Method

Tap-to-translate or a dictionary?

For reading, instant glosses usually teach more. In studies, readers given glosses learned more new words than readers with a dictionary, largely because dictionary users looked so few words up. A dictionary still earns its place when a word has several meanings, or when you want its gender, forms and examples. Most learners do best using both.

Do glosses or dictionaries teach more words?

A gloss is a meaning supplied next to the text: a note in the margin, a footnote, or a translation that appears when you tap a word. The classic comparison is a 1996 study by Jan Hulstijn and colleagues. Seventy-eight Dutch university students of French read a short story with marginal glosses, with a bilingual dictionary, or with neither, and were then tested on 16 unfamiliar words.

The gloss group did much better overall, scoring twice as well as the dictionary group or better on some of the tests. The reason was not that dictionaries teach badly. Of the 24 students allowed a dictionary, 20 used it at least once, and between them they looked up just 38 target words, an average of 1.9 out of 16. When a student did look a word up, though, they were more likely to remember it than the average gloss reader.

Computer glosses show the same pattern on a larger scale. Lee Abraham's 2008 meta-analysis of 11 studies found that computer-mediated glosses had a medium effect on reading comprehension and a large effect on incidental vocabulary learning.

The lesson is about friction. Leafing through a dictionary interrupts the story, so readers skip words or guess, and a guessed meaning can be a wrong one. A gloss costs a second, so it gets used.

Doesn't looking a word up yourself make it stick?

Effort does help memory, which is the strongest argument for the dictionary. Laufer and Hulstijn's involvement load idea holds that new words are retained according to how much need, search and evaluation a task involves. A dictionary look-up adds search and choosing between senses; a tap removes most of it.

Hulstijn tested the idea directly in a series of five experiments in 1992. In the three that compared conditions within the same learners, words whose meanings readers had worked out for themselves were remembered better than words whose meanings were given. But readers left to infer without any clue often settled on plausible wrong meanings. His conclusion for learners reading on their own was that a supplied translation or synonym is the safer option, even though it involves less effort.

A CALL dictionary study by Batia Laufer and Monica Hill adds detail. University learners in Israel and Hong Kong read a text with 12 rare words they could click for a translation, an English explanation, the sound or the word's root, and every click was logged. On a surprise test, the Israeli group recalled an average of 4 of the 12 words and the Hong Kong group 7.45. Learners differed in what they chose to look at, and looking at more than one kind of information tended to go with better retention. The number of clicks, on the other hand, was only weakly related to how many words were remembered, and the authors found no sign that easy look-ups led to shallow learning.

When do you need a real dictionary?

When a word has more than one meaning. A gloss gives you one meaning chosen for the sentence; a dictionary lists them all. In Bensoussan and Laufer's 1984 study of Israeli university students, context helped with only 13% of responses, and words with several meanings were among the most common sources of error. Two lines from a free Contos chapter at A1 show the problem:

Una puerta en el pasillo del tercer piso, entre su piso y el de su vecino, Andrés.
A door in the corridor on the third floor, between her flat and her neighbour Andrés's. Piso appears twice in one sentence: first as floor, then as flat.

From The Extra Key, chapter 1 (A1)

Él toca la guitarra y la música es muy fuerte.
He plays the guitar and the music is very loud. Toca can also mean touches, and fuerte usually means strong; here the guitar decides both.

From The Building on Calle Sol, chapter 1 (A1)

If the meaning a tap gives you doesn't fit the sentence, the word probably has another sense. That is the moment to open a dictionary, and the same goes for any single translation, whoever supplies it.

Tap-to-translateDictionary
SpeedA second, without leaving the pageLong enough to lose the thread of the story
Meanings shownOne, chosen for this sentenceAll of them; you choose
GrammarLittle or noneGender, plural, verb forms, register
ExamplesThe sentence you are readingSeveral, often from real usage
Best forKeeping the story movingWords that keep coming back or don't fit

How do you use both well?

Read a free chapter, tap what you need, and look up one word properly in a dictionary.

Read free in your browser

Frequently asked questions

Is tap-to-translate cheating?

No. In Hulstijn's experiments, readers left to guess without help often learned wrong meanings, and he concluded that a supplied translation is the safer choice when you read alone. Glossed readers in his 1996 study also learned more words than readers with a dictionary.

Why don't I remember the words I tapped?

Because one meeting is rarely enough. A tap gets you through the sentence; remembering the word takes more encounters, in reading or in review. Save the words that matter as flashcards and keep reading the same series, so they come round again.

Is a translation or a monolingual dictionary better?

Each suits different learners. In Laufer and Hill's study, learners differed in what they looked at, and those who combined more than one kind of information, such as a translation and an explanation in English, tended to remember words better.

Can a tap translation be wrong?

It can be the wrong sense for your sentence, as with any single translation. Words like piso (floor or flat) or fuerte (strong or loud) depend on context. If the meaning doesn't fit, check a dictionary.

Does looking up more words mean learning more?

Not directly. In Laufer and Hill's study the number of look-ups was only weakly related to how many words learners remembered. What you do with a word after looking it up matters more than how often you click.

Try the method on a real story

Pick a language and a level, and start chapter 1 in your browser. The first two chapters of every series are free.

Start reading free

Free chapters in every series. No download, no card needed.