Articulation Errors and Second-Language Learners
If a child is having difficulty producing a sound, is it a true error or is it due to influence from his first language?
Let’s begin with a quick look at the following Venn Diagram and then take a stroll through the research.
If two languages share a sound, you would expect that the shared sounds would not be difficult for a second-language learner to produce in their second language. For example, English and Spanish both have /b/ so the word “baby (bebe)” should not be a problem.
If a sound is not shared by two languages, you would expect the second-language speaker to 1) delete 2) distort or 3) replace the sound. This is in fact the case. Let’s use Spanish and take a look at a few examples:
- Deletion: Don’t becomes Don REASON: No final /t/ and no final clusters in Spanish
- Distortion: Spaghetti becomes Espaghetti REASON: No initial /s/ cluster in Spanish
- Replacement: This becomes Dis REASON: No /th/ so the brain chooses the most similar sound from the first language.
Articulation errors & second-language learners: The research
This gets a bit thick but you should get two things from the graphs below: 1) An unexpected craving for Lifesavers Candy. 2) The sounds in the center of the Venn Diagram are in graph 1 and there are very minimal differences between Primary- and Secondary Language Learners of English.
Second-language learners should not have difficulty with the sounds that are shared in both languages.
This was true! Graph 1 shows the shared sounds between both languages and notice that there is very little error (difference in performance). Graphs 2 and 3 show the sounds that are shared sounds that are normally later developing sounds. You can see that younger bilinguals (ages 4-6 years) begin by not being able to master the sound but then catch up at a later age (7-9 years). Graphs 4 and 5 shows the sounds that are unique to English. You can see that the older bilingual group does not look as similar to their English-speaking peers as they did on the shared sounds–especially for the later developing unique sounds. We were so excited at how clearly the data show this and, for those of you who geek out on research, you’ll love these graphs too. To orient you to the graphs: Sounds are listed across the bottom X-axis and percent of correct productions (technically proportion of the sounds produced correctly) are on the Y-axis where “1” equals all correct productions. The four stripes from left to right are: Young Bilingual – Old Bilingual – Young English-Speaker – Old English-Speaker.
How do we do this when working with our own caseloads?
We have a few resources in our library to aid in determining if a sound error is due to second language influence.Just click on the images below to download them.
Can this work with all languages? We believe so. We have launched an effort to gather information on other languages which we will release here in the future. Here is an example of the direction we are heading. We hope to provide information on all of the most common languages that we encounter on our caseload.Download these images as pdfs and find other great materials on our Resource Library!