Skip to main content
Category: Principles
Type: Reasoning Principle
Origin: William of Ockham, 14th century
Also known as: Occam’s Razor, Principle of Simplicity, Law of Parsimony
Quick Answer — The Parsimony Principle, commonly known as Occam’s Razor, states that among competing explanations, the one with the fewest assumptions is most likely correct. Named after 14th-century philosopher William of Ockham, this principle doesn’t prove simplicity is true—it serves as a practical heuristic that favors explanations that require fewer extraordinary claims. Applied correctly, it sharpens scientific hypotheses, reduces overfitting in data analysis, and helps avoid unnecessary complexity in problem-solving.

​
What is the Parsimony Principle?

The Parsimony Principle states that simpler explanations are preferable to more complex ones, all else being equal. When multiple hypotheses explain the same phenomenon, the one requiring the fewest new assumptions is considered most likely—until evidence favors a more complex alternative.
“Plurality should not be posited without necessity.” — William of Ockham
This principle operates on a fundamental insight: every assumption in an explanation is a potential point of failure. Explanations with more assumptions have more opportunities to be wrong. By preferring simpler theories, we reduce the risk of building our understanding on unreliable foundations. However, simplicity is not an absolute truth criterion. The principle is pragmatic: simpler explanations are easier to test, harder to misapply, and less likely to include hidden assumptions. When evidence demands complexity, the principle does not oppose accepting it.

​
Parsimony Principle in 3 Depths

  • Beginner: When faced with competing explanations, ask: “Which one requires fewer assumptions?” The answer points toward the simpler, often more reliable explanation.
  • Practitioner: Apply parsimony to models and frameworks. When adding a new variable or assumption dramatically increases complexity, require proportionally stronger evidence to justify it.
  • Advanced: Understand that parsimony is about explanatory power, not about the world being inherently simple. Use it as a tool for theory selection, not as a metaphysical claim about reality.

​
Origin

The Parsimony Principle is named after William of Ockham (c. 1287–1347), a Franciscan friar and philosopher who taught at Oxford and Paris. While the principle’s underlying logic existed before him, Ockham famously wielded it to cut through theological and metaphysical disputes, arguing against unnecessary multiplication of entities. Ockham’s Razor—the “razor” suggesting it cuts away unnecessary assumptions—became one of the most influential methodological principles in Western thought. Scientists from Newton to Einstein invoked it; philosophers use it to evaluate theories; detectives and investigators apply it to find the most likely explanations. The principle gained particular importance in the scientific revolution. Francis Bacon praised it; Newton made it explicit in his rules of philosophizing: “We are to admit no more causes of natural things than such as are both true and sufficient to explain them.”

​
Key Points

1

Fewer Assumptions Mean Testable Explanations

Simple explanations are easier to test because they make fewer specific claims. Complex explanations with many assumptions can always be patched to accommodate contrary evidence, making them unfalsifiable.
2

Parsimony Prevents Overfitting

In data analysis and modeling, adding parameters without strong justification leads to models that fit noise rather than signal. Parsimony guards against building models that appear accurate but fail in practice.
3

Complexity Must Be Earned

Simplicity is the default; complexity requires justification. When a simple explanation fails to account for evidence, the burden shifts to demonstrating that additional complexity is necessary.
4

Parsimony Is a Heuristic, Not a Law

The principle is a rule of thumb for theory selection, not a logical proof. Sometimes the truth is complex, and evidence must lead us to accept that complexity.

​
Applications

Scientific Research

When competing theories explain the same phenomenon, scientists favor the simpler one until evidence requires complexity. This prevents unnecessary entities and mechanisms from cluttering our understanding.

Data Science

Model selection techniques like AIC (Akaike Information Criterion) and BIC (Bayesian Information Criterion) formalize parsimony by penalizing complexity. Simpler models that perform nearly as well are preferred.

Everyday Reasoning

When diagnosing problems—why is my car making this noise?—start with the simplest explanation (loose belt) before moving to complex ones (engine malfunction). This saves time and effort.

Legal Reasoning

Judges and juries are taught to prefer the simplest explanation consistent with the evidence. The prosecution must rule out reasonable doubt, which often means showing the simplest narrative is impossible.

​
Case Study

The development of heliocentric astronomy demonstrates parsimony in action. Ptolemy’s geocentric model required increasingly complex epicycles—circles within circles—to explain planetary motion. By the 16th century, the model required over 80 epicycles to match observations. Copernicus’s heliocentric model dramatically simplified the explanation: planets orbited the Sun in circular paths. While initially less accurate than the patched Ptolemaic model, it required far fewer assumptions. Kepler’s later refinement (elliptical orbits) added necessary complexity but still dramatically reduced total assumptions. The simpler model won not because it was immediately more accurate, but because it was more testable and led to productive research. When Newton later unified celestial and terrestrial mechanics, the heliocentric model achieved the ultimate parsimonious triumph: one theory explaining both falling apples and planetary motion.

​
Boundaries and Failure Modes

The Parsimony Principle is frequently misunderstood and misapplied. First, simplicity is not the same as truth. The world may be genuinely complex, and no amount of wishing for simplicity will make it simple. The principle is about selecting among explanations, not about the world’s inherent nature. Second, parsimony can be gamed by redefining terms. A theory that appears simple may smuggle in hidden assumptions through vague terminology. What’s “simple” depends on what’s already assumed. Third, sometimes the complex explanation is correct. Germ theory was more complex than miasma theory; relativity is more complex than Newtonian physics. When evidence strongly supports complexity, parsimony must yield.

​
Common Misconceptions

The principle states that simpler explanations are more likely, not that they are always correct. Evidence ultimately determines truth; parsimony is a tool for navigating uncertainty.
Parsimony opposes unnecessary complexity. When evidence supports complexity, accepting it is not a violation—the principle requires that complexity be justified by evidence, not rejected by assumption.
The principle applies wherever explanations compete: medicine, engineering, business strategy, personal decision-making. Any situation with multiple possible explanations benefits from parsimonious reasoning.

Occam's Razor

The most common name for the Parsimony Principle, named after William of Ockham.

Falsifiability

Karl Popper’s criterion that scientific theories must be testable and potentially disprovable. Parsimony helps create falsifiable theories by avoiding unfalsifiable complexity.

Model Selection

Statistical methods (AIC, BIC) that formalize the tradeoff between model fit and complexity.

Hickam's Dictum

“Patients can have as many diseases as they damn well please.” Sometimes multiple simple explanations are needed simultaneously.

Sunk Cost Fallacy

The tendency to continue investing in failing plans. Parsimony encourages abandoning complex explanations that don’t work, avoiding sunk cost reasoning.

​
One-Line Takeaway

Before accepting a complex explanation, ask: “Does this really require these extra assumptions?” Default to the simplest explanation that fits the evidence, and require strong proof before adding complexity.