If you spend time in any large city in the United States, chances are you will hear English spoken in a variety of accents. Some of these accents are just native speakers of English from different regions of the country, while others reflect the speech patterns of people who learned another language as a child and then learned English later.
A foreign accent matters in social situations. The accent immediately marks someone as an outsider, which can lead to distrust. In addition, some native speakers find accents hard to understand, and so the accent can also create difficulties in communication.
Are accents a fixed part of a person’s speech pattern in their non-native language? This question was explored by Matthew Goldrick, Elin Runnqvist, and Albert Costa in a paper in the April, 2014 issue of Psychological Science.
These researchers were interested in whether switching back-and-forth between languages would affect the strength of a person’s accent. To test this possibility, participants were run in Barcelona, Spain. All of them were native speakers of Spanish who began learning English by the age of 4.
In this study, participants saw pictures of simple objects on a computer screen in which the first sound of the word for that object began with a d (as in desk or doce) or a t (as in tin or taza). The picture was surrounded by a colored frame. Participants were instructed to use the English word for one color and the Spanish word for the other color. The key question is whether participants would have a stronger accent on trials on which they switched languages from the previous trial than on trials on which they used the same language on consecutive trials.
It can be difficult to measure strength of accent by ear, but the researchers used a clever method to study the strength of the accent. When Spanish speakers produce the sounds ‘d’ and ‘t’, they engage their vocal cords earlier than when English speakers produce these same sounds. If you look at a digital recording of speech, you can actually see the burst of noise when the vocal cords are engaged. The researchers did an acoustic analysis of the ‘d’ and ‘t’ sounds at the start of each word to measure when the vocal cords engaged.
Overall, speakers did differentiate between the languages. They engaged their vocal cords later when speaking English words than when speaking Spanish words. The language of the previous trial did not affect pronunciation of Spanish words, but it did affect pronunciation of English words. Overall, participants engaged their vocal cords a bit earlier when saying an English word if the previous word had been spoken in Spanish than if the previous word had been spoken in English. That is, the person’s Spanish accent was stronger if they had just said a Spanish word than if they had just said an English word.
One interesting aspect of this study is that some of the words used were cognates. That is, the words for the object were similar in Spanish and English. For example, the Spanish word for dentist is dentista. The effect was particularly strong for these cognates. That is, when participants had just spoken a Spanish word and then they had to speak the English word for a cognate, they had a stronger accent than when they had just spoken an English word.
This result reflects that when people speak multiple languages, they learn information both about the words that are used in that language as well as how to pronounce those words. For cognates in particular, people have experience speaking similar words in both languages. Speaking Spanish activates the Spanish pronunciation of the word, while speaking English activates the English pronunciation. When participants switch languages, they get a combination of these pronunciations, which ultimately affects how they speak those words.