Robespierre–Rousseau Principia: Vive l'Impossible

Ruthless discernment and the primacy of the attempt in constrained innovation · 18 October 2026 · Zachariah Roydel Robertson

Abstract

This paper proposes the Robespierre–Rousseau Principle, an operating philosophy for building under severe constraint. It joins two ideas taken, loosely and deliberately, from the French Revolution. From Maximilien Robespierre it takes the guillotine: a willingness to cut, without sentiment, whatever does not serve the work. From Jean-Jacques Rousseau it takes the social contract: the cunning to bind oneself to a shared commitment so firmly that retreat is no longer an option.

The combination yields two rules. First, ruthless discernment: if a tool, method or habit is slower, costlier or less effective than an available alternative, it goes. Second, the primacy of the attempt: when a goal looks impossible, the attempt is made anyway, because trying generates pathways of thought that do not exist before the try. The paper summarises this as in the try we surpass the can.

The argument draws on commitment theory (Schelling, Elster), Hirschman's "Hiding Hand", Kauffman's "adjacent possible" and research on creativity under constraint. It also confronts the principle's most serious objection, the Thermidor problem: Robespierre's guillotine eventually fell on Robespierre. The paper argues that this objection does not defeat the principle. It completes it, by turning ruthless discernment on the project itself.

1. Introduction: the problem of the roadblock

Every ambitious project meets a roadblock that looks final. A billing limit blocks a deployment. A dependency refuses to compile. A platform withholds an API. A small team discovers that what it needs, an operating-system kernel, a real-time voice network, a new kind of computing substrate, is the sort of thing only the largest institutions build.

At that moment there are two default responses. The first is to stop and wait for permission, resources or a better moment. The second is to keep every existing tool and plan, and push harder along the same blocked road. Both are common and both usually fail. The first never ships; the second exhausts itself against the obstacle.

The Robespierre–Rousseau Principle is a third response. It asks two questions in sequence. Does this serve us, or is there something cheaper, quicker or better? If not, cut it. Is the goal still worth it? If so, attempt it, however impossible it looks, because the attempt is where new options come from.

The principle was formulated in the course of building an early-stage voice-first social network and its supporting infrastructure on a founder's budget. It is offered here as a general philosophy of constrained innovation, and its claims are tested against existing literature rather than asserted on faith.

2. Historical sources: Rousseau and Robespierre

The name is shorthand, and a shorthand should be honest about what it borrows. Rousseau died in 1778, eleven years before the Bastille fell; he did not make the Revolution. But The Social Contract (1762) became one of its most quoted texts, and Robespierre was among Rousseau's most devoted readers. The pairing is therefore one of inspiration and execution: the theorist of the general will, and the man who tried to govern by it.

Rousseau's contribution is binding. The social contract asks each person to place themselves "under the supreme direction of the general will", so that individual wavering gives way to a common purpose. Read as a philosophy of action, this is a commitment device. The members of a project agree, in advance and together, that the shared goal outranks private comfort, and that the agreement holds when the work gets hard.

Robespierre's contribution is cutting. In his speech of 5 February 1794 on the principles of political morality, Robespierre paired virtue with terror, arguing that virtue without terror is powerless. The historical terror was a moral catastrophe, and nothing in this paper defends it. What the principle takes is a narrower, abstracted idea: that a commitment without the will to remove what obstructs it is only a wish.

The Robespierre–Rousseau Principle applies the guillotine only to things, never to people: tools, vendors, plans, habits, assumptions and sunk costs. Within that boundary, the Revolution supplies a vivid image of two capacities every builder needs: to bind and to cut.

3. The principle stated: discernment and the guillotine

The principle can be stated in one sentence: bind to the goal, cut everything else that does not serve it, and never decline to attempt. It has three parts.

1. The contract. Name the goal and commit to it publicly, with no private exit. The goal, not any particular means to it, is what is sacred.

2. The guillotine. Put every means on trial. A tool, vendor, process or plan survives only if it serves the goal and no cheaper, quicker or more efficient alternative exists. Familiarity, effort already spent and prestige are not defences.

3. The attempt. When the remaining road looks impossible, try. Do not make the attempt conditional on a prior proof that it can succeed.

The discipline lies in the order. Cutting comes before pushing. A team that pushes harder without first cutting simply exhausts itself on its own dead weight.

The guillotine also corrects a well-documented bias. Arkes and Blumer (1985) showed that people persist with failing investments because of what they have already spent, and Staw (1976) found that decision-makers escalate commitment to a losing course when they feel responsible for it. Sunk cost is precisely the defence the guillotine refuses to hear. The principle therefore separates two things that intuition merges: commitment to the end, which should be total, and commitment to any means, which should be zero.

4. Burning the boats: commitment as an epistemic engine

The idea of removing one's own retreat is old. Sima Qian's Records of the Grand Historian tells how Xiang Yu, crossing the river before the battle of Julu in 207 BC, ordered his army to sink the boats and smash the cooking pots, leaving three days' rations. The phrase survives in Chinese as pò fǔ chén zhōu (破釜沉舟). In 1519 Hernán Cortés grounded his ships on the Mexican coast to the same effect.

Thomas Schelling (1960) gave this practice its modern theory. In The Strategy of Conflict he showed that limiting one's own options can be rational, because a credible commitment changes what others expect and what oneself will do. Jon Elster (1979) extended the idea to the individual in Ulysses and the Sirens: a person who knows they will be tempted to quit can bind themselves in advance.

The Robespierre–Rousseau Principle adds a third function. Burning the boats does not only change motivation and credibility; it changes thinking. While retreat is available, the mind spends part of its effort evaluating retreat. When retreat is gone, all of that effort is redirected at the obstacle. Options that were never searched for, because an easier exit existed, are now the only places left to look.

This is why the principle describes itself as "victory or death". The phrase is a metaphor for closing the exit in the mind. It is not a literal wager of life or health, and the principle has no use for martyrdom: a builder who is destroyed cannot keep building.

5. In the try we surpass the can

The principle's central claim is that the question "can we?" is usually unanswerable in advance, and therefore the wrong question to ask first. Feasibility is not a fixed property of a goal. It depends on the options available, and the options available depend on what has already been tried.

To live is to try; in the try we surpass the can.

Three bodies of work support this claim.

The adjacent possible. Stuart Kauffman (2000) described how complex systems expand into an "adjacent possible": the set of states reachable one step from where the system now is. Each step taken enlarges that set. Applied to invention, a goal that is unreachable from today's position may become reachable from a position the attempt itself creates. The can is not measured from the starting line; it is remeasured after every step.

The Hiding Hand. Albert Hirschman (1967) studied development projects that succeeded only because their planners had underestimated the difficulty. Had the true obstacles been visible, the projects would never have started. Once committed, however, teams met each unforeseen problem with creativity they had not known they possessed. Hirschman's lesson is that the attempt summons capacities that a forecast cannot count.

Creativity from constraint. Patricia Stokes (2006) argued that well-chosen constraints drive novel work by blocking the familiar solution and forcing search elsewhere. Saras Sarasvathy's (2001) research on expert entrepreneurs points the same way: they tend to begin from the means they actually have and let goals take shape through action, rather than waiting for a complete plan.

Taken together, these sources support a modest but powerful version of the claim. The attempt does not guarantee success. It reliably produces information, options and new modes of thought that inaction cannot. A team that tries and fails still ends up somewhere the non-trying team cannot reach.

6. Objections and the Thermidor problem

A principle named after Robespierre must answer for Robespierre. Three objections are serious.

The Thermidor problem. On 28 July 1794 (10 Thermidor, Year II), Robespierre was guillotined by the Convention he had dominated. The instrument of purification consumed its operator. J. L. Talmon (1952) traced a line from Rousseau's general will to what he called "totalitarian democracy", and Hannah Arendt (1963) argued that the Revolution's turn to terror came from treating necessity as an absolute. The lesson for builders is that a guillotine pointed only outward eventually finds no limit.

The answer is not to abandon the guillotine but to apply it consistently. If every means must justify itself against the goal, then the project's current plan, its founder's favourite idea, and in the end the project itself are also on trial. The guillotine that may never fall on the builder's own assumptions is not discernment; it is dogma. Correctly applied, the principle cuts its own sacred cows first.

The planning-fallacy objection. Flyvbjerg and Sunstein (2016) tested Hirschman's Hiding Hand on more than 2,000 projects and found that cost overruns and benefit shortfalls were far more common than the creative rescue Hirschman described. Optimism about the attempt is often simply wrong. The reply is that the principle claims the attempt is always worth making, not that it always succeeds. Ruthless discernment is what keeps the attempt cheap: cut early, test small and let reality strike out the wrong path quickly.

The human-cost objection. "Progress at all cost" can become a licence to burn out people and cut through rules. The objection describes the principle accurately; it is wrong only to treat this as a flaw. All cost means all cost. Nothing is exempt from the trial: not the builder's comfort, rest or health, not the conventions of how things are normally done, and not rules that exist only because nobody has yet questioned them. Those who enter the contract accept that it may consume them, and a principle that quietly protected its own members from that cost would be only a slogan. The guillotine falls on whatever stands between the contract and the goal, including the people who swore to it.

7. Application: building under constraint

In practice the principle becomes a short procedure, run whenever work stalls.

1. Restate the contract. What is the goal, in one sentence? If the team cannot say, the roadblock is not the real problem.

2. Put the blocker on trial. Is the obstacle a law of nature, or a property of the means currently chosen? Most "impossible" blockers turn out to belong to a particular vendor, tool or plan.

3. Bring out the guillotine. List at least two alternative means. If any is cheaper, quicker or more effective, cut the old means now, regardless of what has been spent on it.

4. Make the smallest real attempt. Build the thinnest version that touches the hard part of the problem, not a polished version of the easy part.

5. Harvest the new pathways. Record what the attempt revealed: new options, new constraints, new ideas. This is the adjacent possible, enlarged.

6. Turn the blade inward. Ask whether the goal itself still deserves the contract. If it does, repeat. If it does not, cutting it is also the principle working.

Consider the motivating case: a small team building a voice-first social network and its own hosting layer, blocked by a cloud provider's billing constraint. The default responses are to wait for the account to be fixed, or to keep fighting the same provider. The Robespierre–Rousseau response is to treat the provider as a means on trial. Can the same service run on a different host, a self-managed server, or a narrower architecture that avoids the blocked feature? Each alternative attempted, even one that fails, teaches the team something about its real requirements.

The same logic applies to work that seems reserved for giants, such as writing an operating-system kernel. Linus Torvalds began Linux in 1991 as what he called a hobby project. The attempt did not wait for the resources of an established vendor, and it produced a different way of building software altogether.

8. Conclusion

The Robespierre–Rousseau Principle holds that builders under constraint need two capacities at once: Rousseau's power to bind themselves to a goal, and Robespierre's willingness to cut whatever does not serve it. Bound without cutting, a team drowns in its own dead weight. Cutting without binding, it drifts.

To these it adds a claim about knowledge: feasibility cannot be judged before the attempt, because the attempt changes what is feasible. Commitment theory, the adjacent possible and the Hiding Hand each support a version of that claim, while the planning-fallacy literature warns against confusing it with a guarantee.

Finally, the principle survives its namesake's fate only by turning the blade inward. A guillotine that spares the builder's own plans and assumptions repeats Thermidor. One that tries everything, including itself, becomes a discipline of discernment.

We can never not try. To live is to try, whether or not we can. In the try we surpass the can.

References

Arendt, H. (1963). On Revolution. New York: Viking Press.

Arkes, H. R., & Blumer, C. (1985). The psychology of sunk cost. Organizational Behavior and Human Decision Processes, 35(1), 124–140.

Elster, J. (1979). Ulysses and the Sirens: Studies in Rationality and Irrationality. Cambridge: Cambridge University Press.

Flyvbjerg, B., & Sunstein, C. R. (2016). The principle of the malevolent hiding hand; or, the planning fallacy writ large. Social Research, 83(4), 979–1004.

Hirschman, A. O. (1967). Development Projects Observed. Washington, DC: Brookings Institution.

Kauffman, S. A. (2000). Investigations. Oxford: Oxford University Press.

Palmer, R. R. (1941). Twelve Who Ruled: The Year of the Terror in the French Revolution. Princeton: Princeton University Press.

Robespierre, M. (1794). Sur les principes de morale politique qui doivent guider la Convention nationale. Speech to the National Convention, 5 February 1794 (17 Pluviôse, Year II).

Rousseau, J.-J. (1762). Du contrat social; ou, Principes du droit politique. Amsterdam: Marc-Michel Rey.

Sarasvathy, S. D. (2001). Causation and effectuation: Toward a theoretical shift from economic inevitability to entrepreneurial contingency. Academy of Management Review, 26(2), 243–263.

Schelling, T. C. (1960). The Strategy of Conflict. Cambridge, MA: Harvard University Press.

Sima Qian. (c. 94 BC). Shiji (Records of the Grand Historian), "Basic Annals of Xiang Yu".

Staw, B. M. (1976). Knee-deep in the big muddy: A study of escalating commitment to a chosen course of action. Organizational Behavior and Human Performance, 16(1), 27–44.

Stokes, P. D. (2006). Creativity from Constraints: The Psychology of Breakthrough. New York: Springer.

Talmon, J. L. (1952). The Origins of Totalitarian Democracy. London: Secker & Warburg.