Criteria of Adequacy for Theories (One-Page Notes)
Overview
A theory is a set of statements that aims to explain some phenomenon or type of phenomenon.
The correctness of a theory is often not evidenced on the surface; historical examples can illustrate this (e.g., the heliocentric model proposed by Copernicus long before it was confirmed to be a more or less accurate model of planetary motion).
When we cannot confirm or disconfirm a theory or its components, we judge its quality using a set of adequacy criteria.
Core question: How do we evaluate theory quality in the absence of direct verification?
Source context: Schick, Theodore Jr. and Lewis Vaughn. How to Think about Weird Things. 8th Edition. pp. 180-190.
Simplicity
Simplicity concerns how many assumptions a theory requires.
Fewer assumptions = simpler theory; a theory that requires many beliefs is not simple.
An assumption is anything the theory requires you to believe in order to accept the theory.
If a theory asks you to believe a lot, or to believe in a lot of things, it is not simple.
Occam’s Razor (a way of stating simplicity): not to multiply entities beyond necessity; in other words, don’t believe in anything you do not have to believe in to explain what you’re trying to explain.
Practical implication: all else equal, a simpler theory is preferable because it commits us to fewer concepts/entities.
Examples:
The heliocentric model (Copernicus) is simpler than the geocentric model (Ptolemy) because it explains phenomena like retrograde motion with fewer complex assumptions (e.g., epicycles).
Evolution by natural selection is simpler than creationism by intelligent design as it relies on a few fundamental principles of variation and selection rather than invoking an intelligent agent with unspecified powers.
The Big Bang Theory is considered simpler than steady-state cosmology because it explains a wider range of observations (like cosmic microwave background radiation and red-shift) with fewer fine-tuned initial conditions or continuous creation assumptions.
Conservatism
A theory is conservative when it does not demand changing many of our well-supported beliefs.
If a theory requires substantial changes to established, well-supported beliefs, it is not conservative.
Rationale: we want new theories to conserve or preserve the good beliefs we already have.
Practical implication: conservation of core commitments helps maintain methodological stability and trust in the scientific or reasoning framework.
Examples:
Newtonian physics was conservative in that it built upon common observations of motion and gravity, even as it introduced new mathematical descriptions. A theory claiming objects fall upwards would be highly non-conservative.
The germ theory of disease was initially non-conservative as it challenged the existing miasma theory but gradually became conservative as sufficient evidence accumulated, changing many prior beliefs about illness.
Quantum mechanics, while revolutionary, was conservative in that it still accounted for the observed macroscopic phenomena explained by classical physics as a limiting case. A theory suggesting that energy is not conserved would be highly non-conservative.
Testability
A theory is testable if it is in principle possible to test it.
Testability may be achievable in the future even if current technology or methods cannot test it today.
Important nuance: testability does not require immediate experimental capability; it requires the possibility of empirical testing in principle.
Practical implication: testable theories invite potential falsification or confirmation through observation or experiment.
Examples:
The theory that increasing CO _{2} in the atmosphere leads to global warming is testable through climate modeling, historical data analysis, and direct measurement of atmospheric composition and temperature.
A theory claiming that all dreams are caused by tiny, invisible, undetectable pixies is not testable because there is no conceivable way to observe or measure the pixies or their influence.
The theory that specific genes increase susceptibility to certain diseases is testable through genetic studies, population analysis, and molecular biology experiments. Conversely, a theory that claims future events are predetermined by an unknowable cosmic fate is not testable.
Fruitfulness
A theory is fruitful if it leads to a lot of new experiments or discoveries.
This criterion captures the theory’s productive power and its potential to generate new knowledge.
Caution: assessing fruitfulness in advance is not easy; predictive success and the generation of new hypotheses are key indicators.
Practical implication: fruitful theories help drive research agendas and technological or empirical advances.
Examples:
Einstein's theory of General Relativity was fruitful, predicting phenomena like gravitational lensing, black holes, and gravitational waves, which subsequently led to numerous experiments and discoveries that confirmed its predictions.
The discovery of DNA's double helix structure was incredibly fruitful, leading directly to the fields of molecular biology, genetics, and biotechnology, opening up avenues for understanding disease, heredity, and evolution.
The theory of plate tectonics was extremely fruitful, explaining phenomena like earthquakes, volcanic activity, mountain formation, and the distribution of fossils across continents, and leading to new research in geology, oceanography, and geophysics.
Scope
A theory has wide scope if it explains more phenomena.
All things being equal, a theory with wide scope is better than one with narrow scope.
An explanation is more powerful when it explains all events of the same type rather than just an individual event.
Practical implication: broad explanatory reach reduces the need for ad hoc additions and increases coherence with related phenomena.
Examples:
Newton's Law of Universal Gravitation has wide scope because it explains both the falling of an apple on Earth and the orbits of planets around the sun.
The theory of evolution has wide scope, explaining the diversity of life, the fossil record, genetic similarities between species, and geographic distribution of organisms. A theory that only explains the existence of a single species in isolation would have narrow scope.
Mendelian genetics has wide scope, explaining patterns of inheritance in a vast array of organisms, from simple plants to complex animals. A theory that only explains the color of one specific type of flower would have narrow scope.
Connections and implications
The criteria collectively promote theories that are simple, non-disruptive to established knowledge, empirically testable, productive, and broadly explanatory.
Philosophical connection (implicit): these criteria align with the scientific ethos of understandable, falsifiable, and progressive knowledge advancement; they interact with debates about falsifiability, coherence, and explanatory power in philosophy of science.
Real-world relevance: choosing theories with these properties tends to favor robust, reliable explanations that can guide further research and practical applications rather than speculative or fragile accounts.
Notes on interpretation and use
When applying these criteria, weigh them against each other: for example, a theory with broad scope but poor testability may be compelling in some contexts, but risky in others.
The balance among simplicity, conservatism, testability, fruitfulness, and scope helps avoid overfitting (too many ad hoc assumptions) while still encouraging innovative explanations.
The criteria are not purely mathematical; they require