“Peeling Facts off the Face of the Unknown”

Revisiting The Legacy: Sherman Kent, Willmoore Kendall, and George Pettee—Strategic Intelligence in the Digital Age

Anthony Olcott

To be “the father of intelligence analysis,” as Sherman Kent has so often been called, means of course that Kent was its founder, but it also means that his intellectual “genetics” lie deep within our enterprise. The Kent legacy has survived because his approach to intelligence analysis served the United States extremely well for a long time. However, as happens when environments undergo dramatic change, successful adaptations for one environment can prove to be much less efficacious—perhaps even fatal—in a new environment. One need only consider the doom hanging over the two professions that Kent held up as the models that he wished the “intelligence profession” to emulate—the “large university faculty” and “our greatest metropolitan newspapers”1 — to see how profoundly our environment has changed.2

The views of Kendall and Pettee found little traction in their day but now seem to have important lessons for how the intelligence profession might change if those of us who practice it wish to escape extinction.

In addition to extinctions, though, profound changes of environment also can reveal the adaptive virtues in structures and approaches that did not thrive in the past. Kent had at least two contemporaries, Willmoore Kendall and George Pettee, who outlined quite different approaches to strategic analysis.3 Their views found little traction in their day but now seem to anticipate in striking ways the vision of the future of the intelligence community published by the director of national intelligence (DNI) in 2008. This suggests the “paths not followed” have important lessons for how the intelligence profession might change if those of us who practice it wish to escape the looming extinction of the tenured faculty member and the professional journalist.

*     *     *     *     *

Things to Remember about Strategic Intelligence

Kent’s best-known public work was prescriptive, not descriptive—that is, Kent wrote it from a kind of self-exile following the postwar dismemberment of his OSS research unit as part of the State Department’s bureaucratic battle to retain ownership of intelligence. That Kent was describing not what existed as he wrote in 1947 but what he urged should exist places in context the attention he devotes to the nature of the intelligence analyst and the relationship the analyst should have with the policymaker.4

Nearly every line of Strategic Intelligence derives from the conviction that for any situation, for every occurrence, for every phenomenon, there exists a single truth.

The policymaker, whom Kent dubbed “the consumer” of intelligence, was thus not so much an abstract entity as it was the foreign service officer who claimed superior knowledge of the countries in which he or she had served and thus agreed to use the products of Kent’s research unit only to the degree that its findings coincided with that officer’s previous understandings (Kent, 114). Kent’s contempt for such consumers was withering: he cited Hitler as the prime example of a consumer who had “brilliant hunches” but who did not try to “analyze the why of his successful intuition,” preferring instead to see his intuition as a “natural, personal, and infallible source of truth” (Kent, 204).

What Kent saw to be Hitler’s mistake—that the dictator had failed to probe more deeply—highlights another point about Kent: his view of reality was profoundly and unshakably Platonic. Nearly every line of Strategic Intelligence derives from the conviction that for any situation, for every occurrence, for every phenomenon there exists a single truth. Kent states this assumption most clearly in a footnote, when he explains:

By “objective situation” I mean the situation as it exists in the understanding of some hypothetical omniscient Being. I mean the situation stripped of the subjective characteristics with which a prejudiced human observer is almost certain to endow it. I use the word “probable” because, whereas knowledge of the objective situation is of highest desirability, any non-omniscient Being (i.e., any frail human being) probably can never apprehend the true objective fact. He should, however, strive until it hurts [emphasis in original] (Kent, 41–42).

Discovering “true objective facts” is not only arduous, but it also is cumulative. As Kent wrote: “Research is the only process which we of the liberal tradition are willing to admit is capable of giving us the truth, or a close approximation to truth” (Kent, 155). The fruits of that research, and thus the policymakers’ ability to approximate the truth, were put badly at risk by the (to Kent) frivolous postwar “demobilization” of his research team. Kent used that term in “The Need for an Intelligence Literature,” which he wrote in 1955 for the inaugural issue of Studies in Intelligence in part to argue that intelligence was now firmly enough established to be considered a profession.

It is worth unpacking what Kent meant. For one thing, a profession, unlike a job, requires special training and a period of tutelage. It is also not something for which all people are fit. “Twenty men with a mental rating of 5 put together in one room will not produce the ideas of one man with a mental rating of 100, and you cannot add minds as if they were so many fractional parts of genius.” (Kent, 174–75). A profession is also something in which people engage for reasons other than money—“People work at [intelligence] until they are numb, because they love it, because it is their life, and because the rewards are the rewards of professional accomplishment” (Kent, “Need”: 2).

Kent gave three reasons for maintaining a space between analyst and policymaker, the most important of which was that “captured” intelligence analysts will end up “swinging behind the ‘policy’ of the operating unit.”

More importantly, however, making intelligence analysis a profession also endowed the activity with self-justifying autonomy. It is the rigor of the selection process and the willing self-abnegation of its rewards that justify the most distinctive feature of Kent’s vision of strategic analysis— the strict separation of analysts and policymakers that remains a unique feature of US intelligence to this day. Kent gave three reasons for maintaining a space between analyst and policymaker, the most important of which was that “captured” intelligence analysts will end up “swinging behind the ‘policy’ of the operating unit” and [thus be] “prostituting itself in the production of what the Nazis used to call kampfende Wissenschaft (roughly, knowledge to further aims of state policy)” (Kent, 200).


[Top of page]


Kent Channeling Walter Lippmann

The source quoted to justify Kent’s remedy for the danger of “prostitution” is Walter Lippmann’s classic of 1922, Public Opinion.5 The passage that Kent cites, which explains that the only way to safeguard “impartial and objective analysis” (Kent, 200) is to keep the “staff which executes” as separate as possible from “the staff which investigates,” as “parallel but quite distinct bodies of men,” appears near the end of Lippmann’s book, in the section “Organized Intelligence.”

Just as Kent’s book was in many ways a reaction to his experiences during World War II, so was Lippmann’s a reaction to his during World War I. Although his blood was not quite as blue as Kent’s,6 Lippmann too was high-born and wealthy, a Harvard man to Kent’s Yale, and just enough older to have been pressed into public service during World War I, as an unacknowledged adviser to President Wilson. Lippmann emerged from the experience with the paradoxical but not uncommon conviction that democracy is too good an institution to be trusted to ordinary people. The problem, he argued in Public Opinion, is that humans can only know well the things that lie in their immediate experience. Beyond that, their knowledge becomes increasingly second-hand, based upon what Lippmann termed “the pictures in their minds,” which leaves them open to error or, even worse, makes them susceptible to outright manipulation.7

Kent’s analysts will have “horse sense,” with “a set of well-stocked and well-ordered brain cells.”

The only way that America’s policymakers might be kept reasonably on task, Lippmann argued, was to create “intelligence bureaus” that would serve in each of the government’s 10 departments as permanent repositories of deeper knowledge. These “intelligence officials” (Lippmann’s term) would be “independent both of the Congressional committees dealing with that department, and of the Secretary at the head of it” so that “they should not be entangled either in decision or in action.” To ensure this freedom from “decision or action,” these groups of “intelligence officials” would have tenure for life, “with provision for retirement on a liberal pension,” regular sabbaticals, and could be dismissed “only after a trial by professional colleagues.”8

[Top of page]


The Liberal Professions

Lippmann and Kent invested their intelligence professionals with so much autonomy because the Platonic model on which they both based their thinking was an innately hierarchical one. The “liberal tradition” to which Kent refers had its roots in imperial Rome, where the things that men might know were divided into “arts and skills” (techne in Greek, hence “techniques” and “technology”), which were appropriate to slaves, and the intellectual realm of free men (liber = Latin for “free”), which included rhetoric, or the power of persuasion, and leadership.9

For that reason, despite all of Kent’s attempts to describe a system of intelligence analysis, what he advocates relies almost entirely upon a self-defining and self-policing elite that, because it engaged in the selfless struggle to draw ever nearer to truth, is not to be second-guessed by politicians or anyone else. To be sure, Kent’s analysts were not to function entirely independently, but their relationship to policymakers was to “stand behind them with the book opened at the right page, to call their attention to the stubborn fact they may neglect.”

The analyst-policymaker model, as Kent makes plain, is that of “professional man” and “client” (Kent, 182). Just like the self-policing tenured “intelligence staff” that Lippmann envisioned, Kent’s analysts will have “horse sense” (Kent, 164), with “a set of well-stocked and well-ordered brain cells” (Kent, 65). They will be “wise men” whose “mysterious inner selves are of the kind which produce hypotheses of national importance” (Kent, 161) and, as such, are not to be confused with German or Soviet “party ‘intellectuals’ purporting objectively to prove such phenomena as Aryan Supremacy,” as Kent wrote in a dismissive footnote (Kent, 200).

Analysts envisioned thus as “professionals,” Kent can belittle the consumer who “insists that no idea is too complicated for the 300-word summary” because, by rejecting the deeper knowledge of the analysts, such a consumer is contributing to the “demoralization of his intelligence staff” (Kent, 176). The consumer who chooses to function without giving the analysts “sharp and timely guidance” is to Kent the main cause of “the worst sickness which can afflict intelligence,” because “when intelligence knows little or nothing of what lies behind a request,” the analysts “lose desire to participate,” become “dumb and unhappy automatons” who “long ago…quit caring” (Kent, 183).

Kent asserts that “pardonably wrong diagnosis and understandably inadequate presentation” must be forgiven, just as one might forgive “the dentist who pulls out the wrong tooth.”

Most tellingly, Kent asserts that “pardonably wrong diagnosis and understandably inadequate presentation” must be forgiven, just as one might forgive “the dentist who pulls out the wrong tooth” or the “lawyer [who] loses a case” (Kent, 194) because “when the findings of the intelligence arm are regularly ignored, the consumer should recognize that he is turning his back on the instruments by which Western man has, since Aristotle, steadily enlarged his horizons—those of reason and scientific method” (Kent, 206). Indeed, Kent argues darkly, unless analysts are spared from all the normal rules, regulations, and considerations of their own efficacy, “the whole question of the preservation of the democratic way may itself become somewhat academic” (Kent, 147).

[Top of page]


One Road Not Taken: Willmoore Kendall

Although Willmoore Kendall is one of four colleagues whom Kent thanks for “readings of the manuscript and many kinds of advice” (Kent, x), it is hard to imagine two men more dissimilar in their backgrounds and beliefs—no blue blood, Kendall was the son of a blind, itinerant preacher who had moved his family from Kentucky to Oklahoma, where Kendall was born.10 Self-taught, Kendall enrolled in university at age 14, won a Rhodes Scholarship, and nearly completed a PhD in French literature before switching to political philosophy. His first book was on baseball and later works included a study of John Locke, articles on the poet John Milton, and a quantitative study of American voting behavior.

By all accounts Kendall was the antithesis of an “organization man,” unable to stay on speaking terms with more than one colleague at a time, and—to judge at least by the transcripts of some his public presentations—savagely condescending to those he considered intellectually his inferiors, so it is unsurprising that Kendall played no discernible role in the formation of the CIA or the foundation of the “intelligence profession.”11

Kendall argued, intelligence needs are strategic, because policymakers faced what Kendall called “the big job—the carving out of United States destiny in the world as a whole.”

The one public instance when Kendall did lay out his thoughts on intelligence, however, makes clear that Kendall would have argued vigorously with Kent. Already back at Yale,12 Kendall reviewed Kent’s Strategic Intelligence for the then-new journal World Politics.13 The review breaks into two parts. The first is devoted to inconsistencies in Kent’s argument, or to places where Kent (in Kendall’s view) fails to draw out the full conclusion of what he has been saying. Many of these criticisms still resonate—Kendall agrees, for example, with Kent’s fear that the security requirements of covert collection would make overt collection difficult, but goes even further, to assert that “our present intelligence arrangements…enormously exaggerate the importance of covert collection, and yet permit it to yield shockingly small dividends.”14 It is in the second part of the review, however, that Kendall critiques what he sees to be Kent’s “state of mind,” which he argues has four major shortcomings (Kendall, 548–52).

First, Kent saw the intelligence needs of wartime and peacetime as essentially the same, which Kendall viewed as a dangerous mistake. Wartime intelligence needs are primarily tactical—the enemy is known and the goal is clear—while in peacetime, Kendall argued, intelligence needs are strategic, because policymakers faced what Kendall called “the big job—the carving out of United States destiny in the world as a whole.” (Kendall, 548). This was not a matter of empirical fact, but rather of what Kendall elsewhere calls “ethics.”15

Although Kendall obviously had views about what that destiny should be, he did not take the triumph of those views as a self-evident scientific “fact,” as did Kent. Rather he defined that destiny as a belief system which, he argued adamantly, was not the property of some small, self-selected elite to establish, but rather was to be decided by what Kendall considered to be sole determinant of “right” in a democracy—the “one-half of the members, plus one” which constitutes the majority of any group. (Kendall, Contra Mundum, 93). The Kent “state of mind,” Kendall argues, views “the course of events” to be “a tape, all printed up inside a machine; and the job of intelligence is to tell the planners how it reads.” (Kendall, 549) Far better, he argues, to understand this course as “something you try to influence.”16 Kendall also rejected Kent’s division of intelligence into “domestic” and “foreign,” because that prevents analysts from examining how US actions might change a given situation.

The second problem with the Kent “mind set” was to see the business of government, and therefore of intelligence, as properly being conducted by professionals, the “producers” being the intelligence officers and the policy planners the “consumers.” To Kendall, this makes analysts “mere research assistants to the George Kennans.” Unlike Kent, the liberal elitist, Kendall was a “majoritarian,” a deeply literal believer in democracy who saw the legitimate functions of US government as being performed by “the people,” as indicated by the majorities of voters who had elected their various representatives.17

Although he had strong beliefs about what is moral and what is not, Kendall regarded his beliefs—all beliefs—as lying outside the “right-wrong” paradigm. “Rightness” is not in Kendall’s view to be determined by an elite corps of enlightened specialists, but rather by a strict majority. To Kendall, the true uniqueness, and the value, of America lay in the fact that it is as close to a genuine democracy as humans had yet produced—meaning that if you disagreed with a national policy, the proper route to changing it lay not in capturing control of a functional bureaucracy, but rather in convincing a majority of the populace to agree with you. Thus the minds in which the pictures should be changed belong to “the politically responsible laymen,” meaning the lawmakers, as well, presumably, as the minds of those who elect them. It is intriguing to imagine the arguments that Kent and Kendall must have had on this score when Kent defended his notion that democracy could be preserved only by a non-democratic, self-selecting and self-policing elite.

Kent was a puzzle-solver.… Kendall, by contrast, was more of a mystery solver, who argued that belief systems are arbitrary constructions that can never be proven to be true or false.

The third shortcoming was Kent’s “crassly empirical conception of the research process in the social sciences,” a political scientist’s oblique dig at the historian’s tendency to equate intelligence value to volume of data. Kendall argued that the research process should also offer scope for “‘theory’ as it is understood in economics and sociology” (Kendall, 551). In the “empiricist” world of Kent, the biggest challenge was to find better ways to “process” what Kendall called “a tidal wave of documents.” To Kendall, the challenge rather is to enable analysts to “work under conditions calculated to encourage thought.” Kendall points out that, although Kent makes many references in his book to social scientists and social science, he “never employs in that connection the words theory and theorist.”

The distinction Kendall was making comes quite close to what Gregory Treverton much later characterized as the difference between a puzzle and a mystery.18 Kent was a puzzle-solver—indeed, his signature gift to friends was a so-called “buffalo block,” a handcrafted wooden puzzle that could only be taken apart, and reassembled, in one, and only one, intricate and precise way.19 Kendall, by contrast, was more of a mystery solver, who argued that belief systems are arbitrary constructions that—importantly—can never be proven to be true or false.20 As he wrote in one essay, “You cannot deduce from an ethical judgment…a canon of scientific proof,” because “Science tells us not what things to do but how to do things” (Kendall, Contra Mundum, 100–101).

It was not the errors of the past that drove Pettee’s concern, but rather his certainty that peacetime would face the United States with even greater challenges than had the just-finished war.

Finally, Kendall accused Kent of “uncritical optimism” in assuming that there would always be adequate numbers to perform the tasks “upon which the intelligence function depends” (Kendall, 552). To Kendall, there could be “no more dangerous assumption.” In fact, Kendall argued, the supply of specialists “in fields other than History” (Kendall’s parting dig at Kent) was critically short, and would grow worse. What Kendall appears to have been arguing was that real “high-level foreign positive intelligence” was not an issue of fact-finding, but rather required coming to understand better how humans make and modify their beliefs.

[Top of page]


Another Road Not Taken: George S. Pettee

Given Kendall’s famously arrogant and contentious personality, it is little surprise that his views on strategic intelligence failed to gain traction. It is less clear, however, why the views of George Pettee were also left by the wayside as the US Intelligence Community was taking its present form.21 Just a year younger than Kent, George Pettee was yet another academic who got put to work in wartime Washington. Pulled from Amherst College’s Department of Political Science, Pettee quickly rose to become chief of the European Enemy Division of the Foreign Economic Administration (FEA), which provided industrial targets for Allied bombers and also tried to limit enemy access to strategic raw materials.

In that sense, Pettee’s wartime activities were similar to those of Kent, who had harnessed open source information to provide operational support to the invasion of North Africa. Unlike Kent, however, who came away from that experience convinced of the value that “objective facts” might offer to strategic analysis, Pettee emerged deeply skeptical of the value of such data—particularly after he could study the US Strategic Bombing Surveys22 that were produced to gauge the efficacy (or, quite often, lack of efficacy) of target recommendations made by him and his colleagues.

That skepticism informs The Future of American Secret Intelligence,23 which Kent acknowledged as “a trail breaker in the literature of strategic intelligence” (Kent, xi). To Pettee, the problem was not that the data he and his colleagues had provided was itself incorrect—they had more or less gotten right the amount of steel the Nazis had produced, the oil they were consuming, and so forth—but rather that those figures had proven to be simply numbers that the analysts themselves had embedded into contexts and assumptions that in fact were completely wrong. As he wrote, “We won the war…in spite of, and not because of, the fact that our intelligence system showed many failings” (Pettee, 2).24

It was not the errors of the past, however, that drove Pettee’s concern, but rather his certainty that peacetime would face the United States with even greater challenges than had the just-finished war. Unlike Kent, who saw “Aryan supremacy” and similar doctrines as manifestly false beliefs that better data could disprove, Pettee was a student of revolutions,25 who had argued, even while an instructor at Harvard, that totalitarianism was a logical and attractive option that could be battled only by overt government counterpropaganda—what he termed “conscious guidance”26 that could provide an even more attractive alternative narrative to those tempted by the blandishments of hostile ideologies.

Like Kendall, Pettee argued that beliefs exist separately from data. In a talk given to the Industrial College of the Armed Forces (ICAF) in 1947, Pettee said that, in addition to their physical economies of production and trade, nations also have what he termed an “ethical economy”27 that can prove even more important than the other two kinds. In his book, Pettee asserts that it was the Nazis’ ability to “predict and influence the process of events on the general economic-political-psychological level” that determined their early success, while “we, at the time, were only surprised and dismayed by both the process of events and the apparent Nazi ability to control it” (Pettee, 29).

“Fundamental lack of doctrine [was a] notable characteristic of wartime intelligence agencies [and] remains…an outward indication of the fundamentally amateurish basis” on which strategic intelligence continued to be approached.

The peacetime challenge, of course, was Marxism, which Pettee saw as a “sociological weapon” that, despite being “about 80 years old today [in 1947]” is “more modern…than anything you have ever known” (ICAF, 9). Pettee warned that there is “a long record of American ignorance or misunderstanding of what makes world politics operate” that can no longer be tolerated if the United States is going to “accept active leadership of the world in seeking positive means to eliminate the causes of war” (Pettee, 39). Pettee warned that “Russia will do her best to use a modern approach to the problems which arise, to call her shots, and if we try to play our part without the tools of thought there will be no question of the outcome” (Pettee, 44).

Like Kent, Pettee saw the solution in research: “there is no case of failure [of policy in the interwar years] that cannot be fully explained by ignorance of economic and political facts and relationships” (Pettee, 44). But unlike Kent, Pettee argued that research, the gathering of facts, is useless if conducted separately from the functions of policy. As he wrote, “The failure to define clearly the special province of ‘strategic’ or ‘national policy’ intelligence…meant in the past that the conduct of the work lacked all the attributes which only a clear sense of purpose can give” (Pettee, 65).

Pettee also urged that intelligence analysis should become a specialized activity, with its own school, case histories for study, professional association, and a journal (Pettee, 99–100). However, Pettee did not wish to see intelligence analysis become a profession, because he saw a more important task—the creation of “a doctrine for strategic intelligence” (Pettee, 97). Unlike a profession, which is self-justifying and self-policing, a doctrine depends upon outcome—as Pettee wrote, strategic intelligence analysis required “a doctrine of method, namely what data to seek as essential to the solution of a problem” (Pettee, 97).

Unlike Kent, whose analysts pursued subjects based on their own “horse sense,” Pettee warned that “individual initiative can only lead to disorder unless the individual understands the purposes and structure of the program in which he plays a part…. The remedy for misguided initiative must be through leadership from above and participation from below in a common doctrine” (Pettee, 97). It was the “fundamental lack of doctrine,” Pettee wrote, that was a “notable characteristic of wartime intelligence agencies” and “remains…an outward indication of the fundamentally amateurish basis” on which strategic intelligence continued to be approached (Pettee, 95).

[Top of page]


Kendall, Pettee, and Vision 2015

Pettee was even more specific about the necessity of linking analysis to purpose [than was Kendall].…”At the top is some kind of national brain taking in data…and making decisions.”

Pettee did not share Kendall’s disdain for the “George Kennans,” the unelected bureaucrats of policy planning, but he did insist that strategic intelligence could only be as good as the policy-formation apparatus that sat atop it.28 In his book’s final chapter, Pettee argued that, even if “all of the problems of the actual processing of intelligence” that he details “can be given radical solutions,” there still remains a final requirement, that there be “officials or offices in the government competent to act upon the conclusions.” (Pettee, 103)

Pettee’s words were different from those of Kendall, but he seems to have had something very like the “carving out of United States destiny in the world” in mind when he wrote that “the United States cannot have a national policy comparable to its commitments unless it has the means to form such a policy and base it upon the best possible knowledge of the facts and circumstances” (Pettee, 117). A government that could “find the truth in a situation, and could recognize the truth and act upon it when it found it” would be “less than a panacea,” but it would be “a positive test and sign that democratic government could learn how to unify intention and consequence” (Pettee, 118).

Pettee was even more specific about the necessity of linking analysis to purpose in the talk he gave to ICAF in 1950, where he outlined an ideal view of how intelligence and policy should function together:

Up at the top is some kind of national brain taking in data on the situation, estimating the situation, and making decisions—the fundamental, high-level policy decisions…. There is a flow of information about the situation from the world to the general staff. That is the intelligence function…. There is also the research function, which is studying things distinct from what these other people study —and are distinct not because they are domestic, not because they are foreign, but because nobody knows them yet. They are peeling facts off the face of the unknown…. Research must provide intelligence with data, intelligence must provide research with data, and things have to correlate at lower levels…. If that cycle—the flow of information, decision, command, and action—is rational, then you are using your resources to suit your purposes, and the consequences are coming out in accordance with your intentions. If the consequences are not coming out in accordance with your intentions, there is…a failure of function somewhere in that cycle.29

The diagram Pettee used to illustrate his lecture (facing page, above) bears a remarkable, and very suggestive, similarity to the depiction of the intelligence process offered at the end of the DNI’s Vision 2015: A Globally Networked and Integrated Intelligence Enterprise, published in 2008 (see right). Although there are some differences in the details, both diagrams share the hierarchical but integrated relationship of analyst and policymaker that Kendall and Pettee both advocated, and which Kent warned against. While paying lip service to the tradition of “speaking truth to power,” (Vision, 8) the DNI’s statement speaks of future intelligence analysis following “a more interactive model that blurs the distinction between producer and consumer” (Vision, 9) and that, rather than “asking customers, ‘What are your intelligence priorities?’ analysts will be expected to engage them with, ‘what do you want to accomplish?’” (Vision, 10)


Intelligence Chart1


Intelligence Chart2

[Top of page]



There are many forces that are converging to bankrupt the professions of the tenured faculty member and the professional journalist, but all of them point in the same direction—that society no longer requires intermediaries to stand between it and knowledge, or it and events, to declare that one set of things (writers, ideas, faiths, political traditions, etc.) is good, important, and true, while all the rest is to be ignored. Although the connection has perhaps become obscured, both the university system, as practiced in the United States, and the US tradition of journalism (since at least the end of the 19th century) derive from the same set of Platonic assumptions as did Kent’s prescriptions for strategic analysis.

The reason that the Kent method worked so well for so long is that, in the bipolar world of the 20th century, the differences between tactical and strategic intelligence were not great. In the Cold War, the presumed enemy was almost as evident as had been the real enemies in the hot war from which Kent, Kendall, and Pettee had all just emerged, and the strategic purposes of intelligence thus seemed clear. Both Kendall and Pettee, however, understood that Marxism was not an error to be countered with better facts, or a “kampfende Wissenschaft” like Aryan Supremacy to be dismissed as evil foolishness.

“Even if strategic intelligence is not exposed to public view, the officials who receive its advice will trust it and respect it in close proportion to its batting average.”

If the world were indeed as Kent imagined it to be, with “objective situations” that could be discerned by “western man enlarging his horizons” through “reason and scientific method,” then strategic analysts should probably be the kind of academic, dispassionate elite whom Kent and Lippmann envisioned. It seems far more realistic, however, to accept that our current global terrain is like that in a quote provided by John Heidenrich, that we are in “a competition ... for the right to win the hearts, minds, and acquiescence of the population,” on a global scale.30

For all their clear antipathy to the emerging Marxist opponent of their day, Kendall and Pettee saw the enemy not as one to be proven to be in factual error, but rather as one with which we were in a competition “for hearts and minds.” What Kendall, and even more so, Pettee recognized in 1950 is that, in that kind of world, the ultimate measure of the strategic intelligence analyst is utility and result. In Pettee’s words, “Even if strategic intelligence is not exposed to public view, the officials who receive its advice will trust it and respect it in close proportion to its batting average” (Pettee, 91). The measure, in other words, is the efficacy of the function, rather than the purported inherent value of those who perform it.

The professions that Kent held up as models for the intelligence analyst—journalist and professor—are vanishing in part because the needs they once met no longer exist and also because competitors have appeared to meet other of those needs faster, cheaper, and in more accessible ways than can professors and the journalists. Vision 2015 appears to recognize that the profession of intelligence analysis faces the same erosive pressures:

We confront the challenge of acting in an environment that is more time-sensitive and open to the flow of information, in which intelligence sources and analysis compete in a public context established by a global media…. The typical customer in 2015 will be a new generation of government decision-maker, accustomed to instantaneous support, comfortable with technological change, and unfamiliar with intelligence as a privileged source [emphasis added]. Such users will expect intelligence to provide customized, interactive support “on demand,” and will expect to be treated as partner—both a source of input and an ultimate intelligence end user (Vision, 5).

It is not clear whether the profession of strategic intelligence will prove any more resilient against that competition than have the professors and the journalists. George Pettee, however, at least provided a yardstick by which our progress, or lack of it, might be measured. As he warned, “The services which must be rendered to US policy by economic and political intelligence will be judged ultimately against the most drastic standards. If in coming years the consequences of American action correspond to American intentions there will be world peace and prosperity” (Pettee, 42).

Author: Anthony Olcott is an analyst in the DNI Open Source Center.

[Top of page]



1 Sherman Kent, Strategic Intelligence for American World Policy, (Princeton, NJ: Princeton University Press, 1949), 74–5. Hereafter citations from this work are given in the body in parentheses.

2 For data on the decline of these professions see and Frank Donoghue, The Last Professors (New York: Fordham University Press, 2009). What professors teach has changed dramatically, as well, particularly in the liberal arts curriculum that was the basis for much of Kent’s assertions about “the liberal tradition.” For more see Harold M. Greenberg, “Intelligence of the Past; Intelligence of the Future,” in Loch Johnson, ed., Strategic Intelligence, (Westport, CT: Praeger, 2006).

3 John Heidenrich does a cogent job of explaining the differences between tactical and strategic intelligence in “The State of Strategic Intelligence,” Studies in Intelligence 51, No. 2 (2007). He argues that strategic intelligence has essentially been forgotten by the IC.

4 The date shows, for example, on page 172, where Kent speculated on whether or not Truman would try to be nominated for president in 1948.

5 A version is available at:

6 Kent’s father, William, was a three-term congressman, while through his mother he was related to Roger Sherman, one of the five drafters of the Declaration of Independence. His uncle founded the Thacher School, in Ojai, California, and his family donated the land north of San Francisco which became the Muir Woods National Monument.

7 Lippmann also used the word “stereotypes” for those “pictures,” repurposing a word that, until he changed it to what we mean today, had meant stock phrases that occurred constantly in a given kind of text and so did not require that typesetters set them anew each time, but rather just keep them as “stereotypes” ready to be slapped into place.

8 Lippman, chapter 26, part 3; taken from an online version, available at:

9 There are many explanations of this point. A recent and quite readable one is W.R. Connor, “Liberal Arts Educations in the Twenty-First Century,” keynote speech at American Academy for Liberal Education, date not given: [PDF 63.9KB*]

10 Much of what follows is based on Jack Davis, “The Kent-Kendall Debate of 1949,” Studies in Intelligence 35, No. 2 (1991).

11 If the “Dialogues in Americanism” offered in Kendall’s posthumous collection Contra Mundum is indeed the transcript of his comments at a debate with historian James McGregor Burns, Kendall would have been a terrifying person with whom to disagree. See Contra Mundum, 266–81.

12 This does not mean Kendall had no ties with the Intelligence Community. His name appears on a number of reports for the Operations Research Organization (ORO) until at least the mid-1950s. The subjects include an overview of China, a study of psychological warfare as waged by North Korea, and the sequence of questions to be used when giving Korean defectors polygraph tests. See precis of ORO reports at:

13 Willmoore Kendall, “The Function of Intelligence,” World Politics, Vol 1, No. 4 (July 1949): 545. Subsequent citations in text.

14 In a footnote to that remark Kendall contrasts open source collection to covert, arguing that “an intelligence agency in France” could learn far more about the United States from an airmail subscription to the New York Times than it might from four French agents sent to “dispense largesse in Washington.” He also speaks of how analysts might have “continuous and instantaneous access” to the “raw data of the developing situation in the outside world” by using “international telephone,” rather than rely on documents and reports forwarded from overseas, which, as he noted, “as a matter of course are out of date.”

15 See the essay “The Majority Principle and the Scientific Elite,” in Nellie Kendall, ed, Willmoore Kendall Contra Mundum, (University Press of America, 1994), 93–102.

16 Kendall makes a useful distinction between unattainable absolute prediction (“General DeGaulle will come to power this day six months”) and the more desirable contingent prediction. (“The following factors, which can be influenced in such a fashion by action from the outside, will determine whether, and if so, when General DeGaulle will come to power.”) Kendall, 549.

17 Kendall argued that America has two majorities, one which determines who will be president, and another, the composite of many smaller majorities, who make up the majority of elected representatives. The two are almost always—and should be—in opposition. See “The Two Majorities,” Contra Mundum, 202–227.

18 Gregory Treverton, “Risks and Riddles,” Smithsonian, June 2007.

19 Harold P. Ford, “A Tribute to Sherman Kent,” Studies in Intelligence, Fall, 1980.

20 This point is argued at some length in Kendall’s The Basic Symbols of the American Political System, Catholic University of America Press, 1970, reissued 1995.

21 Certainly Pettee was less colorful than either Kent or Kendall. He did appear to have an enthusiastic following at Harvard, his alma mater (PhD,’31), who objected when he was denied tenure, in 1941. See The Crimson, 13 March 1941 (at: Pettee’s obituary suggests a productive but quiet life. See New York Times, 29 November 1989.

22 See and

23 George Pettee, The Future of American Secret Intelligence, Infantry Journal Press, 1946. Hereafter, cites appear in text.

24 Pettee was even more insistent on this point in a lecture he gave to the Industrial College of the Armed Forces in 1950, when he argued: “The Japanese underestimated us and most seriously underestimated our economy. The Germans underestimated our economy. They underestimated the Russian economy, and we know very well, also, that the Germans underestimated their own economy. We underestimated the Germans, the Japanese, and the Russians. The British underestimated the Germans, the Japanese, and the Russians.” See [PDF 3.83MB*]

25 His book, The Process of Revolution (Harper, 1938, reprinted Shenkman 1971) is cited by 46 other books and articles, according to Google Scholar.

26 Porter Sargent, War and Education, Boston, 1943, 249 (at:

27 ICAF, 3 January 1947. At url: [PDF 1.13MB*]

28 Indeed, at the 1950 talk Pettee quipped, “There is another side to the old point that war is too important a matter to be entrusted entirely to soldiers. I think it is possible that, under the conditions we have reached, we can add: Peace is too important a matter to be entrusted entirely to civilians” (5).

29 24 February 1950. At url: [PDF 3.83MB*]

30 Heidenrich, “The State of Strategic Intelligence,” citing David Kilcullen, “Twenty-Eight Articles” in Military Review (May–June 2006): 103–8.

[Top of page]


Note: All links above are external. Please read our External Link Disclaimer

* Adobe® Reader® is needed to view Adobe PDF files. If you don't already have Adobe Reader installed, you may download the current version at (opens in a new window). [external link disclaimer]


All statements of fact, opinion, or analysis expressed in this article are those of the author. Nothing in the article should be construed as asserting or implying US government endorsement of an article’s factual statements and interpretations.

Posted: Aug 24, 2009 10:31 AM
Last Updated: Aug 24, 2009 10:31 AM