However, I'd say a lot of follow-up is needed before this can be considered a "conclusion".
First of all, the study was conducted with a grand total of 60 participants, all of whom were college students in Israel of the same socioeconomic characteristics: 20 native-Hebrew speakers, 20 native-Arabic speakers who started learning Hebrew around age 7, and 20 native-Russian speakers who started learning Hebrew around age 13. Enough for a hypothesis for further testing, perhaps, but conclusion? I don't think so.
In any case, here's what they had to do:
All were asked to read out a section from a report in Hebrew, and then to describe – in Hebrew - an image that was shown to them. The pieces were recorded and divided into two-minute sections. Additionally, the participants filled out a questionnaire that measures empathetic abilities in 29 statements.Once that was done, 20 separate native speakers rated how heavy the accents were.
Both Russian and Arabic speakers were shown to have similarly heavy accents. However, the more empathetic the Russian speakers were, the weaker their accents. For the Arabic speakers, however, their level of empathy didn't affect the heaviness of their accents.
From this, the professors of the Israeli university conclude of the Arab students that:
[T]he pattern among Arabic speakers demonstrates their sentiment toward the Hebrew-speaking majority group, and the former consider their accent as something that distinguishes them from the majority.Color me skeptical.
I'd be more accepting of these results if, instead of Russian speakers, they had used Arabic speakers that didn't carry the same sociopolitical baggage—Moroccan Jews, or something like that. As is, this could perhaps show nothing more than the fact that there's some connection between empathy and accent in Russian speakers.
I'd love to see this test repeated where they could get larger, more diverse samples, like in the U.S. or Canada. Would people from locales with widely varying cultures—e.g., Latin America, Japan, India, Croatia, etc.—show the same connection between empathy in their English accents? A study like that would be much more interesting; first establish the connection between empathy and accent, then move on to dissenting other things that might affect it.
Steve focuses on some of the findings' applications to language learners:
To learn a language is to imitate some of the behaviour of another group. You have to act as if you are one of them. . . . It is not enough to practice making certain sounds. You have to dive into the role of being someone else. You have to feel comfortable doing it. That means not translating comfortable sayings, and turns of phrase from your own language. It means forgetting who you are, linguistically and culturally.If this position is right (and my inclination is that it is), it would seem to make sense that actors would be great language learners. I've never heard of any studies looking into that, but Brad Pitt's language-learning abilities don't bode well for the hypothesis.
While I do think their hypothesis is probably onto something, I think it's likely a lot more complicated that "the more the empathy, the better the accent". Mitch, in the comments on Steve's blog, points out a big hole in the hypothesis:
There were prisoners of war and camp inmates who learned German well enough to pass as natives after escape. They probably had zero empathy for their hated guards, but their motivation was incredible--survival.So clearly there are cases of zero empathy, but great accents.
What Makes an Accent in a Foreign Language Lighter? [Newswise]
Accent and empathy [The Linguist on Language]