Knowledge Engine

Do people in different cultures disagree about what counts as knowing? The Gettier car case

·no read data

The idea

For a long time philosophers defined knowledge as justified true beliefjustified true belief The traditional definition of knowledge: believing something that is true, for good reasons.: you know something if you believe it, it is true, and you have good reason for believing it. In 1963 Edmund Gettier published a three-page paper with cases where all three conditions hold but most readers feel the person still does not know. These became known as Gettier casesGettier case An imagined story in which someone has a true, well-supported belief that is true only by luck, used to ask whether that person really knows., and the feeling they provoke became one of the standard examples of a philosophical intuition, a judgement about an imagined scenario that arguments are then built upon.

In 2001 Jonathan Weinberg, Shaun Nichols and Stephen Stich asked a pointed question in the journal Philosophical Topics: whose intuitions are these? They gave university students a Gettier-style vignettevignette A short written story given to participants in a study, who are then asked to make a judgement about it.. Bob’s friend Jill has driven an American car, a Buick, for years, so Bob believes she drives an American car. Unknown to Bob, the Buick has been stolen and replaced by a Pontiac, also American. His belief is true and reasonable, but it is true by luck. Does Bob really know, or only believe?

The paper reported that a large majority of Western participants said Bob only believed, while a majority of East Asian participants, a group of 23, said he really knew. Respondents from the Indian subcontinent were described as differing from the Western group even more. The authors drew a sharp conclusion: if intuitions about knowledge vary by culture, a method that treats them as evidence about knowledge itself is in trouble. The paper became a founding text of experimental philosophyexperimental philosophy The use of surveys and experiments to study how people respond to questions that philosophers argue about..

Why it might change a decision

The practical stake is methodological. If you are a philosopher, a cognitive scientist or anyone who builds an argument on how “people” react to a thought experimentthought experiment An imagined scenario used to test an idea by asking what we would judge or expect if it were real., the result predicts something testable: run the same case on large, culturally varied samples and you should see the split again. If you do not, the safer working assumption is that this intuition is widely shared, and the original difference becomes a cautionary tale about small samples rather than a fact about culture.

How solid is this?

The original cultural difference rests on a single study with small groups, and it has not reappeared when others looked for it. Minsun Kim and Yuan Yuan, writing in Episteme in 2015, gave the same car case to Yale students and analysed 82 East Asian and 58 Caucasian participants. In both groups about 85 percent said Bob only believed (85.4 and 86.2 percent), with no meaningful difference. They note that the effect had not shown up in any sample except the original 23. In the same journal that year, Hamid Seyedsayamdost reported collecting data from four sources, two online and two in person, often with larger samples than the original, and failing to detect significant differences between the ethnic and socioeconomic groups Weinberg and colleagues had compared.

The broadest test came from a team of 46 researchers led by Edouard Machery, published in 2017 in the Journal of Indian Council of Philosophical Research. They collected responses at 24 sites in 23 countries in 17 languages, from Brazil and Mongolia to Lebanon and Japan, and analysed 2,230 participants. When the question allowed the answer that the character “thinks he knows but does not actually know”, roughly 70 to 90 percent gave the Gettier answer at 23 of the 24 sites. The exception was a very small sample of Bedouins in Israel, which the authors themselves warn against reading much into. An earlier paper by much of the same team in Noûs, covering four cultural groups, reached a similar conclusion and floated the idea of a possibly innate core folk epistemologyepistemology The branch of philosophy that studies knowledge, belief and what makes a belief reasonable..

Two caveats keep this from being a clean verdict. First, wording mattered. On a second version of the knowledge question, only 11 of the 24 sites differed significantly from an even split, so how strongly the intuition shows up depends on how the question is put. Second, the replications tested particular cases. They do not show that every judgement about knowledge is culturally uniform, and the 2017 authors present a universal core as a hypothesis, not an established fact.

The strongest competing explanation for the original result is ordinary sampling noise: with 23 people in a group, a majority can tip either way by chance. A second possibility, which the later studies do not fully rule out, is that differences exist for other cases or other populations than those tested. Weinberg and colleagues’ broader point, that philosophers should check whether their intuitions are shared rather than assume it, survives; it is the specific evidence for a cultural split in the car case that has not held up.

Sources

  • Weinberg, J. M., Nichols, S., and Stich, S. (2001). Normativity and Epistemic Intuitions. Philosophical Topics, 29, 429-460. Text
  • Kim, M., and Yuan, Y. (2015). No cross-cultural differences in the Gettier car case intuition: A replicationreplication Running a study again, independently, to see whether the original result holds up. study of Weinberg et al. 2001. Episteme. doi:10.1017/epi.2015.17. Preprint
  • Seyedsayamdost, H. (2015). On Normativity and Epistemic Intuitions: Failure of Replication. Episteme, 12(1), 95-116. Record
  • Machery, E., Stich, S., Rose, D., and 43 others (2017). The Gettier Intuition from South America to Asia. Journal of Indian Council of Philosophical Research, 34(3), 517-541. doi:10.1007/s40961-017-0113-y. PDF
  • Machery, E., Stich, S., Rose, D., and colleagues. Gettier Across Cultures. Noûs, 645-664. Record

Check yourself

Answers are checked in your browser. Nothing is recorded or sent anywhere.

  1. In the 2001 study by Weinberg, Nichols and Stich, what did the Gettier car case result appear to show?

  2. What did Kim and Yuan (2015) find when they reran the car case with 82 East Asian and 58 Caucasian students at Yale?

  3. In the 2017 study across 23 countries, why do the authors put more weight on the 'Knowledge 2' question than on 'Knowledge 1'?

  4. Which is the most accurate reading of the evidence described in the post?

Questions without answers

  1. If a philosophical argument depends on readers sharing an intuition about a made-up case, how many people from how many backgrounds would you want to ask before treating the intuition as shared?
  2. The Knowledge 1 and Knowledge 2 questions described the same story but produced different patterns of answers. What does that suggest about how much a survey result depends on the exact words of the question?

Back to the archive