The Man and the Works

Herbert Alexander Simon (1916–2001) was an American political scientist, economist, cognitive psychologist and computer scientist who spent most of his career at Carnegie Mellon University. He is the only figure in this syllabus to hold both the Nobel Memorial Prize in Economic Sciences (1978), awarded for his research into decision-making within economic organisations, and the Turing Award (1975), computer science's highest honour, shared with Allen Newell for work in artificial intelligence. He also received the National Medal of Science.

The Royal Swedish Academy described Administrative Behavior as epoch-making. His citation count runs into the hundreds of thousands across disciplines that barely speak to one another — economics, psychology, computer science and public administration.

The Demolition: "The Proverbs of Administration" (1946)

Simon's entry into the discipline was an assault. The classical principles, he argued, are proverbs — like look before you leap and he who hesitates is lost, they come in mutually contradictory pairs, and one invokes whichever supports the conclusion already reached. They are unscientific, derived from casual observation and deductive assertion rather than from evidence, and they offer no criterion for weighing one against another when they conflict.

On the four Ps specifically. Simon's attack on Gulick's departmentalisation is more radical than most notes record. He argued the categories are ambiguous: purpose and process shade into one another, and persons and place can be subsumed under purpose. A school health service — is it health (purpose), education (purpose), or medicine (process)? Every real organisation is built on several bases at once, at different levels, so the "principle" tells you nothing about which to use.

What Else Simon Rejected

• The politics–administration dichotomy. Simon rejected the separation, holding that administrative decisions unavoidably contain value elements — there is fusion, not dichotomy. This is a significant departure from Wilson, Goodnow and the classical tradition.
• Efficiency as a self-evident good. He retained efficiency as a criterion but insisted it be operationally defined rather than invoked as a slogan.
• Prescription without evidence. Classical theory is prescriptive; Simon's behavioural theory is descriptive. This methodological switch is the deeper revolution.

The Fact–Value Dichotomy

Simon's positive programme begins here. Every decision has two kinds of premise.

• Factual premises are statements about the observable world. They can be tested and found true or false.
• Value premises are expressions of preference about what ought to be. They cannot be empirically validated — only accepted or rejected.

Since science is built on what can be tested, a science of administration can be constructed only on factual premises. Administration is not wholly fact; therefore only part of administration can be scientific. That is Simon's careful position, and it is more modest than the caricature.

The Simon–Waldo Debate

This is the most consequential quarrel in the history of public administration, and it deserves a section of its own. UPSC has framed a stem directly on it — that Simon's identification of decision-making as the core field is logically acceptable but his positivist underpinning is problematic. That sentence is the debate.

The Setting: Two Books, One Year Apart

Both men were young, both were attacking the same target, and both published landmark books within a year of each other.

• Herbert Simon, Administrative Behavior (1947). The orthodoxy failed because it was not scientific enough. Its principles were untestable maxims. The remedy is rigour: operational definitions, empirical propositions, and a clean separation of the factual from the value component so that at least part of administration can be studied scientifically.
• Dwight Waldo, The Administrative State (1948). The orthodoxy failed because it pretended to be neutral technique while smuggling in a political philosophy. Waldo showed that the efficiency movement carried a whole ideology — business values, Progressive-era faith in expertise, a particular vision of the good society — while presenting itself as mere method. The remedy is not more science but honesty: recognise that public administration is a branch of political theory and argue about its values openly.

Note the symmetry. They agreed the classical school had failed and disagreed completely about why. Simon wanted to purify administration of values; Waldo wanted it to own them.

The Exchange Itself

The confrontation came in the American Political Science Review, 1952. Waldo published Development of a Theory of Democratic Administration, arguing that administration must be reconceived in democratic terms. Simon replied — as did Peter Drucker — and Waldo rejoined. The tone was unusually sharp for an academic journal. Simon accused Waldo of using words loosely and of failing to define his terms; Waldo accused Simon of a spurious precision that defined problems narrowly enough to be solvable while excluding everything that mattered.

The Five Substantive Issues

1. Can values be separated from facts? Simon says yes, analytically — the administrator receives value premises from the political process and reasons scientifically about the factual ones. Waldo says no, because of the means–ends chain: every intermediate end is also a means, so value keeps re-entering at every link. Choosing to study only facts is itself a value choice about what is worth studying.
2. Is efficiency a neutral criterion? Simon retains it, suitably defined. Waldo's whole book is an argument that it is not — the efficiency movement carried the values of a particular class at a particular moment, and the question efficient for what and for whom cannot be answered scientifically.
3. Is logical positivism a method or a philosophy? Simon drew on the Vienna Circle: only logical and empirically verifiable statements are meaningful; ethical statements merely express preference. Waldo's reply is that this is itself a contested philosophical position, not a neutral technique — and one that was already losing ground in the philosophy of science.
4. What is democratic administration? Waldo wanted administration to embody democratic values in its own working — participation, representativeness, responsiveness. Simon thought the idea confused, since democracy in a workplace and democracy in a polity are different things, and efficiency in executing democratically chosen ends is itself a democratic service.
5. What kind of discipline is public administration? This is the real stake. Simon: an administrative science, generic across public and private, allied to psychology, economics and later computing. Waldo: a field of political theory and practice, defined by its concern with the state, justice and the good society, and therefore inseparable from politics.

Who Won, and What Followed

Neither. The discipline split, and the split is still visible in every syllabus including this one.

• Simon's line ran into the behavioural revolution, organisation theory, decision sciences, public choice, policy analysis and artificial intelligence. He himself largely left public administration for management science and cognitive psychology — which is why critics say he did not merely win an argument but took the discipline's best minds out of it.
• Waldo's line ran into the values-and-relevance strand: he convened the Minnowbrook Conference of 1968, from which New Public Administration emerged with its commitment to relevance, values, social equity and change. Waldo also gave the discipline the language of an identity crisis.
• Nicholas Henry's paradigms record the split as history: Paradigm 3 — Public Administration as Political Science and Paradigm 4 — Public Administration as Administrative Science run in parallel through roughly the same decades. Citing this shows you understand the debate produced not one winner but two disciplines sharing a name.

Waldo's own proposed resolution is worth knowing and rarely written. He suggested public administration should be understood as a profession on the model of medicine: it draws on many sciences without being reducible to any of them, and it is unified not by a method but by a practice and a value commitment. A doctor uses chemistry and physiology but is defined by the obligation to heal. On that view Simon's science is a necessary input to public administration but is not, and cannot be, the whole of it.

The Allied Critics

• Norton Long — power is the lifeblood of administration. An agency without political support cannot execute even a perfectly designed programme, so a theory that brackets politics cannot explain what administrators actually do or why they succeed and fail.
• Chris Argyris (1973) — Simon's administrative man describes a passive, dependent, non-self-actualising personality, and by describing it as normal helps to perpetuate it. Simon's reply, Organization Man: Rational or Self-Actualizing?, argued that describing a constraint is not endorsing it. Naming both sides of this exchange is a mark of thorough preparation.
• Robert Dahl, writing before the debate proper, had already set out the three obstacles to a science of administration — values, variable human behaviour, and the differing social framework across societies. His first obstacle is Waldo's whole case in compressed form.

Bounded Rationality

Simon accepted rationality as the object of study and denied that it is ever complete. Absolute or objective rationality would require the decision-maker to know all alternatives, know all consequences of each with certainty, and rank them all against a stable preference ordering. None of these obtains. In Simon's own formulation, the capacity of the human mind for formulating and solving complex problems is very small relative to the size of the problems whose solution real behaviour requires.

Simon's Three Limits on Rationality

1. Skills, habits and reflexes — much behaviour is unconscious and not subject to deliberation at all.
2. Values and conceptions of purpose — the individual's own values may diverge from the organisation's, bounding what they will even consider.
3. Extent of knowledge and information — the decision-maker cannot know all alternatives or all consequences.

To these Simon added limits of time and computational capacity. And in The Sciences of the Artificial he gave the idea its most elegant statement: human rational behaviour is shaped by a pair of scissors whose two blades are the structure of the task environment and the computational capabilities of the actor. Rationality is not a property of the mind alone; it is the fit between mind and world. That metaphor is a genuine differentiator in an answer.

Satisficing and the Three Features — The 2012 Question

UPSC has asked directly for the three features characterising Simon's original view of bounded rationality. They are:

1. Search for alternatives. Not exhaustive but limited and sequential. The decision-maker examines options one at a time and stops at the first that clears the bar. A department seeking a vendor does not evaluate every vendor in the world; it shortlists a few on prior performance and reputation.
2. Satisficing. A portmanteau of satisfy and suffice. The decision-maker accepts an alternative that is satisfactory rather than optimal, because the cost of finding the absolute best usually exceeds the benefit. A District Magistrate siting a school chooses a location meeting the essential criteria of accessibility, land availability and cost — not the theoretically ideal one.
3. Aspiration adaptation. The threshold of "good enough" is not fixed. It rises when outcomes consistently exceed it and falls when they consistently disappoint. An officer who repeatedly exceeds a disposal target raises it; one defeated by resource constraints lowers it. This dynamic adjustment is what allows learning — and, less happily, what allows chronic underperformance to become normalised.

Programmed and Non-Programmed Decisions

From The New Science of Management Decision (1960). Note that Simon presented this as a two-dimensional taxonomy — decision types crossed with traditional and modern techniques. UPSC asked about exactly this two-dimensional structure in 2021, so reproducing all four cells is the answer.

Two further propositions belong here. Higher organisational levels handle proportionately more non-programmed decisions, lower levels more programmed ones — which has direct implications for where analytical capacity should be concentrated. And the managerial task is, so far as possible, to move decisions from the non-programmed toward the programmed category, since programmed decisions carry more rationality and less risk. That is the theoretical case for standard operating procedures — and, in its modern form, for algorithmic decision support.

The Decision Cycle: Intelligence, Design, Choice — and Review

Two Traps Within the Cycle

• Solving the wrong problem. In the intelligence phase, what presents as the problem is often a symptom. Rising costs may be a symptom of poor inventory control. Simon's advice — reason by analogy with previously solved problems — is a heuristic for avoiding misdiagnosis.
• Skipping design. Organisations under pressure leap from problem identification to a familiar solution, generating no genuine alternatives at all. This is the most common and least noticed failure in Indian policymaking, and the reason the design phase deserves the most emphasis in an answer.

Simon's Other Contributions

• Mechanisms of organisational influence. How does the organisation shape the individual's decision premises? Through authority, communication (advice and information), training, the criterion of efficiency, and organisational loyalty and identification — supported structurally by the division of work and standard practices. The organisation does not command outcomes; it supplies the premises from which individuals reason.
• Zone of acceptance. Simon took Barnard's zone of indifference and renamed it. The improvement is real: subordinates are not indifferent to orders they obey — they positively accept them. He carried over the contribution–satisfaction equilibrium intact, adding the notion of the individual's zero point, the best alternative forgone by remaining in the organisation.
• Organisational identification. Simon named this as one of his own novelties. Individuals identify with the sub-goals of their unit rather than the organisation's overall purpose — which produces goal displacement. Note the difference from Merton: Merton's displacement is rules becoming ends; Simon's is sub-goal identification crowding out the whole.
• Decision premises. Simon's reframing of what an organisation actually does — not making decisions but supplying the premises from which its members decide. This is his most durable contribution to organisation theory.
• Attention as the scarce resource. Simon observed that a wealth of information creates a poverty of attention. In an age of dashboards and real-time monitoring, this is arguably his most contemporary insight — the Secretary's attention, not the data, is the binding constraint.

Lindblom's Incrementalism: The Counterpoint

Simon bounded rationality but kept it. Charles E. Lindblom, in The Science of "Muddling Through" (1959), went further: the comprehensive-rational ideal is not merely unattainable but undesirable, and administrators are right to abandon it.

Lindblom's Argument in Detail

• Successive limited comparisons. The administrator compares a narrow range of policies differing only marginally from the status quo. Alternatives radically different are never examined — nor should they be, since analysing them is beyond anyone's capacity.
• Means and ends chosen simultaneously. One does not first fix an objective and then find the means; the choice of policy and the choice of objective are the same act.
• The test of good policy is agreement. Since participants disagree about ends, the only workable criterion is that they can agree on a policy — not that it demonstrably maximises anything.
• Partisan mutual adjustment. In The Intelligence of Democracy (1965), Lindblom argued that coordination among many partisan participants, each pursuing its own interest, produces better outcomes than central analysis. Fragmentation is a feature, not a bug.
• Disjointed incrementalism, developed with David Braybrooke in A Strategy of Decision (1963), and refined in Still Muddling, Not Yet Through (1979).

Simon Against Lindblom — The Comparison Table

The Other Responses

• Amitai Etzioni — mixed scanning (1967). The synthesis. Use the rational mode for fundamental, direction-setting decisions and the incremental mode for the bit decisions that follow within that frame. His analogy: a weather satellite with a broad-angle camera scanning everything at low resolution, and a zoom lens examining in detail only the areas the broad scan flags.
• Yehezkel Dror — the normative-optimum model (1968). The sharpest attack on Lindblom: incrementalism is inherently conservative and works only where existing policies are broadly satisfactory, change is incremental and problem-continuity holds. None of those conditions obtains in developing countries needing transformation. Dror's alternative combines rational and extra-rational elements — judgement, intuition, creative invention — across eighteen phases in three stages: meta-policymaking, policymaking and post-policymaking. For India, Dror's critique is the more directly usable one.
• Cohen, March and Olsen — the garbage can model (1972). In "organised anarchies", problems, solutions, participants and choice opportunities float independently and are coupled largely by accident. Solutions look for problems as often as the reverse.
• Graham Allison (1971). Three lenses on the same decision — the rational actor, organisational process, and bureaucratic politics models. A reminder that Simon's frame is one of several.
• Kahneman and Tversky. Simon's successors in psychology. Prospect theory, heuristics and biases, and the System 1 / System 2 distinction develop bounded rationality into an empirical research programme; Kahneman received the Nobel in 2002. Gerd Gigerenzer takes the opposite tack with ecological rationality — fast and frugal heuristics are not failures but adaptations to environments, which is arguably closer to Simon's own scissors metaphor.

Criticism of Simon

• Positivism. The central charge, made by Waldo: values cannot be excised, efficiency is itself a value, and the fact–value dichotomy imports a philosophy while claiming to be method. UPSC has framed a stem around precisely this.
• Power is absent. Norton Long — administrative decisions are shaped by power, and a theory that brackets politics cannot explain them.
• Bounded rationality is still rationality. Sociological and political critics note that culture, hierarchy, personal relationship and political pressure do not fit the intelligence–design–choice sequence at all. In Indian conditions, add hierarchical deference, inter-departmental rivalry and political interference — each distorting a specific phase.
• Too general to guide. The model describes the structure of the process and says nothing about the substance of a good decision in any particular context.
• Argyris's charge. Administrative man describes a passive, non-self-actualising personality and, by describing it, normalises it. Simon replied that describing a constraint is not endorsing it.
• He destroyed more than he built. A common assessment: "The Proverbs" was devastating and the positive science of administration Simon promised never fully materialised. The behavioural revolution changed the questions more than it produced settled answers.
• Neglect of the environment. Like the classicists, early Simon studies the organisation largely in isolation — the objection systems theory and Riggs's ecological approach later formalised.

Simon and Paper 2: Indian Administration

Cross-Links Across Paper 1

UPSC Previous Year Questions — Verified

Practice Questions For Test Series

1. "The principles of administration are proverbs." Examine Simon's demolition with reference to at least three contradictory pairs.
2. "Even bounded rationality is still rationality." Assess the sociological critique of Simon's framework.
3. Examine the Simon–Waldo debate and its consequences for the possibility of a science of administration.
4. "The means–ends chain makes the fact–value dichotomy untenable." Discuss.
5. Compare Simon's satisficing with Lindblom's incrementalism as accounts of how administrators actually decide.
6. "Incrementalism is a strategy for societies that do not need to change." Examine Dror's critique with reference to India.
7. Evaluate Etzioni's mixed scanning as a synthesis of the rational and incremental models.
8. "India's weakest decision phase is design, not intelligence." Critically examine.

Answer-Writing Toolkit

Two Introductions

• Demolition hook: "Simon entered the discipline by calling its principles proverbs — maxims that come in contradictory pairs and cannot be tested. Having removed the foundation, he supplied a replacement: the decision as the unit of administrative analysis."
• Realism hook: "Simon's premise is a statement about the size of the human mind relative to the size of its problems. From that single observation follow bounded rationality, satisficing, administrative man, and the standard operating procedure."

Two Conclusions

• Balanced: "Waldo was right that the fact–value dichotomy conceals a philosophy, and Long right that a theory without power cannot explain administration. Yet Simon changed the questions permanently: no serious account of administration now begins from an optimising actor, and every argument about standard procedure, evidence-based policy or algorithmic decision support is conducted in his vocabulary."
• Indian: "India has built an impressive intelligence apparatus and a functioning choice apparatus, and left the design phase almost empty. Alternatives are recorded rather than generated, and evaluation rarely returns to inform the next decision. Simon's diagnosis is precise, and Dror's warning is the necessary supplement: a society that must transform cannot rely on incremental adjustment."

Ten Power Keywords

• Proverbs of administration — principles in contradictory pairs.
• Factual and value premises — the two components of every decision.
• Means–ends chain — every end becomes a means.
• Bounded rationality — intendedly rational, limitedly so.
• Satisficing — satisfy plus suffice.
• Aspiration adaptation — the moving threshold.
• Administrative man — against economic man.
• Programmed and non-programmed — with traditional and modern techniques.
• Zone of acceptance — Simon's improvement on Barnard.
• Scissors metaphor — task environment and computational capacity.

Frequently Asked Questions

What is bounded rationality?

Bounded rationality is Simon's proposition that real decision-makers are intendedly rational but only limitedly so. Rationality is bounded by skills, habits and reflexes; by values and conceptions of purpose that may diverge from the organisation's; and by the extent of knowledge and information available, to which Simon added limits of time and computational capacity. Because complete optimisation is impossible, decision-makers work with a simplified model of reality.

What is satisficing?

Satisficing is a portmanteau of satisfy and suffice. It describes choosing an alternative that is satisfactory rather than optimal. The decision-maker sets an aspiration level, searches alternatives sequentially, and stops at the first that clears the bar, because the cost of finding the absolute best usually exceeds the benefit. In Simon's illustration, maximising means finding the sharpest needle in the haystack; satisficing means finding one sharp enough to sew with.

What is the fact–value dichotomy?

Simon held that every decision rests on two kinds of premise. Factual premises are statements about the observable world and can be tested as true or false. Value premises express preferences about what ought to be and cannot be empirically validated. Since science rests on what can be tested, a science of administration can be built only on the factual component. The means–ends chain complicates this, because every end becomes a means to a higher end, so intermediate links carry both elements.

What is the difference between programmed and non-programmed decisions?

Programmed decisions are repetitive, routine and structured, with a definite procedure already worked out; traditional techniques include habit, standard operating procedures and organisation structure, and modern techniques include operations research and computer simulation. Non-programmed decisions are novel, unstructured and consequential; traditional techniques include judgement, intuition and executive selection, and the modern technique is heuristic problem solving. Higher organisational levels face proportionately more non-programmed decisions.

What are Simon's phases of decision-making?

Intelligence, in which the environment is scanned and the problem identified and defined; design, in which alternatives are invented, developed and analysed; and choice, in which one course is selected. Simon later added review or implementation, in which past choices are assessed and their outcomes feed back into future decisions. The phases are sequential but not rigidly so, and the decision-maker may loop back at any point.

How does Lindblom's incrementalism differ from Simon?

Simon retains rationality as the ideal while showing it is bounded in practice; Lindblom rejects the comprehensive-rational ideal as neither attainable nor desirable. Simon's account is psychological and centred on the individual decision-maker; Lindblom's is political and centred on bargaining among partisan participants. For Simon a good decision satisfices; for Lindblom a good policy is one participants can agree on. Yehezkel Dror criticised incrementalism as inherently conservative and unsuitable for developing societies, and Amitai Etzioni proposed mixed scanning as a synthesis.

Conclusion

Simon's achievement was to replace a vocabulary. Before him, administration was described in principles that could not be tested; after him, in decisions that could be studied. Bounded rationality, satisficing, the aspiration level, programmed and non-programmed decisions, the zone of acceptance and the idea of the organisation as a supplier of decision premises are all now simply how the subject is spoken.

The criticisms hold and must be conceded. The fact–value dichotomy conceals a philosophy, as Waldo showed; power is missing, as Long insisted; and bounded rationality remains rationality, which is why Lindblom's politics and Dror's demand for transformation are necessary supplements rather than rivals. But the measure of the achievement is visible every time an administration writes a standard operating procedure, commissions an evaluation, or argues about whether the evidence supports a policy. Those are Simon's questions, and nobody has found better ones.