The DI 10 Years After Reorganization
Stresses, successes, and the future
The DI 10 Years After Reorganization
David W. Overton
Ten years have passed since the Directorate of Intelligence (DI) was reorganized primarily along regional rather than functional lines. Technically, the reorganization was of the National Foreign Assessment Center (NFAC) into the DI. Because the DI was a name of much longer standing (January 1952-October 1977), I have chosen to simplify this discussion by talking about the old DI and new DI, with October 1981 as the breaking point between the two.
At the time, the objectives of the reorganization were to increase the share of multidisciplinary output and to provide better-tuned support to policymakers. No one would argue that the transformations in process or product have been costless, but most would agree that the outcome has in the main been consistent with what was first intended. The DI product gets generally good grades from our customers, and those who produce it are typically more comfortable today that it is complete than they might have been a decade ago.
Whether the design and process of reorganization of the DI begun in 1981 was carried out in the most effective way turns out today to be largely a moot point, except in management courses, simply because new configurations proved necessary to deal with changing times and demands. In hindsight, it is hard to imagine that the challenging situations we were forced to deal with in the 1980s — the conflict over the Falklands; the war between Iran and Iraq; the US interventions in Grenada and Panama; deepening problems in managing the political economics of Third World debt; and the resource crisis of meeting both defense and consumer demands in the USSR — or those we face in the 1990s — the transformation of socialist systems; the revivification of international institutions; growing gaps between rich and poor and free and oppressed; proliferation of sophisticated weaponry to irresponsible regimes — could be adequately addressed without a reorganization like the one we underwent.
My assessment of the reorganization reflects the peculiar perspectives of my experiences as: a DI research director shortly after the famous speech by then Deputy Director for Intelligence (DDI) Gates on the inadequacy of our analysis and product; a senior economist in the DI as economic issues became increasingly important as a share of our work; a member of the Product Evaluation Staff (PES) from 1983 to 1985 and from 1989 to the present; and the Deputy Director for Curriculum of the Office of Training and Education (OTE), a position that allowed me to see some of the persistent management and analytic training problems that attended the new configuration of the DI. All these things have helped make me aware of some important aspects of the emergence of a new DI culture over the 1980s and 1990s, but they may also have made me too attached to what the DI has achieved to be wholly objective about the flaws.
1981: Inventing a New DI
The issue of DI reorganization had been visited with and without consultants several times before enough momentum built for action. Most of the dinosaurs of DI management were resistant to the idea of reconfiguration along geographic lines.
A study on the reorganization produced by OTE traces the idea of geographic organization of the DI back to at least 1973 and refers to serious examinations in 1976 of the possibility by DDI Sayre Stevens and by contractor Arthur D. Little Associates. The words of the study have a certain haunting quality to those who were part of the old DI:
Several of the studies submitted to the DDI [by his subordinates] identified the same problem as the central flaw with the functional model: the Directorate did not appear capable of interdisciplinary research and analysis, at least not consistently, at a time when the foreign policy questions were becoming more complex, and consumers' questions more sophisticated. Further, some observers noted, many DI analysts were concerned that the demands of current intelligence production — with its fast pace and short time horizon — were being met at the expense of long-term research.
Besides the bureaucratic inertia and sense of fiefdoms common to all organizations, I believe there was another basic impediment to reorganization along geographic lines. The products of the DI were not as diverse as they are today, but the differences in product emphasis in particular offices were profound. For example, in the Office of Economic Research (OER), where I was a division chief at the time of reorganization, current intelligence largely meant production for the International Economic and Energy Weekly. We produced only sporadically for the daily publications, which, in turn, were the bread and butter of many of the political analysts. Unlike the Office of Political Analysis, we did not have monthly publications. On the other hand, OER put out a considerable amount of both hardcopy and typescripts tailored to the particular needs of economic policymakers. Though changes were beginning to take place in political and sociological analysis propelling them more toward longer-term research, the emphasis on research was much stronger in the economic, engineering, military, and scientific areas.
The product differences were complemented by differences in the ways in which offices related to their customers. Then as now, an unfortunately large number of customers saw themselves as their own best political analysts, while those who were interested in economic topics were fairly receptive to new ideas or the results of rigorous research.
The military analysts — again, then as now — had a good deal of competition from other parts of the Intelligence Community. One net effect of all this was that the economists tended to interact more closely and frequently with the policymaker, and they were rarely concerned that their results were being "politicized," but the political analysts had to be more skeptical of the ends to which their work would be put.
Despite the obstacles to breaking down functional organization, there was some slight progress in that direction with the establishment of several centers before the reorganization. These included a Cuba Analytic Center, an International Narcotics and Terrorism Center, a Southwest Asia Analytic Center, and a Center for the Analysis of Personality and Political Behavior. Moreover, about a year before the reorganization, multidisciplinary task forces were put together to track the Polish debt crisis and developments in the USSR.
The key turning point, however, was DCI Casey's increasing concern for better service to the new Reagan administration policymakers, especially on the USSR, which led to the appointment of John McMahon as the Director of the National Foreign Assessment Center (NFAC), in April 1981, with the mission of improving the DI's analytic product. Through he considered other options, McMahon gravitated back to the geographic model. In July 1981 he told a meeting of his office directors that he was going to reorganize NFAC — to once again be known as the DI — along geographic lines no later than 1 October 1981.1 To some, it seemed that this would require a massive overhaul, might not eventuate in anything particularly better, and would take more personnel resources than the DI could possibly muster.
1982-1984: Getting Beyond Words
If McMahon's determination to get on with reorganization was necessary to budge the old cultures over, the hump of inertia and complacency, the insights of DDI Gates on the needs of the customers were critical to shaping a new product. In fact, since he took over as DDI only a few months after the reorganization and quickly addressed nearly the whole officer corps of the DI on his new agenda, Gates can lay particular claim to being the father of the new product.
There were, and are, two schools of thought about whether shaping of the product went too far. Those few on one extreme of the spectrum who believe politicization — or more politicization — was part of the outcome tend to forget how often pre-reorganization product was conceived at Headquarters without benefit of knowing the policymakers' activities or needs, packaged according to the convenience of the producer, and left on doorsteps downtown in the orphan basket it deserved. All too often, I saw a sort of arrogance in what the DI produced and how it was delivered that was first cousin to the hauteur of Detroit's automakers when they argued that they knew best what kinds of cars American consumers needed.
And yet, in fairness, even those most sympathetic to the new approaches in DI production and marketing suffered some pain in trying to make them work. There were few good models from outside the Agency to tap into in order to get a sense of how effective multidisciplinary analysis took place. The academic world had for years been talking about — and raising funds on the strength of — multidisciplinary projects, but most of the ones that were successful were in narrow areas of the physical sciences. As a result, progress in this area for the DI was to occur largely through trial and error based on what we and the customer felt about the end-product. And this meant the need for closer relations with the customers, a move that the most conservative DI analysts saw as "caving in" to the policymakers' prejudices.
Even for those who were attuned to the fine line between providing good policy support and kowtowing, simply learning the new writing styles and terminologies was a tedious chore. To distinguish more clearly between what we knew and what we thought, Gates pushed the analysts to use much more sourcing and statements of methodology. As an office research director, I was constantly reminding analysts that appendices and a few well-crafted footnotes on sources or methods might relieve them of the "obligation" of justifying every single sentence or paragraph with a source comment. Developing the combination of an engaging writing style and a fairly rigorous method of identifying sources and methods did not come quickly or easily for anyone.
Another area that needed serious development was our knowledge and skills in identifying outside experts who could help us think through issues and undertake research for us. In the 19 years I had been in the DI before the reorganization, I had only interacted with 35 to 40 people outside the US Government who worked on topics of interest to me. For most DI analysts, the network of outside experts was even thinner. In early 1982, when the DDI asked that the offices write essays on key outsiders working on relevant fields and list names of those with whom we dealt, many in the DI began to fear that we might actually have to deal with the world beyond us.2
DI economists had been among the more, outwardly oriented analysts before the reorganization. It was not uncommon for the old Office of Economic Research to bring in academics a few times a year to comment on our product or help us work on complex issues. And the DI economists also made regular use of the former Domestic Contacts Division and other channels to reach out to specialists in the academic, business, and banking worlds. Still, the rate of outside contacts was nowhere near as frequent then for economists as it is today. In the last two years, for example, I have been invited to DI-sponsored conferences with outsiders on economic topics at the rate of about one a month, and the number of outsiders present at any one event has run between five and 25.
Another wrenching experience for those who had lived under the old DI was the rigor of the construction of the program of analysis in the new culture. I recall sitting through a program review in the earlier days in which a senior analyst was confident that, upon some minutes of reflection, he had a good concept for a paper about Soviet educational aid in the Third World because he had thought as far as the title: The Little Red Schoolhouse. That approach had no place after the reorganization, when building an annual program became an event of several months and engaged all layers of supervision in the DI.
Before the DI reorganization, a reasonable annual rate of production for an experienced analyst had been seen as a combination of one or two longer hardcopy publications, perhaps a half-dozen typescript memorandums, and a solid file of current intelligence dotted across the year. In fact, the rate of output of products slowed in early 1982 as managers and analysts tried to come to grips with the new ground rules for formats and the new approach to programming. By the time the reorganization took hold, expected and actual output rates were considerably higher than before.
Perhaps the worst new development of the new programming/production environment from the perspective of those who had been fully comfortable with the old culture was the notion of accountability. On 12 February 1982 the DDI sent a memo to all office directors and staff chiefs that was the charter of PES. The first sentences of that memo included these words: "I attach the highest importance to the quality of our research and production effort, and have established the Product Evaluation Staff to advise and work with you and make recommendations to me concerning the overall DI research program and how to improve it. ... It [PES] will have two major responsibilities — research planning and evaluation of our product." Thus, imbedded in the new research planning process was the responsibility to count beans and track results.
The Path to Today
The range of subjects the DI subsequently had to cover was more extensive than any of us imagined in the early 1980s, and the ability to do this work was shaped by the new configurations and commitments of the first days of the reorganization. Some additional changes in organization and format were in store, but they usually blended well with what the chefs had put in the pot in 1982.
Areas of conspicuous new strength abounded:
- The regional offices working on the Third World and the Office of Global Issues (later to become the Office of Resources, Trade, and Technology) were able to move beyond the baseline of our earlier analysis of LDC debt to say more trenchant things about how well LDC governments would manage their problems and how world financial markets would react to them.
- The regional offices working on the G-7 countries were able to extend political/economic analysis forward in ways that made our output on the EC and trade/financial relations with East Asia more sophisticated.
- Growing steadily from simple biographic sketches in the early 1980s, our work in the area of leadership analysis progressed to more comprehensive assessments of individuals' and organizations' strengths and weaknesses.
- From a limited base for military analysis on the Third World, the DI developed enough capability to produce significant intelligence on the Falklands crisis, the war in Chad, civil strife in Angola, the Iran-Iraq War, the continuing bloodbath in Ethiopia, the invasions of Grenada and Panama, and the Persian Gulf crisis.
- Essentially from scratch, the DI built an ability to do military-industrial analysis on major countries other than the communist ones.
- The DI developed from the bottom up a new expertise on advanced technologies that had both national security and major commercial applications, and it advised policymakers on critical new developments in this field.
Beginning essentially in the mid-1980s, we enhanced our ability to analyze in the areas of counternarcotics, counterterrorism, and counterintelligence through development of centers that drew from both the DI and the Directorate of Operations (DO) for their personnel. This approach has meant better exchanges of information between collectors and analysts, although sometimes at the cost of capturing analytic time for operational support. Generally, the center experiments have worked out well in helping to build or increase specialized knowledge. There is, therefore, some basis for optimism that the new DCI Nonproliferation Center will also expand the range and depth of what we do in that field.
The Age of ADP
While this and much more was afoot, the DI passed through one of the most encompassing changes in its history. More than any other time, the 1980s were the watershed of broad acceptance and use of automated data processing. While we think of SAFE and other sophisticated changes in the use of the mainframe computers as signs of this change, two of the most important new departures were of the garden variety. One was electronic mail. It broke with the traditional ways of doing work because the number of potential discussants moved from the small numbers of one's acquaintances and immediate colleagues to all those you might simply hear of or choose to reach through an electronic conference or alias. While undertaking a study on ADP in the DI in the mid-1980s, I was surprised to find, for example, that the unclassified version of a DI paper on the Soviet computer industry had been shared via INTERNET with a former DI co-op knowledgeable about the topic who had returned to his campus. The author maintained that he had gotten some useful remarks from the co-op as well as from those here at Headquarters with whom he shared it electronically.
The second change with enormous implications was the Agency's acquisition of commercial databases that gave us electronic access to all sorts of unclassified material. Thus, when PES did a study of DI current intelligence production in the mid-1980s, it was able to call up quickly what the press had been reporting on the same issues through LEXIX/NEXIS. Similarly, the typical DI economist vaulted from having to read through hundreds of pages a quarter of The Journal of Economic Literature to find what he or she needed to getting a librarian to do a run in DIALOG.
One of the key concerns at the time of reorganization had been that the geographic model would lead to degradation of specialized products. On balance, that does not appear to have been the case. In various reviews, PES has found instances in which the DI product would have been stronger if the junior analysts had someone from their own discipline overseeing their work. Weighing heavily against this kind of defect, however, is the fact that the range of specialized products for the DI has increased steadily over the past 10 years. One has only to think of what we produce today on economic/technology issues, narcotics production and finance, terrorist networks, weapons proliferation, and counterintelligence concerns compared to what we produced in 1980 to realize that it is possible to have both more multidisciplinary work and more specialized analysis. One small clue to that increasing diversity is the fact that the number of functional labels by which PES categorizes DI product in its database is about twice what it was in the early 1980s.
This is not to say that we have avoided all the problems that concerned us when the DI took on a new look in 1981. I recently asked a former Director of OER to look over a substantial number of articles from the International Economic and Energy Review in order to compare it to earlier editions. He gave it high marks for thoroughness of research, writing style, and quality of the graphics and packaging. On the other hand, he said it struck him as less premonitory and less willing to take risks than what he remembered of the 1970s. I believe he is right, and I also believe that some of the conservatism of the new era is driven by management that is generally less comfortable with the concepts that are familiar to economists.
Costs and Benefits
Human Costs. The DI has paid high human costs in moving from where it was in 1981 to where it is today. Some analysts, particularly in the regional offices, who could not adjust to multidisciplinary work either left the DI or found that the numbers of jobs for which they were suited and their career advancement opportunities dwindled rapidly.3 And those who had counted on their supervisors and their professional mentors being one and the same now found career development more complicated as economists, say, reported to military analysts who in turn reported to political analysts.
Even the definition of appropriate work activities became more muddled. Military analysts, who often had proportionally greater need for extended formal training after they entered on duty, were now subject to criticism from political scientist managers, much of whose training came on the job. Economists, who typically make extensive use of private sector experts and source materials, had to defend the outside meetings and conferences they attended and their recourse to unclassified data when we already had "the embassy stuff." Both economic and military analysts, who rely heavily on automated, numerical databases that they must develop and adapt, were now open to the charge that they spent too much time "playing with the numbers."
Certainly in the case of the economists, the costs of career development for the DI became increasingly explicit over the 1980s. What formerly took place as an exchange between a branch chief and an analyst now became sets of courses and workshops and career counseling across organization boundaries. It became increasingly difficult in a DI office for a new economist to know what training was important to developing the skills most needed on the job. And the rewards that had accrued to same-discipline managers who tutored and developed people proved harder to specify and implement as bureaucratic boundaries intervened.
A partial correction to these problems was the outcome of two PES studies undertaken in 1989 and 1990. One study looked at the product of and conditions for economic analysis in the DI, and the other was a similar study on military analysis. One outcome of the studies was the appointment of senior military and economic referents to help on issues of professional communication and career development for these two groups of analysts throughout the DI. So far, this approach seems to have achieved a better transmission of relevant information than had been true before.
Resource Costs. In some areas, it is hard to sort out how much of the increased resource commitments for analysis related to new demands rather than the increased expense of doing business in the new configuration. In others it is not. Either way, there were some resource tendencies that we need to be aware of as we look forward.
In thinking about resource costs, it is important to remember that DCI Casey increased the size of the Agency and the territory in which it operated. Thus, it is equally hard to distinguish accurately or in any net sense between new things we did because we thought they would be helpful and new things we did because the customer demanded them. (Showing that we can do one new thing on our own often stimulates outside interest in having us do several others that are related.)
In addition to greater training and new ADP features, four areas come to mind as increased costs of doing business in the new DI: more people; more recruiting; more space; and more time invested in meetings. On people, once you undertake to do multidisciplinary work on any country, the requirement for specialists from various disciplines rises. In the economic area, for example, there are today twice as many economists working on sub-Saharan Africa as there were before 1981, and they are still stretched thin. The difference is that, before the reorganization, OER moved people around much more from country assignment to country assignment as particular narrowly focused economic issues arose.
Similarly, military analysts did not have to have a deep knowledge of the countries they worked on. In a brief stint as a division chief in late 1981, I inherited from the old Office of Strategic Research the lion's share of military analysts doing research on the Third World. They constituted a small branch. Considering their numbers, they would have drowned in a task like tracking the Persian Gulf crisis.
Recruiting is, of course, the other face of having more people. Recruiting activity blossomed in the 1980s for two reasons. The accounts we covered grew in number and complexity, and this required more people in the pipeline. As the population of people outside the Agency who fit well the changing nature of the DI jobs tended to decrease, recruiting became somewhat more difficult and increasingly demanded a fairly heavy effort of DI people in concert with Office of Personnel recruiters. As we face reductions over the next several years, the need for large numbers of applicants will decline. But the sustained high effort to find the right people will not.
There is also a strong correlation between people and space. If we had needed a clear sign that space would be a problem, it came swiftly. The departure of the former Office of Soviet Analysis from the Headquarters compound was the first salvo, but others were to come as we searched for places to house the new centers in the mid-1980s. The pressure of space issues was heavily driven by increased automation of information processing, an enhanced capability which was critical to dealing with more and more complex policymakers' questions.
I would argue that there is also a strong connection between people and space issues and the increased numbers of large meetings we conduct. I recall the time before the DI reorganization when, if I wanted to get away from phones and noise to review a paper, I had a better than even chance of finding OER's sumptuous conference room unoccupied. Today a conference room readily accessible on short notice is like a fresh supply of toilet paper at state stores in Moscow: the word travels fast.
What has happened? Communication among greater numbers of people on increasingly complex topics and within more offices/centers than in the pre-reorganization DI is a part of the answer. The other part is that the daily cycle of production meetings that was central to the current intelligence-dominated political analysts is now de rigeur for the whole DI.
People and Product Benefits. For anyone who has participated in studies of the DI production effort, a message that rings loud and clear is that most DI analysts find working in this organization intellectually stimulating.4 This is not necessarily individual rationalization of why they have stayed on, for younger analysts and student interns also give the work high marks for chances to learn and grow. Moreover, I have listened to visiting academics say over and over again how impressed they are to discover how challenging and exciting the DI's work is.
Because I have heard the same academics say that we are light years ahead of them in multidisciplinary work in the social sciences, I tend to believe that it is just this object of the DI reorganization that has done a lot to make the work more fun and rewarding. Indeed, the distinction between what we do here and what we might be doing if we applied our talents in academe was well captured in a comparison made by an Ivy League college professor who visited the Office of Leadership Analysis for discussions. After complimenting our analysts on their collegiality within and across offices, he described the atmosphere of the typical university campus.
We're high-tech cave dwellers. I don't know what's going on in the cave next to mine, and the guy in it doesn't know what I'm doing.
The cross-fertilization among disciplines has served us especially well in the newer, expanding areas of our work, such as counterterrorism or environmental aspects of foreign policy. But it has clearly also helped in increasing the sophistication of our traditional country and regional analysis. It is not hard today to find multiple instances in any year of DI papers that either would not have been possible or would have been very difficult to produce before the reorganization.
Both in PES and with fellow economists over the past two years, I have had the challenge of worrying about the issues we will face in the first half of the 1990s. What is striking to anyone who undertakes to look back and forward at this watershed is how dramatic the potential for change is in the next five or so years Marxism-Leninism bids fair to disappear from most parts of the world, except, of course, from college campuses. Tyrants who appeared capable of dominating many of the world's poorer countries are cringing in front of a wave of support for democracy. The death of communism and the rise of democracy will, as comparable experience in various parts of the Third World has shown, be attended by serious pains and instability.
If the economic and political revolutions in countries around the world are obvious new developments, subtler but equally important ones loom.
- The institutional base that permitted the unprecedented postwar expansion of global trade and finance is undergoing rapid changes, the end of which we can only dimly predict.
- Dangerous, unsophisticated people are finding it easier to gain access to dangerous, sophisticated weapons.
- The ferment in both regional groupings and technological development may profoundly change both the way economic relations are conducted and the power positions of the actors.
Comparative Advantage in the 1990s
In this changing environment, we would be well advised to focus our thoughts on what constitutes our comparative advantage as a source of information and insights to the policymaker. I am struck by four advantages we enjoy that we have to press to their fullest. I list them in what I judge to be their decreasing order of importance:
- The ability to acquire, process, and manipulate large quantities of data/information.
- The time to undertake serious, deep, premonitory intelligence.
- Access to sensitive information not generally available.
- A reputation for producing readable, attractive products.
Information handling. The clear message of the 1980s and the picture we face today is that the availability of unclassified data and analysis pertinent to policymakers will continue to increase. Those who are simply trying to keep up with what is becoming available from the former communist countries have evidence enough that this is going to be a challenge. Because this information is as likely to be available to CNN, think tanks, and universities as to us, we are all in a position to exploit it. Virtually no one else, however, has the computer capacity, experience, or broad access to the full range of commercial databases that we do, and we can make this work to our advantage in assuring that policymakers come to us to collate and analyze the raw material.
We have this same capability with respect to classified data and mixed databases. The groans about the ups and downs of SAFE in the early 1980s have given way to a much clearer recognition that we served ourselves well in this investment.
The clear strength we have in developing analytical software is sometimes overlooked. My experience, for example, is that we have yet to demonstrate TRADAR, the automated system for collating and analyzing trade data to outsiders, that they fail to be impressed with it.
Time. In the early 1980s, Bob Gates spoke to large groups of university placement directors on several occasions. When he did, he would astonish them by saying that we were among the few US agencies still undertaking long-term forecasting in the realms of foreign policy and national security. He was right then, and the story is simply more accurate today.
In addition to the recognition that we have a forecasting capability, we are credited with having time to think seriously about difficult topics generally. A senior US Trade Representative official told me some months ago that the institutional products we were doing in the realm of leadership analysis could only be done in real time at CIA.
We have to remember, however, that when we get locked into sitreps for sitreps sake or produce a flyweight typescript memorandum to pad an analyst's production file, we are frittering away this advantage.
Access to sensitive information. This was our primary comparative advantage when I joined the Agency some 30 years ago. It is still important, and it is often dramatically so in key moments of international tension or negotiation.
But times have changed. Much more is openly available than 30 — or even five — years ago. Some of the issues that demanded clandestine collection have been seriously reduced in scale or disappeared entirely.
This imposes a greater burden on both analysts and collectors to pinpoint what they really need from clandestine collection and to know when new or expanded open sources become available. I believe we are pointing in the right direction in this effort, but we still have a long way to go. A recent high-quality HUMINT collection plan for economics tells me we can get to where we need to be. The repeated silliness in the press about our engaging in industrial espionage says, however, that some do not believe we are there yet.
Readable, attractive products. Whenever I think about the generally high quality of the layouts and graphics in our formal publication, I recall a State Department classmate of mine in State's Senior Seminar showing me some maps that he particularly liked. "Our mapmakers really do attractive work, don't they?," he asked. I did not have the heart to point out the CIA product number in the lower right corner.
But this is no time to get cocky. Recently, the typescript memorandum has tended to displace our more formal publications. Only if you have the perspective of seeing all of what the DI puts out, as we in PES do, do you realize how easy it is for quality standards to appear to slip if someone is not watching out for formats as well as substance. In my estimation, the DDI has been right on the money in trying to assure that there are some standards for appearance in our informal publications as we drift down the river of desktop publishing.
The Task Force on Intelligence Production has made some wise and provocative recommendations in its report. I do not agree with all of them, including in particular the theme of pushing more of the responsibility for product review down to the offices. That said, I was pleased to see that they recognized the need to standardize formats for informal product.
Our maps, charts, and products that more than met the consumers' needs for clarity and convenience have been praised for years in the US Government. On the other hand, despite what consumers say about being willing to take our product in any format or condition, I can testify to the fact that a few slips in this area bring out the boo-bears and cries of sloppy work.
Applying Lessons from the Past
As tighter budgets press on us in the coming years, we have to remember the most important lesson we have learned from the DI reorganization. Nothing — larger sums of money, additional space, greater favorable recognition, less criticism from the Congress or the press — is, by itself, as important to us as the mental health and welfare of the people who do our work. They have to be able to develop knowledge and skills we can only dimly perceive today. And they have to be able to attack, regroup, and reattack issues that will be constantly changing form and reappearing in new settings.
This central fact of the importance of the DI's people calls for more training, diverse job opportunities, and sensible rewards for good service. The people of the DI need to know that they can grow, develop, and apply newfound skills. I am confident that enhanced language training, the Analyst Overseas Program, opportunities for external training, rotations to policy organizations, and more chances to deal with the private sector are taking us in the right direction.
I am less confident that we understand the importance of remaining flexible about how we do our work. The events in the former communist countries have precipitated various reorganizations into new groups, divisions, branches, and working groups. Some of the change has been strategic and some knee-jerk. It is too soon to say whether it has enhanced or diminished our flexibility in putting out the product.
But the message of the DI reorganization of 1981 has less to do with wiring diagrams and new names than it does with the flexibility of people themselves. And foremost among those who have to be flexible are our managers. Throughout the 1980s, the participation of senior management in the production and review processes provided an important opportunity for analysts to speak up the line of command. One hopes that the push for decentralization of review does not take a toll on this kind of communication.
To move forward smoothly into the next century, DI managers have to be willing to solicit new ideas from those they supervise, who, after all, know best where the shoe pinches. The DCI has taken an important first step in this direction in convening task forces that were charged to challenge our working assumptions. The DI should promptly exploit every good idea the task forces advance.
1. In the end, the substantive area of the DI least affected by the reorganization was scientific and engineering research and analysis. Various earlier changes had resulted tin the collocation of the DI's scientists and engineers, and their leadership was persuasive in arguing that the maintenance of professional skills required that they continue to sit together. Over the period 1981-92, the most striking changes in this area had to do with additions to OSWR's responsibilities in technology transfer and proliferation issues.
2. A notable exception to this inwardness was Soviet military and economic analysis, where we had long had relations with the academic world. In the context of the recent charges that we failed to take advantage of superior wisdom outside the Agency on the Soviet economy, this fact of a long relationship makes the criticism especially ironic.
3. Until comparatively recent times, this sort of retooling was less of a problem in Soviet analysis, where the need for specialized knowledge on comparatively narrow aspects of the military, internal politics, and the economy preserved the demand for analysts from the "old school." For functional accounts that survived the 1981 reorganization, there was generally less of a problem of adaptation, but, even here, analysis of issues like trade or financial policies, for example, led rapidly to the need for a better understanding of country politics.
4. All good generals know that an army where no one grumbles is in serious trouble. The complaint from the minority in the DI is that the individual accounts are becoming "too narrow." What this seems most often to mean is that those concerned are not getting the space or frequency of publication they feel they deserve, especially in current intelligence. This, the other face of sustaining multidisciplinary capabilities on a wide range of countries, is a real balancing and adapting challenge for DI managers.