The Fallacy of Inconsistent Aggregate Judgments
This fallacy occurs when one assumes that majority opinions on two separate issues logically combine to yield a majority opinion on a third issue, ignoring that the groups forming each majority may be different. It often leads to misinterpretation of public opinion and aggregated data.
THE IDEA
The Fallacy of Inconsistent Aggregate Judgments
This fallacy arises when we treat the majority as a single, consistent entity and assume that its preferences on different issues stack logically. In reality, the majority on issue A may be composed of different individuals than the majority on issue B, so there is no guarantee that a majority supports the combination of both positions.
Example from Atkins v. Virginia (2002)
The U.S. Supreme Court considered whether executing prisoners with intellectual disabilities violated the Eighth Amendment's ban on cruel and unusual punishment, which depends on "evolving standards of decency" reflected in state laws. Justice Scalia argued that only 18 of the 38 states that permit capital punishment had barred execution of people with intellectual disabilities—a minority. The majority counted 30 states opposed (18 that specifically barred it plus 12 that prohibit capital punishment entirely). The source criticizes Scalia’s exclusion of non-death-penalty states and favors the majority’s count of all states. The fallacy lies in assuming that because a majority of death-penalty states do not differentiate, and a majority of states accept capital punishment, then a majority must approve of executing people with intellectual disabilities—which is false because the sets of states overlap differently.
Another example: The 1992 U.S. Presidential Election
Neither George H.W. Bush nor Bill Clinton received a majority of votes. That alone does not show that a majority preferred neither candidate. A voter who chose Perot could still have preferred one of them over the other; a vote total does not reveal that ranking.
Key Insight
Aggregate judgments (e.g., public opinion polls, state legislation) must be analyzed with care: the whole is not a simple sum of its parts when groups shift between issues.
When to use it
Interpreting public opinion polls on multi-issue policies.
Legal reasoning about 'evolving standards of decency' from state legislation.
Statistical analysis of survey data to avoid ecological errors.
Political strategy when building coalitions across different voter groups.
FROM THEORY TO EVERYDAY LIFE
In practice
Atkins v. Virginia and Aggregate State Counts
In 2002, the Supreme Court ruled against executing prisoners with intellectual disabilities. Scalia noted that a majority of states (38) permitted capital punishment, and only 18 of those barred the practice. The majority counted 30 states opposed: 18 that specifically barred it plus 12 that prohibited capital punishment entirely. The source uses the disagreement to show why majorities drawn from different groups cannot simply be combined into a single collective position.
The 1992 Presidential Election
Neither Bush nor Clinton received a majority of votes. It does not follow that a majority preferred neither candidate: someone voting for a third candidate might still have preferred one of them over the other. Separate vote totals do not establish the combined preference.
A MOMENT TO REFLECT
Put it to work.
Think of a political debate where one side says 'most people support X' and 'most people support Y,' then concludes most people support both X and Y. Why might this be a fallacy?
How would you correctly determine whether there is a national consensus on a controversial issue like the death penalty? What data would you need to avoid inconsistent aggregate judgments?
Consider a survey that asks two questions: 'Do you support higher taxes on the wealthy?' and 'Do you support increased spending on education?' If 60% support each, can you conclude that 60% support both? Why or why not?
References
- Atkins v. Virginia — U.S. Supreme Court (2002)
- Penry v. Lynaugh — U.S. Supreme Court (1989)
- Trop v. Dulles — U.S. Supreme Court (1958)
- Harmelin v. Michigan — U.S. Supreme Court (1991)
CONTINUE THE CONVERSATION
Source texts
- How Not to Be Wrong The Power of Mathematical Thinking (Ellenberg, Jordan)