Optical illusions and magic tricks can illuminate how the brain's visual systems are wired; NOVA scienceNOW's Magic and the Brain showed how. But can we learn about language processing in the same way? Colin Phillips thinks so. Phillips studies "linguistic illusions" to learn more about how our brains process words in real time. Here are a few linguistic illusions to try out, based on some that he shared with us at the AAAS. Skip to the continued link for explanations.


More people have been to Russia than I have.

The key to the cabinets are on the table.

While she was taking classes full time, Russell was working two jobs to pay the bills.

I'm not one to attribute every activity of man to the changes in the climate.

It is unlikely that Congress will ever pass that bill.

No peeking ahead. Ready? Here are the explanations.

More people have been to Russia than I have. This reads just fine, but what does it mean? Take a minute here to think about it. Got it? It means nothing. The sentence is a comparative illusion: It requires the reader to compare two items that can't be compared logically. So why was your brain temporarily fooled? Linguists like Phillips are trying to find out.

The key to the cabinets are on the table. You probably noticed the error in this grammatical illusion, but you would have noticed it much more quickly if I'd given you this sentence: "The key to the cabinet are on the table." Why? Your mental grammar filter allows "cabinets are" but blocks "cabinet are" even though the subject of the sentence is the (singular) key, not the cabinets.

While she was taking classes full time, Russell was working two jobs to pay the bills. This sentence doesn't break any rules, but you probably experienced a "boggle"--a little moment of mental recalibration--when you hit the word "Russell" because the first half of the sentence primed you to expect a female name. Indeed, when language researchers ask test subjects to read this sentence word by word, the readers stall momentarily on "Russell." Linguists call this a gender mismatch effect.

I'm not one to attribute every activity of man to the changes in the climate. This one comes from Sarah Palin, and the remarkable thing in this case is that it did not cause a boggle--you probably got her meaning (even if you didn't agree with it) despite the fact that the sentence is flipped on its head.

It is unlikely that Congress will ever pass that bill. The surprising thing about this very ordinary sentence is that the sentence that should have the opposite meaning, "It is likely that Congress will ever pass that bill," clangs so dissonantly. The word "ever" is a negative polarity item and (in English, at least) has a direct line to the brain's grammar police.

Why we do fall for these illusions? Why do we recognize immediately when some grammatical rules are broken but let other infractions slip? Phillips and his colleagues believe it has something to do with memory. As we proceed through a sentence, whether spoken or written, we store information from the beginning of the sentence so that we can use it to make sense of what comes later. Using linguistic illusions, Phillips hopes to discriminate between different strategies the brain might be using to store and retrieve this data. Want to learn more? Check out Phillips' paper on the subject (pdf).

User Comments:

This is so interesting! Thanks for posting about it!

I appreciate the post on language, but have to dispute the premise that the examples cited are illusions. Just because linguists can apply an abstract symbolic structure to some of the observable artifacts of language does not mean that people necessarily know or use these abstractions. Language is what people use to think and communicate, and if formal linguistics cannot capture real language use (as these 'illusions' suggest), then that seems more likely to indicate limitations with academic linguistics than real problems with the way that people use language. Fault tolerance in comprehension (e.g. illusions 1 & 4) just means that comprehenders normally try to comprehend _meaning_ and don't necessarily care that much about superficial grammatical/lexical/morphological forms. The third 'illusion' ("she... Russell") basically reflects the ecological invalidity of the experiment: an out-of-context sentence that introduces one character as a mere pronoun, and then a second character as a proper name. The last 'illusion' ("ever...") in particular may not fit with systematicity dogma (i.e. everything has to work according to the rules and characterizations that linguists have specified), but remains perfectly acceptable to any non-grammarian, even after extended reflection; regardless of how a grammarian thinks 'ever' _should_ work, the example demonstrates that real people think differently. Comprehension of theoretically anomalous utterances can help us understand what the mind is actually doing during comprehension, but if people's behavior does not follow linguistic rules, then that only indicates a failure of the rules to adequately describe behavior.

blog comments powered by Disqus