We cannot reason about reality without some way of deciding what is real.
Every time we accept an explanation, reject a claim, or decide that something exists, we are applying a criterion for reality—whether we recognize it or not.
The danger is not simply having the wrong criterion. It is using one without knowing what it is.
An unexamined criterion allows us to mistake familiarity for truth, authority for evidence, or the limits of our perception for the limits of reality.
In The Fabric of Reality (1997), physicist David Deutsch argues that we must:
“develop reasons for accepting or rejecting the existence of entities that may appear in contending theories”
He proposes a criterion based on explanation: when our best explanation requires an entity with complex, autonomous behavior, we have reason to regard that entity as real.
This is a powerful proposal because it allows us to recognize realities we cannot directly observe.
But Deutsch’s criterion is not the only possible one. Other approaches emphasize observation, experimental intervention, causal effects, or empirical evidence.
The deeper principle is that our judgments about reality need standards that can themselves be examined.
This has three practical implications.
First, make your criterion explicit. What would justify accepting something as real? If you cannot explain your standard, you cannot reliably evaluate whether you are applying it.
Second, apply it consistently. A criterion that changes depending on which conclusion we prefer is not a reliable standard. We must be willing to apply the same reasoning to claims we like and claims we dislike.
Third, subject the criterion itself to criticism. A standard may exclude something real or accept something imaginary. We therefore need to examine its limitations and compare it with alternatives.
Deutsch recognizes this problem. He writes:
“the list of acceptable criteria for reality must be open-ended too”
This reveals an important consequence: disagreements about reality are sometimes disagreements about the standards used to judge it. More evidence will not necessarily resolve them until those standards are made explicit and examined.
A criterion for reality is not a machine for producing certainty. It is a means of making our judgments accountable to reasons and criticism.
We need criteria to judge reality, and criticism to judge our criteria.
That is how we avoid replacing the search for truth with confidence in our own assumptions.
References:
- Deutsch, D. (1997). The Fabric of Reality. Chapter 4, “Criteria for Reality,” pp. 85–86.
- Mossé, B. (2025). How We Know What’s Real: Deutsch’s Criterion for Reality.