xi Jinping’s most trusted advisor in the Central Military Commission. Clearly, based on his extensive knowledge of all aspects of U.S. national security policy, access to the most sensitive classified intelligence, insight into the thinking and personal habits of senior leadership and detailed understanding of the strengths and vulnerabilities of both the U.S. and its allies, the level and scope of information supplied could irrevocably shift the balance of power in China’s favor. Any number of emoluments provided as compensation for the hypothetical perfidy would pale in comparison to the potential advantages to be gained.
During the era in which the espionage chapter of The Art of War was most likely composed, this was an all too common occurrence. In addition to the previously mentioned Su Qin, other notable senior officials who switched sides to the detriment of their native state include: Wu Zixu, Bo Pi, Zhang Yi, Shang Yang, Fan Ju, Lü Buwei, Han Feizi, Li Si and Zhao Gao. Historians of this period extensively document the misfortune and devastation these double agents caused their former masters by offering their knowledge and rolodex to the highest bidder.
Becoming a double agent, though, was also a highly risky venture if political fortunes shifted out of one’s favor. Of the examples, four were brutally executed (Shang Yang was ripped apart by oxen while Li Si was chopped in half at the waist following mutilation), three were compelled to commit suicide rather than endure torture and execution, and most faced repeated assassination attempts while in service to their adopted state. Few died peacefully in their sleep.
In hindsight, the chapter’s three explicit admonishments to spare no expense in rewarding double agents, coupled with the insistence that these seditious individuals serve as the sole gatekeeper for information flow on behalf of the sovereign and commander, should give us pause. The espionage apparatus being proffered almost too conveniently serves the direct interests of a duplicitous official needing to quickly consolidate a secure base of power within a new and perilously uncertain political patronage system. If this theory has validity, this would make the espionage chapter not only part of the oldest military treatise in the world, but perhaps also the oldest defense contractor lobbying prospectus in existence.
Circumstances have changed substantially over the interceding millennia. The expectation that one could simply land an operative so consequential that it would fundamentally alter the balance of information between two great powers is a rather thin reed to hang the hopes of a modern espionage system. Intelligence today is a game of steadily accumulated inches, not Hail Mary passes. Despite portions of The Art of War remaining critical to understanding contemporary warfare and strategy, the text’s theory on espionage appears to be based less on universal strategic principles than on a practical means to survival within the confines of an all-encompassing and brutal civil war.
Intelligence ultimately derives its value based upon timing. Our respect for the strategic canon should not blind us to the fact that its works were composed during vastly different historical periods. While carefully derived tenets translate well within a modern setting, other pieces of advice have simply grown stale with age. Sun Tzu’s The Art of War deserves both our continued admiration and attention, but as a fictional disciple of the Chinese general warned about investments, “don’t get emotional about stock, it clouds your judgement.” It’s not that Sun Tzu’s theory on espionage is better or worse than Clausewitz in terms of guiding modern intelligence practices. As Gordon Gekko might acerbically note, it’s simply a dog with different fleas.
End Notes
[i] “In short, most intelligence is false, and the effect of fear is to multiply lies and inaccuracies. As a rule most men would rather believe bad news than good, and rather tend to exaggerate the bad news.” Carl von Clausewitz, edited and translated by Michael Howard and Peter Paret, On War (Princeton: Princeton University Press, 1976), p. 117. To be fair, Clausewitz’s disparagement of intelligence applies mostly at the tactical and operational levels of war and reflects the often crude methods of intelligence collection available to commanders in the midst of early nineteenth century battles. It is unfair to say that Clausewitz categorically dismisses the importance of intelligence at the strategic level. For instance, his concept of identifying the enemy’s center of gravity presupposes a detailed intelligence assessment of the enemy’s overall capabilities and strategic situation.
[ii] Roger T. Ames (translator). Sun-Tzu: The Art of Warfare. (New York: Ballantine Books, 1993), p. 113.
[iii] Samuel B. Griffith (translator). Sun Tzu: The Art of War (London: Oxford University Press, 1963) p. 44.
[iv] E. Bruce Brooks. “The Present State and Future Prospects of Pre-Han Text Studies.” Sino-Platonic Papers 46 (July 1994), p. 59.
[v] The oldest extant copy of The Art of War, known as the Yinqueshan (Silver Sparrow Mountain) manuscript, was recovered from an excavated tomb discovered in Shandong province in 1972. The text is thought to be dated to approximately 140 BC, roughly three and a half centuries after the original text is traditionally believed to have been written. See Ames, p. 15.
[vi] Mark Edward Lewis. “Writings on Warfare Found in Ancient Chinese Tombs.” Sino-Platonic Papers 158 (August 2005), p. 5.
[vii] Ames, p. 171.