75 – A Strategy of Decision: Policy Evaluation as a Social Process (Braybrooke and Lindblom, 1963)

Reference:

Braybrooke, David and Charles Lindblom
“A Strategy of Decision: Policy Evaluation as a Social Process”
The Free Press, Macmillan, New York, 1963
(Ch.4-6 and Chapters 9-10)

Topic: The theory of “disjointed incrementalism” in policy decisions.

Summary and citations:

Chapters 4-5-6: Lindlom
• “disjointed incrementalism”
• “Decisions effecting small changes:… When is a small change? …“small” depends on the value attached to it…. Distinction between “small” and “large” change is the difference, as it is sometimes put, between structural changes and changes within a given structure…”small” change is a change in a relatively unimportant variable or relatively unimportant change in an important variable. … “Increment of change”…. To be a large or non-incremental change… a small or incremental change…. The difference is one of degree. . – p62-64
• Incremental changes: 1) social change that largely repeats; … the more repetitive the change ,…the more it is incremental; 2) non-repetitive change – p65
• Incremental politics: … an indefinite sequence of policy moves (p70)… moving away from known social ills rather than as moving toward a known and relatively stable goal (p70). … In short, the incremental character of this political pattern is central and fundamental, even if it does not wholly characterize it. (p73)… In incremental politics, political parties and leaders compete for votes by agreeing on fundamentals and offering only incrementally different policies in each policy area in which they wish to compete. (p73)… In addition, policy-making proceeds through a sequence of approximations (p73)…. In short, incremental policies follow one upon the other in the solution to a given problem. (p73)… Just how far the practice of incremental politics permeates the activity of or government can be illustrated even by decisions that, on superficial inspection, appear to be non-incremental *** (p75) i.e.: decision to convert from peacetime to wartime…. Waves of reorganization also characterize the history of other wartime agencies…. Gulick:”sequence of evolution” or “rhythm of growth and adaptation”. (p77)… The incremental character of political policy-making is often disguised,[…] by much talk of plans and planning. (p77)
• p.78: Types of decision-making
• Quadrant 1: high understanding and large change: Revolutionary and utopian decision-making. Analytical method: none.
• Quadrant 2: high understanding and incremental change: Some administrative and “technical” decision-making. Analytical method: Synoptic
• Quadrant 3: low understanding and incremental change: Incremental politics. Analytical method: disjointed incrementalism (among others).
• Quadrant 4: low understanding and large change: Wars, revolutions, crises, and grand opportunities. Analytical method: not formalized or well-understood.
• “As to what the adaptations are, they become familiar as soon as they are named. Our purpose in collecting them together under the label of disjointed incrementalism is only, as we have said before, to show that, taken together as a mutually reinforcing set of adaptations, they constitute a systematic and defensible strategy.” – p82
• The strategy outlined
1. Margin-dependent choice: “… they derive information about alternatives from historical experience, from contemporary experience in other societies or locations, and from imagination stimulated by experience” –p84 – “Incremental evaluation is clearly quite different from the construction of a rational-deductive system.” –p86. “The resolution of a conflict over two values is not expressed by a principle, as in the rational-deductive method, nor by a priorities list –nor is it implicitly embodied in a ranking of social states. IT can best be expressed by stating how much of one value is worth sacrificing, at the margin reached in a given situation, to achieve an increment of another.” P88
2. Restricted variety of policy alternatives considered: “…incremental alternatives… while…an analyst is often without adequate information, theory, or any other organized way if dealing systematically with non-incremental alternatives” – p89
3. Restricted number of consequences considered for any given policy: “… do neglect the unimportant consequences of policies, but among those they concede to be important, they often rule out of bounds the uninteresting (to them), the remote, the imponderable, the intangible, and the poorly understood, no matter how important.” P 90. “One striking feature of disjointed incrementalism: that what is omitted is often quite as important as what is considered” –p93
4. Adjustment of Objectives to policies: “although there is a fundamental sense in which ends govern means, there is an equally fundamental sense in which the proximate ends of public policy are governed by means” –p93 “…characteristics: a) some of the objectives became relevant only after the means was decided upon, and b) the group excluded certain objectives relevant to foreign aid policy only because they were irrelevant to the one means decided upon.” –p97. “The objectives are discarded, not the policy” –p97.
5. Reconstructive treatment of data
6. Serial analysis and evaluation: “…policy-making under the strategy proceeds through long chains of policy steps. This serial procedure is an important feature of the strategy in its own right” – p99. “They do not organize to solve a problem and then disband. Instead, they take it for granted that their problems will rarely be “solved” but only alleviated.” –p100
7. Remedial Orientation of Analysis and evaluation: “Since policy analysis is incremental, exploratory, serial, and marked by adjustment of ends to means, it is to be expected that stable long-term aspirations will not appear as dominant critical values in the eyes of the analyst.” -102. “… Encourage the analyst to identify situations or ills from which to move away rather than goals toward which to move” –p102. “…They are less concerned with pursuing a better world than with avoiding a worse” –p104
8. Social fragmentation of analysis and evaluation: “disjointed: …various aspects of public policy and even various aspects of any one problem or problem area are analyzed at various points, with no apparent co-ordination and without the articulation of parts that ideally characterizes subdivision of topic in synoptic problem solving” –p105. “…advantage of preserving a rich variety of impressions and insights that are liable to be “co-ordinated”…There are circumstances to which no one plan is especially suited.”- p106
• “It is easy to misconceive the strategy as having an indelibly conservative color” –p106. “Strategy specifies nothing about the speed with which change is to be carried on” –p109
• “It is also unquestionable that the strategy accomplishes an enormous simplification of the analytical task, compared to the rational-deductive ideal, the method of the welfare function, or the general synoptic approach” –p111
• “With respect to evaluation and analysis, it was shown to be incremental, restricted (both with respect to variety alternatives and to variety of consequences for each alternative), means-oriented (in a number of complex senses), reconstructive, serial, remedial, and fragmented.
• The synoptic ideal, does not adapt in any specific way to (p113):
• Intellectual capacities, information, and costs of analysis
• Adaptation through incremental and limited analysis: simplification through omission (reducing the demand for intellectual capacity and information); the role of theory (“analyst… can dispense with comprehensive theories..because he limits his analysis to changes in which the variables involved are fewer and less consequential…He does not require full information” –p118)
• Adaptation through means-orientation and fragmentation: “close reciprocity between means and ends…. Unnecessary” p119 – “Values…are seriously reconsidered with each new policy choice because values and policies are chosen simultaneously” p120
• Adaptation trough serial, remedial, and reconstructive features: “…strategy does not try to achieve an optimum or maximum… he does not try to “solve” his problem.. At the most, he attempts comprehension of a suitable remedial next step in a series.” p121 – “He disregards many other possible moves because they are too costly (in time, energy, or money) to examine.” –p123
• The problem of neglected consequences: “…harmful failures in analysis are “corrected” despite the relative narrowness of analysis by any one analyst by any one analyst at any one time” – p124
• Further adaptation through fragmentation: “Organizations, as everyone knows, are often dedicated to interests or values that the members wish to protect. What one center ignores, another does not. Values neglected at one point in the policy-making process are central to the analysis carried on at another point” – p128
• Inadequately formulated values:
o “Strategy makes the most of existing constraints on policy to reduce the areas on which social disagreement is possible” –p134. “because the strategy is remedial and serial, it takes advantage of the frequency with which we find ourselves agreeing on what we are against” –p134. “we agree in large part on a “next step”” –p134. “because the strategy is reconstructive, we add that it discourages intransigence in the defence of values” – p135
o “We are saying that where analysis and policy-making are serial, remedial, and fragmented, political processes can achieve consideration of a wider variety of values than can possibly be grasped and attended to by any one analyst or policy-maker. It is the accomplishment at the political level that makes agreement among analysts less necessary” p138

Chapters 9-10: Braybrooke

o “…on the one hand, the collapse of utilitarism; on the other, the rise of emotivism” p204
o “Utilitarism… insists more strongly …on forcing moral judgments to the test of facts” p206; “…explicate the criteria for intelligible moral judgments” p 206; “…several features of the strategy of disjointed incrementalism make it easy to associate the strategy with utilitarism” p206
o Emotive theory of ethics: “The thesis is that moral locutions are intelligibly used in accordance with the present conventions of ordinary language, if they are used only emotively or imperatively –that is, to do no more than express emotions and convey prescriptions” p210
o Enlightened peremptory objections: “The judgments in question are all peremptory ones” –p213; “…peremptory rules such as these: that one should keep promises, even when the consequences of doing so in a particular instance would be less happy than the consequences of breaking them;…Subscriptions to such rules almost suffices by itself to qualify a person as morally enlightened. A literal application of utilitarianism might nevertheless conflict with them. Some of them are at the root of long-standing objections to utilitarism.” –p213
o The felicific calculus, it will recalled, presumes that the utility of an actions (or policy) can be reduced to the “value of a lot of pleasure or pain” –p214
o “it is supposed that, according to utilitarism, the average or total happiness of the group is to be increased, regardless of how many or how few people are made happy, some perhaps at the expense of making others unhappy and regardless even the continued presence or absence of certain members.” –p216
o Overcoming peremptory objections in alliance with strategy: “Utilitarism can avoid these difficulties and a number of others by abandoning the felicific calculus and allying itself, at least for the present, with the strategy of disjointed incrementalism.” –p217; “the census notion already incorporates in practice some provisions for fair and equal treatment that the felicific calculus, as it is usually interpreted, lacks.” P217
o “…strategy helps to save utilitarism from many conflicts with peremptory rules by insisting on a remedial orientation” –p219
o “When it is allied with the strategy of disjointed incrementalism, utilitarism thus anticipates some peremptory moral judgments that have in the past been made objections to it.” –p219
o “Utilitarism, utilitarism with the strategy better than utilitarism with the calculus thus supplies the demand constituted by the principles of justice” – p222
o “The strategy of disjointed incrementalism is, in ways both morally convincing and philosophically illuminating.” –p223
o The questionable feasibility of the calculus:
o “Possible or not, however, an entirely convincing design for the felicific calculus has not yet emerged from these efforts” p 227.
o “subjective utility”: one person’s satisfaction with a certain policy as against another person’s dissatisfaction”p229
o “the strategy responds to this need in two ways: by redesigning policies so that, as far as possible, they satisfy multiple demands, and by supersession. Supersession in effect establishes an order of priority among different values.” P229
o The problem of consequences:
o “”utilitarism requires that persons evaluating policies consider the consequences of those policies as they affect happiness.” –p230; “there are two things that have to be done in answer to the objection about consequences. First, it must be shown that utilitarism can be associated with reasonably specific ways of selecting the consequences that are to be considered in evaluating actions and policies. Second, it must be shown that utilitarism deals as prudently as possible with the inevitable fact that some important consequences may be overlooked (partly because not all the consequences will have been formulated).” –p233
o “use of strategy does not imply that the analyst ignores information… only that he defines his task so as to leave out information that will have to be dealt with in the course of other tasks.” –p234.
o “The strategy of disjointed incrementalism maximizes the chance, and thus it mitigates fears about undetected consequences. As we have said, it encourages extensive and varied discussion yet provides means of giving the discussion focus” p.238
o “…They are safer because they leave so much more room, in the way of other factors that can be adjusted, for remedying unforeseen consequences.” P239
o Shift or reference groups: “The problem in question is one of identifying suitable reference groups, the consequences for which are to determine the rankings of policies”. p240; “… in the absence of challenges, the references groups should be the ones used in the past” –p242
o Social pluralism and use of information: “….the shift of reference groups would take place between one group and a larger group that wholly includes it, ultimately reaching a group as comprehensive as “society as a whole”” –p243

Personal comments, interesting issues and findings:
• Incrementalism versus synoptic: reminds me of my engineering studies. It is the same difference between, algebra and numeric calculus. Algebra gives you an exact result, through the application of a resolution process and numeric calculus gives an approximate result, following iterative calculations. Numeric calculus can be applied in very complex mathematical problems, because it pretends to trend to the right solution in progressive steps. Algebra finds just the right result or the process is stuck (for example, due to mistakes in concepts, in the applications of rules or due to complex and time-consuming process).
• The comparison of incrementalism versus synoptic also reminds me of the extreme positions of “the paralysis by analysis” and the “extinction by instinct” (see article of Ann Langley about these two extreme strategic approaches). Incrementalism is more agile and allows a quick action (that can be corrected or readapted later). A synoptic approach requires much more previous knowledge before action is taken.
• Both books:
o They both agree on the fact that considerations of consequences are quite simple (p96 in Cyert & March) and that will be faced when the consequences appear (not when forecasted).
o Incrementalism in actions
o Problems fo neglected consequences: harmful failures are corrected after implemented
o Limited analysis disregarding possibilities
o Decisions are taken without adequate information