Il y a toujours un chemin.

HumanFrench

Un français patient, bienveillant et humain.

LE LABO · The exam · Nº 28

I have nothing to say in the speaking test: build a shell of ready-made phrases, then fill it with your own life

By Débonnaire · 3 min read

The problem

The examiner hands you the prompt: “Talk about an activity you enjoy.” You enjoy plenty of things. But in French, at this moment, your head is an empty room. You start “J’aime…”, you hunt for the word, you find it, you lose the sentence. Every silence feels a minute long. On the way out, you think of everything you could have said. It is not that you have nothing to say: it is that you are building every sentence from scratch while someone watches.

What the research found

Frank Boers and colleagues (Boers et al., 2006) followed 32 Belgian students of English over 22 hours of lessons. One group worked on the same texts as the other, but with one instruction: notice and remember formulaic sequences, the chunks native speakers say in one breath, like “in my opinion” or “what I like about it is”. At the end, two examiners who did not know who belonged to which group rated an oral interview. They judged the “chunks” group as more fluent and more at ease, and the more chunks a student used, the higher the rating. The explanation is mechanical: a memorised chunk is retrieved whole, with no construction effort, which frees attention for what you actually want to say.

That leaves the freeze itself. Ali Al-Hoorie (Al-Hoorie, 2018) pooled 32 research reports and 32,078 learners in a meta-analysis, a study that adds up the results of other studies. He measured the link between the “ideal L2 self”, the picture you have of yourself speaking the language one day, and the effort learners say they put in. The correlation reaches r = 0.61, a strong link: people who can clearly see themselves speaking French work harder at it. The link with actual grades is much weaker, r = 0.20, and that needs saying. But for a candidate who blanks, the idea is useful: a concrete image of yourself talking, about your own subjects, feeds the effort of practising.

Put together, the two findings give a strategy: a shell of ready-made phrases, learnt whole, that you fill with your real life, the part you can picture precisely.

Do this today

Pick three ready-made openers and say them aloud until they come out without thinking: « Ce qui me plaît, c’est… », « Par exemple, le week-end dernier… », « Pour être honnête… ». Then fill them with something true: your neighbourhood, your job, what you did on Saturday. Record yourself for one minute on your phone. Listen back without judging. Tomorrow, same openers, a different topic. The shell stays, the content changes.

How our lessons use it

The activity « Je m’appelle Nadia » (A1) ends with “Dis-le à ta façon”: you borrow the shape of Nadia’s sentences and say three things about yourself, your name, your city, what you like on Saturdays. That is exactly a shell filled with your life. The shadowing version of the video « L’ascenseur » pauses so you can repeat each sentence aloud: you collect chunks that are ready to come out. Start with Je m’appelle Nadia (A1).

The honest limit

Boers’s study involved 32 students, in English, with two raters: it is small, and a replication has been called for. Al-Hoorie’s meta-analysis measures self-reported effort, not a score, and a correlation is not a cause. Neither concerns the TCF or the TEF. These are solid leads for practising, not guarantees of a grade.

Your questions

What is a "formulaic sequence"?

A chunk of words that French speakers say as one piece, like « à mon avis », « ce qui me plaît, c'est » or « en ce qui concerne ». You retrieve it whole instead of building it word by word.

Doesn't reciting memorised phrases sound fake?

If you recite a whole text, yes. If you use ready-made openers to carry your own ideas, you simply sound like someone who is talking.

What does r = 0.61 mean?

It is a correlation, between 0 and 1. Here, the clearer learners' picture of themselves speaking the language, the more effort they reported. That is a strong link, but not proof of cause.

References

Boers, F., Eyckmans, J., Kappel, J., Stengers, H., & Demecheleer, M. (2006). Formulaic sequences and perceived oral proficiency: Putting a Lexical Approach to the test. Language Teaching Research, 10(3), 245–261. doi:10.1191/1362168806lr195oa

Al-Hoorie, A. H. (2018). The L2 motivational self system: A meta-analysis. Studies in Second Language Learning and Teaching, 8(4), 721–754. doi:10.14746/ssllt.2018.8.4.2

Try the free lessonRecorded coursesComing soonNext article →

You may also like