Recently a friend, who I’ll call Patricia, described to me a scene which she felt portrayed a specific divergence in non-verbal communication. If you like, before I write the details, imagine the following situation:
A friend is picking you up, and you see that she has already gotten there and is seated in her car in anticipation of your arrival. → How does your body express itself as you approach her?
Patricia, the driver, told me that she had noticed that a Japanese woman walked in quick, short steps, shoulders slumped forward, with her head down. In contrast, another Japanese woman strode toward her, head up, back straight, looking directly at her. To Patricia, the first woman seemed more Japanese in her bodily manifestation, and she interpreted the woman’s posture, gait, and gaze as a conveying graceful, wordless apology at making her wait that felt culturally Japanese compared to the other woman’s body language.
Just like when I go shopping for something, like a tweed jacket, and don’t find one but come home satisfied nonetheless with a different blazer, research expeditions triggered by real-life observations often lead not to research directly related but to topic-adjacent studies, which are likewise pleasing. And, of course, I continue to keep an eye out for that “tweed jacket”.
I love Patricia’s observation and I agree with her reading of the woman’s (probably unconscious) intention. Unfortunately, I couldn’t find any research related to the expression of speech acts like apologies, gratitude, requests, or refusals via gait and posture, but I did find an interesting study on the interpretation of various body stances in different cultures. Andrea Kleinsmith is a specialist in human-computer interaction, with a focus on ways of modeling and measuring non-verbal communication. She and two colleagues, P. Ravindra De Silva and Nadia Bianchi-Berthouze, conducted a study on cultural divergence in the evaluation of different types of body postures that was published in Interacting with Computers. Their study reveals interesting distinctions between Americans, Japanese, and Sri Lankans.
The researchers asked actors to wear a specified suit with 32 markers attached to it in posture-generating positions, such as joints and body segments. Then they were told “perform an in-place posture” as if they were feeling each of the following emotions: anger, fear, happiness, or sadness. The postures were captured by eight cameras and in total the researchers obtained 108 postures expressing these emotions. The data was used to create faceless avatars, all with the same body. These avatars were shown to American, Japanese, and Sri Lankan research participants, who were shown one at a time randomly and asked to choose which among eight feeling was represented: angry (okotta), upset (nayanda), fearful (osore), surprised (odoroki), happy (tanoshii), joy (yorokobi), sad (kanashii), or depressed (yuutusuna).
Kleinsmith and her colleagues first compared to what degree the three cultural groups of respondents agreed with the actors regarding the emotion portrayed. (Of course, it’s impossible to say the actors were correct in their interpretations, but their intuitions of how to express these emotions served as a base to compare how much each cultural group agreed with these depictions.) Across the board, the agreement with the actors’ intended posture was highest for the combined frequencies of sad (merging “depressed” and “sad”), with at least 60% of the members of each cultural group concurring with the representation the actors aimed at, followed by happy (adding together “joy” and “happy”), with at least 54% agreement in each group. But even for the conflated totals of the remaining two emotions, there was solid agreement for “fear”/ “surprised” (at least 44%) and ”angry”/”upset” (at least 45%).
Nor was there much of a divergence between groups in their frequencies of agreeing with the actors’ intentions: for sad there was only a 5% gap, for happy 6%, and for both fear and angry 11%. There were, however, two significant differences in the sub-categories of the merged emotions. While the Sri Lankans and Americans were just about as likely to call the anger postures “angry” or “upset”, the Japanese observers were almost three times as likely to interpret the postures as “angry”. On the other hand, while the Japanese and Americans did not have much of a difference in their slotting between ”fear” and “surprised” for the designated fear postures, twice as many Sri Lankans called it “surprised” rather than “fear”.
Then the researchers got down more to the nuts and bolts, examining each group’s agreement with the precise markers of the representations. Using a Mixture Discriminant Analysis (MDA), Kleinsmith and her colleagues were able to isolate 17 features that were most significant in discriminating the emotions for each cultural group. Only three of these features were salient in all three cultures. The Japanese and Sri Lankans shared three, Sri Lankans and the Americans had another three in common, four were only pertinent for the Japanese, one for only the Sri Lankans, and three only for the Americans.
Unfortunately, the researchers didn’t explain all of these difference, but here are some of the interesting findings:
Arms extended along the body coupled with a bent head were the most salient expressions of sadness for the Japanese participants. While the bent head was also an important element for the Americans and Sri Lankans, the arm along the body was not.
Conversely, the Americans and Sri Lankans perceived a posture with the face slightly raised and the hands close to the chest as demonstrating sadness, but the Japanese did not.
Extended arms were one aspect of posture that the Japanese interpreted as happiness. Arms raised to mid-level could also play a role in expressing happiness or fear for the Japanese, but the Sri Lankans perceived arms extended down or raised mid-level as angry postures.
On the other hand, for the Sri Lankans, fear was always associated with the elbow lower than the shoulder, and very low for sadness.
The Japanese equated distance between heels with fear, but the Sri Lankans viewed it as indicative of anger.
Keeping someone at arm’s length, sticking your neck out, keeping your chin up, giving someone the cold shoulder—our language attests to the role of posture in emotive communication, but ultimately body is in the eye of the beholder.
I had to dig out my copy of Kiss Bow Or Shake Hands just to follow along! You must love that book.
Yes it seems like culture and not just language makes it really hard to fit in to this society. Interesting!