Sociolinguistics

Hedging and Gender in the STEM Community

Eric Chen, Abbey Mae Gozon, Khoi Nguyen, Paul Vu, Julia Wang

Hedging is an aspect of language that is easy for non-linguists to overlook. These terms are used to apply uncertainty to a statement, to make it seem less assertive. The question we seek to answer is, do women make more use of hedging than men do? Specifically, we seek this in the context of an environment where more is expected of women than of men. In this experiment, we take a look at the presence of hedging in the speech of female STEM students. These participants participate in interviews about the subjects they study, and then afterwards take a short survey in order to determine what it is that the participants believe is the root cause of their own hesitations. The recordings of the interview portions are scanned for hedges that are measured as uncertainty in the participant’s explanations. A numerous presence of which would imply that the speaker is not completely sure that they are correct and are choosing to leave room for themselves to err. This study intends to find out whether or not women hedging more than men contains more substance than is implied by the stereotype.

[expander_maker id=”1″ more=”Read more” less=”Read less”]

Introduction + Background

Entering new environments with high expectations can be difficult for anyone. This is especially relevant in academic environments, where imposter syndrome and the resulting stress are abundant. As a general stereotype and resulting from years upon years of patriarchal oppression, it is believed that women hedge more than men. This also carries the implication that women are less intelligent. Now we know from a lot of other research (and a thing I like to call common sense) that simply being a woman does not make a person less smart than anyone else. Though why are we looking at female STEM majors in particular? The number of women in stem has been far less than the number of men in the field for many years in our past. In recent years, many more women have been joining the STEM field, choosing to become STEM majors. Their numbers have been steadily increasing every few years from what it was in the past, but also has not been increasing fast enough to be considered a “boom” of any sort. Does the high expectation for the small percentage of women entering these male-dominated fields make them feel less confident in their abilities to acquire/distribute STEM-related information? Could the use of more hedging be intentional, and used as a sort of cushion for being able to make mistakes? It is also possible that the content within STEM fields, which is generally considered difficult to grasp, may make it harder for people to assert their knowledge of the subject. This semi-spontaneous interview test attempts to collect data that can be used as evidence to answer these questions.

Methods

To collect our data, we used a sample of twelve STEM majors at UCLA: six male, six female. We conducted a short interview, asking questions that would illicit hedging. The questions were:

    1. What is one STEM class that you’re currently taking?
    2. Can you explain something that you’re learning in that class?
    3. What is the most difficult thing you have learned in that class?

These questions were designed to provoke deeper thought and test mastery of the participant’s field of study. After the interview, we then explain to the interviewees our study and hedging, specifically what it is and how someone might use it in certain scenarios. We give this explanation to allow them to reflect on the subject and determine how much they think they use it and why they use it as a linguistic tool. We also do this after conducting the interview so the participants are unaware of the topic, giving us the most genuine, unaltered responses. Finally, we give them a post-interview survey to answer based on their recent reflections on a scale from 1- 10. The questions include:

    1. How often do you think you use linguistics hedging?
    2. How confident or capable do you feel in your field of study?
    3. How often do you feel condescension or face condescending remarks said to you in your field?
    4. How much do you believe the environment you face (and the amount of discrimination / condescension) in your field has contributed to this your confidence in said field?
    5. How much do you believe the level of confidence you have affects the number of hedges you use?

With this whole study, we are able to collect genuine responses of hedging from the interview and perspectives on hedging from the participants with the survey.

Results

Each interview from the twelve STEM majors lasted around two to three minutes. We noticed that the STEM majors commonly used hedges: “like”, “possibly”, and “may”. From the twelve interviews, we noticed that the most common hedge used by both genders was the word “like”. The STEM majors used the word “like”, not to show comparison, but to express vague statements. For example, one of the interviewees said:

“… Learning how you compose and create a CT scan from like Fourier transforms …”

Here, the interviewee used hedging to evasively state that she was learning how CT scans are created through Fourier transforms.

To analyze the interviews, we counted the occurrences of hedges in each interview and calculated the average frequency of hedges per gender.

Figure 1: Hedges Counted per Interview

The data revealed that females had an average of 6.83 hedges per interview while males had an average of 8.83 hedges per interview. Although females had a lower average of hedges than males, we noticed that the amount of hedges for both males and females were fairly similar with the exception of a few outliers. The maximum amount of hedges used was 18 hedges by a male, which is much higher than the amount of hedges counted for the other males. The minimum amount of hedges used was two hedges by two females. Because of the outliers and the fairly similar amount of hedges, it is hard to conclude that gender causes a change in frequency of hedges. Our data suggests that it is possible that the frequency of hedges is related to the individual’s competency in the subject rather than gender. To have more conclusive results, we should have interviewed a large amount of people, but because our sample size was too small and hedging counts were fairly similar, it is hard to definitely conclude that gender affects hedging usage.

When we analyzed the questionnaire data from our post-interview surveys, we discovered several notable observations. One of those observations was that women felt significantly more adversity than men. For the question “how often do you feel discriminated against or underrepresented in your field?”, we found that women felt approximately 7 times more discrimination than men. Similarly, for the question “How often do you feel condescension or face condescending remarks said to you in your field?”, we observed women feeling about 3 times more condescension than men.

Figure 2: Perceived adversity by men and women

From those two questions alone, the data suggests that gender disparity continues to exist and has propagated into the UCLA community as well. Specifically, it seems that how much women experience discrimination today has not changed enough especially when it comes to factors such as earnings and promotions in the workplace.

Furthermore, when we analyzed the data from the question asking how confident they felt in their field of study, we observed that women felt more confident than men by a slight margin of 0.5. However, the data from the “How much do you believe the environment you face (and the amount of discrimination / condescension) in your field has contributed to your confidence in said field?” showed that women gave less credit to their environment. Surprisingly, the data suggests women believe they developed their confidence outside of the environment in their respective fields. This begs the question, if this is the case then where exactly are they getting their confidence from and why is this the case?

Discussion and Conclusions

Our hypothesis was inconclusive that women in stem hedge more than men since the data that we received was inconsistent. Even though women in the survey reported a higher frequency of using hedging the frequency of hedges in the interviews conducted was around the same. Therefore, we cannot conclude from our study that women in STEM hedge more because of under representation and lower confidence levels in their field. However, there are many parameters that could influence our findings such as our small sample size, as there were only 12 people interviewed and surveyed in total. Another factor is that the people we used in our experiment were people we were familiar or friends with thus they might have been more comfortable around us, which could influence the frequency of hedges they used.

In the future we could possibly recreate the study but with a larger sample size which would even out the outliers in our data. We could also sample random people which we have not met previously. Furthermore ,we could research the differences in hedging between gender in different majors of the STEM field, such as computer science, mathematics, physics, etc, and observe whether the field you are in can affect the difference in hedging frequencies of men and women. We could also conduct the study among people interested in STEM of different education levels, such as in high school, and different colleges.

Through examination of hedging we can have a better understanding of the effects of gender in the STEM fields and language use. A continuation of this study could be very beneficial to stem majors when considering the role of gender in the stem field and the level of confidence they portray. In particular this research can be important when considering the work force, specifically that women are less likely to ask for raises and promotions. This could be tied into hedging since hedging relates to being uncertain and less confident in one’s ideas. We believe that this is a very important area of research with a lot of potential to explore the effects of gender in STEM and look forward to future contributions regarding this topic.

 

[/expander_maker]

 

 

Fun, Cool, Hip Title Here: AAVE Usage in Twitter Memes

Nick Ushiyama, Stella Oganesyan, Ava Boehm, Rachel Lee, Alesha Vaughn

Love them or hate them, almost everyone active on social media has come into contact with memes at some point. Chances are, one or more of those memes used a variety of English called AAVE, or African American Vernacular English. This variety originated from working-class African Americans and displays words (lexicon), word order (syntax), word pronunciation/spelling (phonology), and word combination (morphology) different from the Standard American English (SAE) taught in schools (Rickford et al., 2015). In our study, we tried to better understand how and why meme-makers switch between AAVE and SAE in their posts. We expected meme-posting Twitter users to use switching as a way to signal to readers that their posts should be read within the unique guidelines of meme-culture humor. For our research, we collected hundreds of memes and distributed a survey to see how people interpreted the switches. The results confirmed our expectations.

[expander_maker id=”1″ more=”Read more” less=”Read less”]

Introduction and background

A meme is a piece of cultural information that holds certain ideologies or behavioral concepts and is transmitted from person to person. The word ‘meme’ stems from the Greek word ‘mimeme.’ The root mim- means to mimic and the English suffix –eme is used to imply a unit of linguistic information, as seen in words such as phoneme and lexeme. The term ‘meme’ was coined from ‘gene’ and similar to a biological gene, the nature of a meme is to mutate or replicate when being transferred from person to person. The world of social media is full of memes as they are seen as a major part of today’s popular culture.

We noticed that a good handful of popular memes contain AAVE regardless of whether or not the original poster was a member of the Black Community. These memes were quite popular, too, which makes the use of AAVE within memes apparently index ‘coolness’ or ‘hipness’. There also wasn’t just one part of AAVE that memes utilized, but instead integrated syntactic, lexical, phonological, and morphological aspects of the dialect.

Figure 1a: An Example of AAVE Switching Involving a Syntactic Feature (“he b getting yelled at”)
Figure 1b: An Example of AAVE Switching Using a Lexical Feature (“the class was wildin”)

AAVE has been studied pretty extensively by linguists in the past. Of the studies that are relevant to our project, most of them show different ways that AAVE contributes to identity. That is, they show that people use it to communicate things about themselves to others. Those things could be anything from membership in social groups (Rickford et al., 2015; Anderson, 1999; Labov 1973; Sweetland, 2002), to particular attitudes (Ilbury, 2020). However, almost none of this research looks at AAVE on social media, let alone in Twitter memes.

For youths, social media is quickly becoming one of the richest sites for creating cultural connections. As such, the linguistic norms that are founded there can quickly become widespread. Our work addresses this understudied, but extremely significant, domain of AAVE usage. We set out anticipating that meme-creators would incorporate AAVE in their posts to tell readers that those posts should be read and interpreted as memes.

Methods

Occurrences in Memes

Before we tested our hypothesis, we first had to figure out what kinds of switches were occurring between SAE and AAVE. As such, we collected instances of AAVE usage in memes by visiting meme-posting pages on Instagram. We recorded whether these AAVE features were syntactic, lexical, phonological, or morphological in nature, and we also considered what topics the memes addressed. Figure 2 below explains which topics we observed.

Figure 2: Topic List and Definitions

Survey

Upon gaining a lay of the land, a survey was designed. We sent it out in order to receive data that would allow us to address our hypothesis. In the survey, participants first provided consent to publish their (anonymously-attributed) data. They then stated their age and level of experience with memes.

Following this demographic collection portion, the participants were exposed to examples of memes in which one AAVE feature (and therefore one switch) was used. They were then asked…

    • Whether they believed the usage of AAVE was ironic (disingenuous) or not given a poster’s race (African American and non-African American).
    • What they believed the posters were trying to do by switching from SAE to AAVE
    • Whether they believed their answer to (2) would change if the poster’s race was the opposite of that presented in (1).

At the end of the survey, we asked them to respond to the following question if they had identified any switch as ironic: “If you said that some usages were ironic, do you think that irony is meant to indicate something about how the humor in the posts should be interpreted?” This allowed us to directly address our hypothesis.

Results/analysis

Occurrences in Memes

The meme data consisted of the type of linguistic feature involved in the switch from SAE to AAVE and the topic that the meme addressed. We calculated the number of occurrences for syntactic, lexical, phonological, and morphological features per topic, and the results are presented below in Figure 3.

Figure 3: Raw Number of AAVE Occurrences per Topic

We then calculated the percentages of each occurrence per topic, and these results can be seen below in Figure 4.

Figure 4: Raw Percentages of AAVE Occurrences per Topic (NOTE: overall here means all topics combined)

In terms of the overall number of AAVE features observed, the data showed a clear preference for AAVE syntactic features, followed by lexical, phonological, and morphological features. This order of preference occurred in 4 out of the 11 identified topics. The most popular topic of memes was daily routines, while the least popular was the occupation topic.

From our survey, we received a total of twenty seven completed responses. Twenty six participants were in the age range of 19 – 29 (approximately the same age as meme-posters), and one older participant (age 38) was also included in the data given their experience level with memes. Out of the four possible meme experience levels, only 3 were observed (options 2, 3, and 4). Figures 5a and 5b summarize their meme experience:

Figure 5a: Breakdown of Participant Meme Experience
Figure 5b: The Meme Experience Levels we Observed

We then averaged irony scores for each example among meme-experience groups, age groups, and overall. The irony score represented how strongly participants believed the poster’s switch would occur as a natural tendency as opposed to a conscious choice. Except for the fourth example (which participants did not view as having a switch at all), irony scores were greater when the poster was assumed to be non-African American. The meme-experience group who chose option three had higher irony scores than those who actually made memes. That said, this difference was not statistically significant according to an F-test and a ttest between the two groups. This data can be seen below in Figures 6a and 6b.

Figure 6a: Irony Score Per Participant Age
Figure 6b: Irony Score Per Meme Experience Group

We then analyzed short answer responses, which consisted of what participants believed switches indicated about the humor of the examples. We boiled down their statements into ‘themes’ of explanation and counted how many responses fell into these themes. We specifically focused on themes relating to humor and noted how strongly these were represented among the three present meme experience levels. A summary of the response data can be seen below in Figures 7a-e.

Figure 7a: Short Answer Data Summarized – How Many Different Themes (Dispersion) and How Many Rejected Responses (n/a portion)
Figure 7b: Short Answer Data Summarized – Ratio of Humorous Themes to Total Entries Under Varying Poster-Race Assumptions for Each Example
Figure 7c: Short Answer Data Summarized – Different Meme Experience Levels’ Ratio of Humorous Themes to Total Valid Entries for Each Example (NOTE: red cells are option 4 group, white cells are option 3 group)
Figure 7d: Short Answer Data Summarized – Mode (Most popular Theme) and Values of Mode For Each Example
Figure 7e: Short Answer Data Summarized – Disagreement in What Switches Meant for Each Example

As seen in Figure 7e, we calculated the degree of disagreement on what switching meant for each example. Generally, there was less disagreement when participants were told that the poster was not African American, and overall disagreement increased in later examples considerably.

Finally, we sorted responses to the final question, regarding what ironic switching was meant to indicate about how humor should be interpreted. Not every participant was instructed to answer this question, only those who indicated that ironic code-switching to AAVE was present in the previous examples. Out of the 21 responses that were eligible, 85.71% of participants believed ironic switching indexed something about how the humor of the meme should be evaluated.

The most popular response was a positive confirmation of the question. The most popular elaborated response stated that switching to AAVE signaled to read the post as a meme. To be read as a “meme” is best explained by one participant’s response:

“Yes, I believe that switching to AAVE shows to users that it is not a formal post but instead casual, humorous, and meant to be related to.”

Discussion and conclusions

Our most significant finding was the general consensus that meme-posters use AAVE to indicate how the humor in their posts should be interpreted. And indeed, our hypothesis was confirmed: participants directly stated the switch to AAVE was done so the humor of memes would be evaluated along comedic standards specific to memes (as opposed to stand-up or sketch comedy). This would suggest that AAVE has become associated with humor. And to be sure, there are negative consequences to this association. The variety could be portrayed as something humorous, lighthearted, and not to be taken seriously. One of our participants in fact commented that AAVE’s appearance in memes is justified because “certain vernacular have a playful connotation that doesn’t imply seriousness.” Obviously, this would pose a problem for those who use the variety in their daily lives, in that their speech would be trivialized and even seen as unfit for participation in larger economic and civil institutions.

Our raw data also suggested that neither age nor meme experience significantly affected the likelihood to see irony in AAVE usage. At least one of our examples however was flawed and may have skewed the data. And in fact, given final question responses, it’s likely that being in the higher meme experience group did make participants slightly less likely to view switching as ironic. It is tempting to draw the conclusion that greater experience with memes in turn translates to lower likelihood to view AAVE usage as cultural appropriation. One participant in the option four group actually recognized that SAE to AAVE switching could constitute appropriation. However, they also noted that it is unlikely that there are ill intentions around the usage itself. They believed that though meme culture may inadvertently stigmatize the variety, the community itself is not systematically “anti-black.” All of this said, we cautiously state here that the negative consequences of AAVE usage in memes do not escape some members of the meme community but also that they don’t view their actions as malicious. As such, it’s unlikely that AAVE usage will cease any time soon.

The greatest number of AAVE features found in memes were syntactic features, the first three survey examples (containing the two syntactic switches) displayed greater numbers of humorous entries, and these first three also included lower levels of disagreement towards the meanings of switches. This suggests that AAVE syntax is not only more heavily associated with memes but is also the most used type of feature in communicating information about humor. And indeed this aligns with what Sweetland (2002) claimed regarding AAVE usage: AAVE syntax was the primary means of linguistically indicating a belonging to the AAVE speech community. Meme posters are arguably not, however, trying to indicate belonging to the AAVE speech community, so there are two likely implications this finding could have. Perhaps the users are trying to imitate and evoke stereotypes regarding African Americans. Conversely, the users could be attempting to signal in-group status of their own. That is, they could be trying to say “I’m a member of the meme community, too!” by switching. We make no conclusions here since we lack evidence to prove either, but leave readers with the understanding that, regardless of humor, there are real world consequences to this type of usage.

 

References

Anderson, E. (2000). Code of the street: Decency, violence, and the moral life of the inner city. W. W. Norton & Company

Ilbury, C. (2020). “Sassy Queens”: Stylistic orthographic variation in Twitter and the enregisterment of AAVE. Journal of Sociolinguistics, 24(2), 245-264. doi:10.1111/josl.12366

Labov, W. (1973). The linguistic consequences of being a lame. Language in Society, 2(1), 81- 115. doi:10.1017/s0047404500000075

Rickford, J. R., Duncan, G. J., Gennetian, L. A., Gou, R. Y., Greene, R., Katz, L. F., Kessler, R. C., Kling, J. R., Sanbonmatsu, L., Sanchez-Ordoñez, A. E., Sciandra, M., Thomas, E., & Ludwig, J. (2015). Neighborhood effects on use of African-American Vernacular English. Proceedings of the National Academy of Sciences, 112(38), 11817–11822. https://doi.org/10.1073/pnas.1500176112

Sweetland, J. (2002). Unexpected but authentic use of an ethnically-marked dialect. Journal of Sociolinguistics, 6(4), 514–538. https://doi.org/10.1111/1467-9481.00199

 

[/expander_maker]

, ,

Speech Patterns as Identity Constructors Across Social Media Platforms

Alissa McNerney, Akina Nishi, Ryley Park, Nicolas Simone, Fontanna Yee

As slang and social media usage has risen in popularity in recent years, we wanted to explore how different patterns of slang would change a speaker’s identity on different social media platforms. Although we initially thought that examining slang alone would give us a good picture of how social media identities were created, we soon realized that slang usage was part of the story, but not entirely dependent on the social media platform. This discovery allowed us to pivot towards analyzing not just slang but also how prosody, speech-related information such as intonation and gestures, also contributed to identity construction. By conducting a case study of TikTok influencer @sirthestar across three social media platforms, TikTok, YouTube, and Twitter, and analyzing both written and spoken content, we concluded that greater usage of slang and prosody contributed to creating a more comedic identity on TikTok and YouTube and lesser usage contributed to a more social activist identity on Twitter.

[expander_maker id=”1″ more=”Read more” less=”Read less”]

Introduction and Background

Slang has long been an integral part of colloquial speech and plays an important role in social communication especially among younger people, as younger generations seem to be the greatest users and creators of new slang words (Zhou, 2013). Currently, a visible linguistic trend is the frequent usage of slang usage on social media platforms, both because of the younger demographic and because the informal dialogue on social media translates well to the nonstandard forms of slang (Teodorescu, 2015). Oftentimes, different subgroups end up creating their own slang usage (Zhou, 2013), and our research question centered around how slang would define certain subgroups on social media. More specially, we asked how users constructed different identities through their linguistic variation on different social media platforms. Initially, we hypothesized that a user’s identity could be influenced by the particular slang terms they used more frequently on a social media platform, and that slang usage would be different depending on the platform it originates from. However, after initial data analysis, we pivoted to include prosody in our analysis, encompassing non-lexical speech-related information such as pitch, intonation, articulation, and gestures, which contributed equally as much to identity construction as slang (Shih & Kochanski, 2002). Our final revised hypothesis was that a user’s identity could be influenced by the varied prosody and other speech patterns, as well as frequency of particular slang terms, they used on each social media platform.

Methodology

For this research study we conducted a case study of TikTok influencer @sirthestar by examining his written and verbal communication on his TikTok, Twitter and Youtube accounts. Tiktok is a video sharing social media platform where users create a wide variety of different types of videos, with over two billion downloads across the world (Zukin, 2020). Like other forms of social media, TikTok has been the origin of new slang terms that have expanded in usage beyond the platform, especially because internet slang can develop and spread quickly because of the viral nature of online content (Zhang, 2016). We decided to analyze the speech patterns of only one person to minimize non-linguistic external factors that could affect the user’s identity. Although most well-known for storytimes on TikTok, Sir is also active on both Youtube and Twitter, which made him an ideal candidate.

We searched for both written and spoken samples from these social media platforms, and using our slang references, analyzed linguistic variation in each of these platforms. We collected data by using the closed captioning system in YouTube, written speech on Twitter, and self-captioning and our own transcriptions of TikTok dialogue. Before pivoting, our intention was to categorize the types of slang we found, such as alphabetisms, blends, clippings, and reduplicatives (Kulkarni & Wang, 2018). However, for Youtube, we could not find enough video data to make any conclusions about slang usage, and when we parsed the data of Twitter and TikTok to find significant slang terms, we saw inconclusive results because slang alone was not enough to distinguish between Sir’s identities on TikTok, Twitter, and YouTube.

Figure 1: Bar Graph of Frequency of Slang Usage (TikTok vs Twitter)

Because of this, we switched gears to include more than just slang as part of our research. With a new direction, we looked at each platform’s data and divided the samples into “comedic” and “non-comedic” content to take a look at the tone presented by Sir. We determined this by identifying if a tweet or piece of dialogue from YouTube or TikTok was meant to elicit laughter and be entertaining or not, and then identified changes in slang, prosody, and other speech patterns that contributed to creating “comedic” or “non-comedic” content on each social media platform.

Results

Twitter

Figure 2: Non-Comedic Tweet. Sir takes a stance on Black Lives Matter.
Figure 3: Comedic Tweet. Sir shares comical experience.

Twitter showed us a pretty even split between comedic and non-comedic tweets. This was mostly due to Sir’s numerous tweets on Black Lives Matter, and his purpose of spreading awareness. For “comedic” content, Sir tweeted with themes of relatability and silly snippets of his life. As for speech patterns, comedic tweets had simple words and were only a few lines long. There was not conclusively a disregard for grammar, because there was still proper capitalization and usage of punctuation like “…” or “*scrolling twitter*”, but it was fashioned casually to be humorous. Meanwhile, non-comedic tweets were usually more than one line long, and had more proper punctuation and capitalization. Some common themes included social activism, self-esteem/positivity, and gratitude towards his fans. An example is his tweet on September 17- “Something I’ve learned…don’t search for or force love, it’ll only hurt you in the long run. Focus on loving yourself, be in your bag. This creates a positive energy which attracts positive relationships ✨” This is not to say, however, that these kinds of rules applied to strictly all of Sir’s tweets, as we saw that both comedic and non-comedic tweets included use of emojis, like in the above examples, and use of all-caps written speech to emphasize emotion.

YouTube

Figure 4: Screenshot of one of Sir’s comedic videos, as shown by facial expression

On YouTube, we noticed that there was an imbalance between comedic and non-comedic videos uploaded. While Twitter was a platform that @sirthestar utilized to voice his opinions on more serious matters, YouTube was a platform where we regularly saw @sirthestar telling stories for entertainment. His language contains many slang words, and his speech pattern of repetition appeared frequently throughout his videos. One example was a quote from a video on July 24 – “Ain’t nobody gonna cheat on us, cheat on us, treat us like sh*t, none of that, none of that. What we gotta do is we play the game, we play the game, and if they think they win. NO! They didn’t win, tie them up!” This was most likely due to the fact that YouTube videos were much longer than the time duration given on TikTok, and therefore @sirthestar was able to take his time to relay his messages with more humor. However, this longer time duration also appeared as a hindrance in terms of collecting data. While TikTok and Twitter content did not contain too many words, the YouTube videos had twelve to fifteen minutes worth of spoken language, which made collecting data from multiple videos too difficult. Overall, we could see that the repetition of certain phrases that appeared on YouTube was not present on the other two platforms.

TikTok

Figure 5: Pie chart of ratio between comedic and non-comedic TikTok videos

TikTok was where Sir also had a large majority of comedic versus non-comedic videos. Of the thirty TikToks that we examined, there were only a few that focused on serious topics like politics, racism, and homophobia, but even in these videos, Sir maintained his comedic persona. This persona was evident through his use of slang and prosody in his videos. On TikTok, Sir used the most slang by far, with words like “bitch” and “y’all” being used frequently in his videos. Along with this slang, Sir would commonly clap and stress certain words in his TikTok videos for emphasis on his jokes. An example of this was in a TikTok where he said “ Mind you, I already had a long ass day, and getting cat-called was not gonna end good for him [clapping between each word].” Sir’s use of prosody and articulatory gestures were present in the majority of the TikToks we analyzed, and unlike other platforms, there was plenty of slang usage in the majority of his TikToks. TikTok was also the platform where Sir used the most stream-of-consciousness type of speaking by utilizing run-on-sentences and speaking quickly, which emphasized that every video was a performance, and he maintained a very exaggerated, entertaining persona even when he was speaking on a sensitive topic. These linguistic features contributed to his comedian persona that he developed on the application in a different way than his other social media platforms.

Conclusion

Our findings were able to help us understand how one’s identity differed depending on which social media platform they were utilizing. By researching @sirthestar’s profiles on Twitter, YouTube, and TikTok, we saw that he used Twitter for more non-comedic content, and thus there was proper grammatical usage, punctuation, full sentences, and less slang, although there was still usage of emojis in both comedic and non-comedic content. His speech patterns on Twitter proved to be more formal and demonstrated that it was a place where he could be more serious and voice his opinions on social activism. On the other hand, on YouTube and TikTok, @sirthestar uploaded content for entertainment purposes. After seeing the higher use of slang and prosody in YouTube and TikTok, we were able to conclude that he has a comedic persona on these platforms, although the persona was created in different ways on each platform. On YouTube, his comedy came from word repetition, partially because of the longer timeframes, whereas on TikTok, he had the higher frequency of slang words as well as articulatory gestures. Although we had to pivot about our original hypothesis, we learned that it wasn’t completely wrong. Slang usage is not dependent on the social media platform but rather plays a role alongside prosody and other speech patterns in developing a comedic persona on YouTube and TikTok.

 

References

Kulkarni, V., & Wang, W. Y. (2018). Simple models for word formation in English slang. arXiv preprint arXiv:1804.02596.

Shih, C,. & Kochanski, G (2002). Section 1: What Is Prosody? Prosody and Prosodic Models, www.cs.columbia.edu/~julia/courses/CS4706/chilin.htm.

Teodorescu, H. N., & Saharia, N. (2015, October). An Internet slang annotated dictionary and its use in assessing message attitude and sentiments. In 2015 International Conference on Speech Technology and Human-Computer Dialogue (SpeD) (pp. 1-8). IEEE.

Zhang, L., Zhao, J., & Xu, K. (2016). Who creates trends in online social media: The crowd or opinion leaders? Journal of Computer-Mediated Communication, 21(1), 1-16. doi:10.1111/jcc4.12145

Zhou, Y, and Fan, Y. “A sociolinguistic study of American slang.” Theory and Practice in Language Studies 3.12 (2013): 2209.

Zukin, M. (2020, Aug 05). TikTok Age of in the Quarantine. Variety, 46-49. https://search.proquest.com/docview/2434859387?accountid=14512

[/expander_maker]

Does She Listen to ‘Girl in Red’? Linguistic Markers in WLW Flirting

Tiffany Dang, Brianna Lombardo, Carlos Salvador Vasquez, Denisa Tudorache, Yuyin Yang

The present article focused on linguistic markers that are adopted by the Women Loving Women (WLW) population when identifying potential members of the WLW community. More specifically, this study focused on the strategies used by members of the WLW community for identifying fellow WLW with the intentions of pursuing a romantic or sexual relationship. Through analyzing popular YouTube videos featuring strategies on flirting with WLW, our first study captured the common beliefs regarding the need to take an extra step, and the possible methods on identifying WLW before taking any romantic or sexual advances. Followed-up by semi-structured interviews in study two with UCLA students who self-identify as WLW, we were able to examine the accuracy of the tips offered by the YouTube videos. This allowed for further investigation on the existence of specific linguistic markers adopted by WLW when flirting. We found that both popular YouTube videos and participants both discussed the need for WLW to take an extra step before they can comfortably pursue another woman and tend to make a conscious effort to not be too direct.

[expander_maker id=”1″ more=”Read more” less=”Read less”]

Introduction and Background

While there have been past studies done on examining the speech of gay men, particularly the California vowel shift among gay men (Podesva, 2011), and one that revealed a concept of gay-dar, the belief that gay men possess an ability to pick out each other in a crowd (Shelp, 2003), little research has been done on uncovering linguistic patterns within the Women Loving Women population (WLW). A member of the WLW community is loosely defined as anyone who identifies as a woman and differs from the mainstream preferences in terms of their sexual practice and identity (Eliason & Morgan, 1998). Due to being seen as deviant from the mainstream practices, they may feel the need to take different approaches when making romantic pursuits in order to establish a mutual understanding of their interest in women when talking to another individual. As WLW may often struggle with compulsory heterosexuality, the fear of being perceived as predatory, as well as the potential dangers that come with revealing their sexuality, we aimed to investigate whether there were any linguistic markers adopted among the members of the community to aid in implicitly seeking each other out. This study explores the ways WLW work around the potential barriers they face when pursuing romantic interests and when revealing their identity in hopes of gaining insight on ways to improve the inclusivity of a general community. We hypothesized that WLW would adopt practices where they refer to certain WLW-group-specific terminologies or features before making romantic or sexual advances towards another woman.

Methods

Study 1 collected people’s lay knowledge on identifying WLW by looking at popular YouTube videos that featured strategies on how to initiate romantic/sexual advances with a WLW. We found three relatively popular videos created by members of the WLW community who also covered a large realm of dating advice and made a list of those that were related to indexing sexual identity. In addition, we watched two videos that featured heterosexual dating advice and made note of the advice given to men to romantically or sexually pursue other women. By comparing the two lists of notes, we were able to identify potential strategies that are WLW group-specific.

Study 2 consisted of two semi-structured interviews that took place and were recorded through Zoom. We interviewed a total three members of the WLW community, with two of them being in a committed relationship. They were primarily asked to describe and draw from their past experiences. The interviews were guided by six open-ended questions (see Appendix A) with the interviewer following up with questions when necessary. Our questions focused on the WLW’s description of their experiences in establishing mutual interest in women using non-direct measures. Participants were recruited using snow-ball sampling and all answers were kept anonymous. After the interviews, we listened to the audio recordings and made notes of the different ways WLW chose to index their sexual identity as well as the cues they used to determine the sexual identity of their romantic interest.

Study 1 Results

In Study 1, we were able to uncover several recurring themes. One point made consistently across multiple videos was that the WLW always felt the need to immediately make their sexuality known once they realized they had feelings for the other party.

Reasons for this were that they did not want to confuse the other party into thinking that they just wanted a female friend, and they also did not want the other party to assume that the speaker is straight and think differently of them. WLW worry about giving ambiguous signals if they were to not reveal their sexual identity soon enough, which leads to the subtle incorporations of various cues in conversations, such as mentioning the pride parade, to demonstrate their sexual identity.

They also made mentions of lurking through the other party’s social media for signs pertaining to possible membership of the WLW community to know whether it would be appropriate for them to make romantic advances. WLW also tend to be cautious in making advances as they adopt a “flirting by not flirting” technique. This allows them to slowly determine if the other party has reciprocated romantic feelings without being too overbearing and only continue to proceed if there is a positive response.

Figure 1: A selection of videos on WLW flirting used in Study 1

WLW flirting:  Video 1      Video 2     Video 3

In contrast, when we explored flirting advice geared towards men to pursue women, there was no  mention for men to index their sexual identity to women before flirting or at any stage of the courting process. The videos generally focused on advising men to be indirect to increase excitement in women and how to appear playful and masculine.

Figure 2: A selection of videos on heterosexual flirting used in Study 1

Heterosexual Flirting: Video 1      Video 2

Although there was some overlap in advice given to women to pursue other women and given to men to pursue women, such as being subtle and indirect, the reasoning behind it was different,  and a clear difference was the need for WLW to drop hints about their sexual identity. Because there tends to be less confusion in intentions when a male approaches a female, neither party is advised to hint at their own sexual identity nor advised on how to determine the other party’s sexual identity. In contrast, a common theme across videos geared towards WLW is to use references to hint at their own gayness or try to determine whether the other party is gay before advancing.

Study 2 Results

Interview 1

A summary of common themes that arose in Interview 1 are presented in Table 1 below along with some illustrative examples given by the interviewee.

Table 1: Recurring themes and examples from Interview 1

Interview 2

To illustrate the results derived from Interview 2, Table 2 consists of the most important statements made by both Subject 1 and Subject 2 in the conversation. It is important to note that Subject 1 and Subject 2 have been in a WLW relationship for over a year. When answering the interviewer’s questions, they both reflected on when they first met and how this has changed or remained consistent. The middle column consists of what they answered similarly.

Table 2: Noteworthy excerpts from each subject of Interview 2 and areas of overlap

 

Study 2 Analysis

From our interviews we gathered that the majority of strategies available for Women-Loving Women to identify and flirt with other WLW are mostly non-linguistic in nature. In both interviews, WLW referred to style of dress as a primary identifier for fellow WLW. These and other aspects of popular WLW culture were also drawn upon during the flirting itself, which leads us to one overtly linguistic flirting strategy we found was used by WLW– compliments. Compliments between WLW referenced nonverbal yet mutually understood markers of WLW identity, so they were used to confirm sexuality and communicate an attempt to flirt, in addition to their function as simple compliments. Importantly, compliments between WLW and platonic ones between heterosexual women were said to differ solely in their content and not their form. We conclude that this arises from a need or desire for WLW to flirt “under the radar” to avoid the very real danger of homophobia and bigoted comments.

We also noted the potential for confusion and ambiguous interpretations of these, arguably necessary, nonverbal flirting methods. Subject 1 even described a trend among WLW to pull back on “standard” physical or verbal affection (at least among other WLW) as a way to avoid creating confusion since more open displays of platonic affection are expected among groups of women. This may contribute to a societal perception of WLW as being “colder” or “more masculine.” Future studies might investigate whether or not this is true among a larger sample size.

Figure 3: A meme employing WLW popular music artist ‘Girl in Red’ to euphemistically index a WLW identity

 

Discussion and Conclusions

Our ultimate takeaway from these interviews was a strong indication that, motivated by a possible fear of negative attention, members of WLW groups feel the need to be covert in romantic contexts. As a result of this covertness, we noticed a trend of relying on nonverbal cues (like clothing choice) more than an awareness of individuals phonetically or lexically indexing their “gayness.”

Even in situations where an individual might directly state “I like girls,” the implication of “I’m romantically interested in you” often remains covert. This gives the other individual a choice as to whether or not an interaction is romantic in nature, but can end up causing some confusion. Thus arises the stereotype that WLW do not flirt. In many cases, their advances can easily be interpreted as platonic interaction among women in a society where affection among women is more normalized than among men, and where revealing your sexuality to the wrong person can have negative repercussions.

Further Reading Recommendations: Although we did not cover this information in our study, there have been numerous studies done on the language WLW may use that distinguish their patterns from heterosexual women. Robin Lakoff in Language and A Woman’s Place (1975), defines stereotypical “women’s language features (WL)” as those associated with “heterosexual women’s performance of femininity.” She contrasts this with the existence of typical “men’s language features (ML),” thus creating a binary of “women’s speech v men’s speech.” It would be interesting to use this and analyze whether women in the WLW community use either one or both of the language features, and whether this could be a distinguishing feature.

 

References

 Eliason, M.J., Morgan, K.S. Lesbians Define Themselves: Diversity in Lesbian Identification. International Journal of Sexuality and Gender Studies 3, 47–63 (1998). https://doi.org/10.1023/A:1026204208243

Lakoff, Robin (1975). Language and A Woman’s Place. Language in Society, Vol. 2, No. 1, 45-80.

Podesva, R. J. (2011). The California vowel shift and gay identity. American speech, 86(1), 32-51.

Rich, A. (1980). Compulsory heterosexuality and lesbian existence. Signs: Journal of women in culture and society, 5(4), 631-660.

Rieger, G., Linsenmeier, J. A., Gygax, L., Garcia, S., & Bailey, J. M. (2010). Dissecting “gaydar”: Accuracy and the role of masculinity–femininity. Archives of Sexual Behavior, 39(1), 124-140.

Shelp, S. G. (2003). Gaydar: Gaydar. Journal of Homosexuality, 44(1), 1-14.

 

[/expander_maker]

, , , , ,

Do you LOL out loud? Screen time influence on internet slang irl

Julia Baylon, Rachel Rim, Carolina Guerrero, Hebbah Elokour, and Caelynn Hwang

If you’re a college student reading this, you are a key individual in the composition of the Gen Z identity. Gen Z, today’s 18-23-year-olds, is defined by its fervent slang usage as well as its notorious association with and attachment to technology. Commonly used components of technology in Gen Z include social media platforms, such as Instagram, Snapchat, Twitter, TikTok, and Discord. This study investigates the relationship between the two variables of time exposed to social media and slang-impacted conversational speech style. As slang rapidly evolves to shape communication and accommodate the construction and expression of individual identity, we begin to ponder, where does this language style come from, and to what extent does it influence our diction? Through conversational analysis and survey/questionnaire conduction, we hope to understand the explicit impact of social media on conversational slang and whether or not the results support our hypothesis, which argues that greater usage of slang on the internet and/or overall internet presence will result in a higher frequency of slang used in real life.

[expander_maker id=”1″ more=”Read more” less=”Read less”]

Our burning question

In a direct relationship with the evolution of social speech, internet slang has transformed from being a marked informal language style to an everyday form of communication. Internet slang influences both the communicative behavior that individuals in society practice along with their daily usage of language. Previous research on first language acquisition reveals that exposure to the respective language is a crucial component to the extent of production ability. Implementing the same argument, we hypothesize that exposure to the internet, where there is an abundance of slang, will affect the frequency that internet-users produce slang in conversation. This study ultimately seeks to understand our proposed research question that asks:  Does the time exposed to social media affect an individual’s conversational speech style and does this influence bleed into their different social media platforms varying informality? 

How did we approach this?

In this study, undergraduate university students will be the target population; the age range is between 18 to 22 years old. Data collection will occur in two parts, first, case studies of three to four recorded conversations, each at least 15 minutes long, will be observed to determine the frequency of slang words used by each participant. The latter half of the data collection will comprise a self-report survey, in which the participant will disclose their social media, respective user handle, and average screen time for each platform. The survey will then prompt participants to reflect and report individual uses of abbreviations, acronyms, and other slang. It concludes with a free response question of whether they think that internet slang influence has a positive or negative impact on speech.

Ultimately, we aimed to examine any correlations between screen time, app usage, self-reported analyses of slang used in real life, and actual usage of slang in real life. The participants involved in this study were unaware of the purpose of the study beforehand, as well, to elicit as natural-sounding conversation as possible. Furthermore, the participants were all friends of the co-authors, to hopefully eliminate any biases potentially introduced by the onset of an “interview-like” setting. The participants were all also friends of the co-authors since during quarantine we were admittedly unsure how to go about encountering people we didn’t already know and recording natural conversations between them.

Our Findings

We asked individuals if they used Instagram, Twitter, LinkedIn, and Facebook. If they answered “yes”, they reported their average weekly screen time based off their devices’ reports or guestimate. 

Figure 1: Self-reported screen time.                                                                                               * Reported additional socials. Only Subject F reported screen time for Facebook (2 hrs)

We asked the participants if they used any other socials not mentioned in the survey and, if the did, to list them and report their screen time.

Figure 2: Additional socials. Only the subjects who reported any screen time are displayed.

 

Each of the 10 subjects was asked to assess whether they used Internet slang in their spoken language. The results are displayed here in this pie chart, with the overwhelming majoring responding “yes.”

Figure 3: Self-reported usage of slang in everyday speech

 

Each of the 10 subjects was asked to assess whether they substituted words in their spoken language. Results appear to be mixed, but again with the majority indicating that they do so at least “Sometimes.”

Figure 4: Self-reported usage of abbreviations in everyday speech

 

Although Subject A(she/her) had a pretty high average screen time (see Figure 1), the frequency of internet slang used in speech and posts is very little. Looking through her different social media platforms we found that in 2020, 0% of her Instagram posts, 0% of her LinkedIn posts, and 2.5% of her Twitter posts contained slang. About 45-55 minutes of recorded speech were collected for Subject A; this data was collected in the form of a phone call with a friend and the conversation topics were school, quitting a part-time retail job, a trip   to a museum, and apartment life. The results showed very little slang used (3 instances) in the form of emphasized agreement. When the friend expressed being happy living away from home and great dislike at going home for long periods of time, Subject A said “Facts! Period.” There was just one other instance of internet speech used but it was within the same conversation. While analyzing the recorded speech, we were surprised that, in the conversation about the job she hated and quit or the conversation about her trip to a museum, she did not use any slang. The tone of the first was annoyed/frustrated, while the second was very emotional and exciting. However, the Subject used the slang when the tone of the conversation was more neutral; perhaps it was the realization of the neutrality and therefore it was an effort to make the conversation more lively, or it just so happens that she thinks moving out and getting an apartment was the best decision of her life and gets excited when others feel the same. The interesting part was that after the recording was taken and the nature of the experiment revealed, Subject A laughed and expressed that she gave us a lot of material but was surprised when informed that she had only used slang 3 times. This overestimation of slang was seen among the other participants as well (see Figures 3 and 4). In the post-conversation survey, when asked if the influence of internet slang had a positive or negative effect, she answered,

“I think a little bit of both. It’s fun to use and say when with friends or in public. It also helps to reduce tension in conversations or with strangers. However, it is embarrassing and worrisome when people only speak like that or use the slang in essays, emails, etcetera.”

Additionally, Subjects E(he/him) and H(he/him) were also of note due to their significant presence on gaming platforms such as Discord (see Figure 2). During roughly 17 minutes of recorded speech, we were surprised to learn that there was little slang used (sick made an appearance a few times, as well as the occasional expletive), given that the gaming culture has evolved to contain speech community-specific language, which was confirmed in the post-conversation study and their Discord posts. Although their slang was not as active in their other social media platforms, Subject E had 68.7% of posts show slang and Subject H had 31.8% of posts show slang on their Discord (ACM Studio). This large difference between them can be attributed to Subject H’s position of co-presidents of ACM Studio, so his posts tended to be announcements, therefore, a bit more formal. The context of the conversation was generally relaxed and took place while cooking dinner. The topics of conversation strayed away from gaming topics, as there was a third individual present (not included in the study, as she is a co-author of this blog) who isn’t as big in the gaming community. Instead, the conversion-focused on classes, general Instagram-related topics, career focuses, and the like. However, in the post-conversation “debriefing” survey, the participants indicated a higher usage of gamer slang when talking to other gamers/about gaming-related topics. This included words such as pog, gg, F (..to pay respects), and broken (as in someone or something who is unfairly overpowered). As was discovered with Subject A, Subjects E and H also overestimated the extent to which they used slang in their everyday speech, indicating on the post-conversation survey that they believed their speech to “often” or at least “occasionally” use slang words.

The larger sociolinguistic picture

When trying to rationalize how little slang was used by the subjects, the words of sociolinguist Vera Regan came to mind. During her TEDxDublin, “What your speaking style, like, says about you” talk, she discusses how people’s language reflects their values, their goals, and their perception of themselves (2014). So, although Subject A is constantly exposed to internet slang, she will only choose the words that best represent her, as she is right now and who she wants to be in the future. As researchers, it highlighted for us how language is not only a tool for communication but also a form of creative expression that can hint at identity markers (like where we work, where we are from, our interests, etc.) like our style and choice in clothes.

In conclusion, throughout our study, we came to three main observations. The first observation states that some participants may have overestimated the extent to which they use slang in their speech. According to Kim and Ra, a higher frequency of the use of slang in one’s speech might be highly correlated to the limited means of communication that is available to reflect or express emotions, whether it is in written text or on social media (Kim and Ra, 2003). On the other hand, in real life, conveying emotions can come in the forms of facial expressions, voice tones, etc. Future studies investigating the effect of the use of emojis and uppercase letters would be interesting and may be useful to expand on the topic. Based on the data that we collected, our second observation stated that gender has little to no observable variance in our results. Our last observation indicates the conclusion that the topic of the conversation only seems to apply to the usage of speech in specific contexts. For instance, the speech observed of participants during playing a certain online game has sparked specific slang used in that specific game or common around social media platforms.

For future direction, it would be interesting to perform a study relating to specific speech communities. Our study unintentionally tapped on the gamer community, and an in-depth study on the usage of gamer lingo exclusively, correlated to the context in which it’s used (both online and in real life) could provide more detailed insight.

 

Acknowledgments

We would like to acknowledge everyone who participated in our study, especially the conversation participants who, after agreeing to participate, were told: “We’re gonna record our conversation for a bit. Don’t worry about it, okay? And no, sorry, we can’t tell you why until after. Just talk normally.” and then had to deal with the resulting confusion for a few minutes after.

 

References

An H. Kuppens (2010) Incidental foreign language acquisition from media exposure, Learning, Media and Technology, 35:1, 65-85, DOI: 10.1080/17439880903561876

Kim Y.S., Ra DY. (2003). Constructing an Internet Chatting Dictionary for Mapping Chatting Language to Standard Language. Lecture Notes in Computer Science, 2713. DOI: 10.1007/3-540-45036-X_72

TEDx Talks. (2014, November 21). What your speaking style, like, says about you | Vera Regan | TEDxDublin [Video]. YouTube. https://youtu.be/jAGgKE82034

[/expander_maker]

, , ,

“It’s not always negative, but sometimes it is”: Exclusivity in sororities vs. cottagecore communities

Sandra Fulop, David Huang, Yinling Li, Joyana Rosenthal

An important part of college is finding a space to belong. For marginalized students such as LGBTQ women, this can also be the most difficult part. Although there are often groups such as Gay-Straight Alliances or LGBTQ resource centers, these revolve entirely around the LGBTQ identity. But general women’s spaces, such as sororities, are notorious for being less accepting and more exclusive of marginalized identities. This presents an issue for LGBTQ women, who may struggle to create an identity outside of being LGBTQ while avoiding prejudice from groups meant to include all women. Our study focused on the vocabulary choices of LGBTQ women when discussing their own women-centric spaces, specifically Panhellenic sororities or cottagecore communities. We discerned how comfortable and included they felt in their respective spaces and how they felt others perceived them inside and outside that group. We created vocabulary categories to differentiate between inclusive/in-group and exclusive/out-group language, averaged the frequency of use across each group, and compared them. We found that LGBTQ cottagecore women expressed much more comfort in the space they belonged to, while LGBTQ sorority women swept their marginalized identities under the rug and focused on out-group perceptions and stereotypes.

[expander_maker id=”1″ more=”Read more” less=”Read less”]

Introduction and Background

LGBTQ college women deserve for women’s spaces to be accepting of their identities. This research concerns how comfortable they are in the groups to which they belong, as measured by which categories of words they use during an interview. The target population was women aged 17-22 who identified themselves as members of women-centric spaces, specifically sororities or cottagecore communities. Cottagecore is a primarily online community catering to LGBTQ identities, which idealizes nature and has a distinct fashion aesthetic. Sororities are a part of Greek life, a space designed for women at universities to meet and support one another through college. Although both these spaces are focused on the wellbeing of women, they are catered to different perspectives. While sororities typically operate under fraternities’ male gaze, cottagecore communities tend to be created by and tailored towards LGBTQ women. We were interested in how the vocabulary of LGBTQ women in these spaces reflected how comfortable they were in their chosen community.

Since individuals’ environments create the framework of how they perceive and are perceived by others (Greco, 2012, pg. 567), we chose two women-centered communities to investigate how the space an LGBTQ woman belongs to affects how they view others and themselves. It has been previously shown that students at campuses without Greek life are more accepting of LGBTQ identities (Hinrichs & Rosenberg, 2002, p. 69), while LGBTQ students in Greek life often feel excluded due to heteronormativity (Fine, 2011, p. 534). LGBTQ students have also faced hostility in residence halls (Evans & Broido, 2009, p. 40), leaving LGBTQ women a small number of communities in which they can feel comfortable. We posited that these women might turn to online communities such as cottagecore, where they can feel more accepted than they would in most in-person spaces. This specific research was informed by a study that showed how women’s words linguistically reflect dependence on men (Lakoff, 1973, pg. 46). We believed that this would be true of sorority women who are counterparts to male fraternities. Much of their identity as a sorority woman likely reflects the opinions and perspectives of those men in fraternities. However, because cottagecore communities have no reason to focus on such a direct male presence, we hypothesized that those women’s vocabulary would not corroborate Lakoff’s results. There is a gap in the literature comparing on-campus women’s spaces to off-campus women’s spaces. However, because cottagecore originated as a safe space for women and those identifying as LGBTQ, we hypothesized more vocabulary surrounding comfort and inclusivity within this community. This directly contrasts sororities, where we expected that LGBTQ women would discuss more exclusion and focus more on out-group perceptions.

Methods

We interviewed two LGBTQ women, each in cottagecore communities and sororities. The researchers asked each participant the same set of interview questions about their identities and perceptions, how represented they felt in their spaces, and how others on campus viewed them and their communities. We divided inclusive and exclusive languages into specific categories, such as stereotyping and out-group vocabulary, and averaged the frequency of each category’s use between the two members of each group, then compared them to the other group’s use.

Results and Analysis

To analyze the linguistic data we collected, we created seven different language categories to analyze what each woman said about the community to which she belongs. The first was the inclusive language, which was any word or phrase representing inclusion. For example, referring to their space as a “community.” The second was warmth/comfort, which was a word or phrase referencing a feeling of safety or happiness in their space. Examples used were “comfortable” or “kind.” The third was any mention of the LGBTQ community, such as “lesbian” or “LGBTQ community.” These were the categories that demonstrated inclusivity and comfort in their space.

The more negative or exclusive categories were the following: exclusive language, any word or phrase representing exclusion of a particular person or group of people, such as “bigotry.” The next was stereotyping or labeling, any language referencing or creating a stereotype or label. Some examples we heard were “dumb party girl” or references to a “curated image.” The next was out-group language, anything referencing a group’s perception that the woman herself did not belong. Some examples were “fraternities” and “art hoes.” The last category under exclusive language was mentions of heterosexuality or cisgender people. This is not a negative concept, but since we measured comfort and confidence in LGBTQ women, we looked for positive references to their own identities rather than those of other people. We noted mentions of homophobic exclusivity. These references included phrases such as “heteronormativity” or “cisgender ideals.”

In Graph 1, each category is represented by a double bar graph for the two groups’ average frequency of use. We calculated the number of words from a specific category to the total words they used in every category and averaged it between the two women in each group. We found that cottagecore women used inclusive language at an average frequency of .401, and sorority women at .165. Cottagecore women used warmth/comfort language at an average frequency of .305, sorority women at .08. Cottagecore women mentioned the LGBTQ+ community at an average rate of .125 and sorority women at  .015. Cottagecore women used exclusive language at an average rate of  .075, and sorority women at .265. Cottagecore women mentioned stereotypes and labels at a rate of .045, and sorority women at a rate of .195. Cottagecore women used out-group references at an average of .02 and sororities women at .2. Lastly, we found that cottagecore women mentioned heterosexuality and cis-gendering at a rate of .02, with sororities at .03. 

Graph 1 – Use of linguistic variable terms

We also noted phrases that did not fit into a specific inclusion/exclusion category but were telling about the interviewees’ perceptions. One example is that, when we interviewed the participants, we referred to both cottagecore and sororities as “communities” explicitly. However, the sorority women consistently referred to their space as an “institution” or “organization” (Fig. 1), whereas the cottagecore women-only referred to their space as a “community” (Fig. 2). We found it interesting that the sorority women chose to use words that, instead of indicating closeness to the space they belonged to, were indicative of something separate from themselves.

Figure 1 – Transcription of L

 

Figure 2 – Transcription of B

 

The sorority women were also more hesitant to critique their space. For instance, when asked about the exclusivity of their space, one sorority woman stated the following:

Figure 3 – Transcription of J

 

Additionally, both sorority women answered the question regarding a feeling of acceptance and representation as an LGBTQ woman in their space with uncertainty (Fig. 4). In contrast, the cottagecore women answered the same question with definitive yeses (Fig. 5).

Figure 4 – Transcription of J

 

Figure 5 – Transcription of S

 

Finally, we concluded that cottagecore women were much more comfortable in their community than the sorority women were. They used inclusive language at almost 2.5 times the rate of sorority women and warmth/comfort language at about 3.8 times the rate of sorority women. They also referenced out-groups much less and tended not to stereotype others. Sorority women used exclusive language at 3.5 times the rate of cottagecore women and referenced out-groups (usually groups relating to men such as fraternities) at ten times the rate of cottagecore women.

Discussion and conclusions

Based on our data, the sorority women were much more focused on others’ perceptions of them, likely because they are conditioned to focus on an image as counterparts of male fraternities. In contrast, the cottagecore women felt freer to focus on their own safe space within their inclusive community. We concluded that cottagecore women’s common mentions of being LGBTQ show that they are more comfortable than sorority women, who acknowledged heteronormativity and never mentioned being LGBTQ after the initial statement of their identity. This demonstrated that cottagecore women are less likely to internalize bias against the LGBTQ community, while sorority life normalizes these biases. Repeating this study with larger sample sizes at different universities might also yield more nuanced information regarding regional and cultural differences concerning this phenomenon.

Following this, we believe that college-age LGBTQ women need more spaces geared towards them to feel comfortable and safe because sororities do not seem to give LGBTQ women the room to explore and express their LGBTQ identities. Colleges should look towards cottagecore as a model for on-campus casual interactions between LGBTQ women in a space that is acceptable but not focused on being LGBTQ. As campus communities try to be more inclusive of all identities, it could also be beneficial for sororities to undergo more implicit bias and diversity training.

Although our research focused on LGBTQ women because we assumed marginalized identities would express more apparent feelings of exclusion, it could be that women in sororities might speak a certain way due to the space to which they belong. It might not be related to their sexual identity. For further research, it would be informative to look at whether there is a difference in how straight women and LGBTQ women, both in sororities, use these same vocabulary categories.

References

Evans, N. J., & Broido, E. M. (2002). The Experiences of Lesbian and Bisexual Women in College Residence Halls. Journal of Lesbian Studies, 6(3-4), 29-42. https://doi.org/10.1300/J155v06n03_04

Fine, L. E. (2011). Minimizing heterosexism and homophobia: constructing meaning of out campus LGB life. Journal of Homosexuality, 58(4), 521-546. https://doi.org/10.1080/00918369.2011.555673

Greco, L. (2012). Production, circulation and deconstruction of gender norms in LGBTQ speech practices. Discourse Studies, 14(5), 567-585. https://doi.org/10.1177/1461445612452229

Hinrichs, D. W., & Rosenberg, P. J. (2002). Attitudes Toward Gay, Lesbian, and Bisexual Persons Among Heterosexual Liberal Arts College Students. Journal of Homosexuality, 43(1), 61-84. https://doi.org/10.1300/J082v43n01_04

Lakoff, R. (1973). Language and woman’s place. Language in Society, 2(1), 45-79. https://doi.org/10.1017/S0047404500000051

[/expander_maker]

“Sorry, I Didn’t Quite Get That: The Misidentification of AAVE by Voice Recognition Software”

Shannon McCarty, Lam Pham, Alora Thresher, Alexandria Wasgatt, Emma Whamond

This study investigates the transcription accuracy by AI speech recognition systems using natural language processing when interpreting standard American English dialects (SAE) versus African American Vernacular English (AAVE). We inspect the percentage of misidentified words, and the degree to which the speech is misidentified, by AI speech recognition systems through analyzing authentic speech found in YouTube videos. The accuracy of voice recognition with respect to AAVE will be determined by selecting for distinct AAVE features, such as G-dropping, the [θ] sound, reduction of consonant clusters, and non-standard usages of be. The methodology includes feeding YouTube clips of both SAE and AAVE through an AI speech recognition software, as well as examining YouTube’s auto-generated transcripts, which are created by automatic speech recognition based on the audio of the YouTube video. The purpose of this study is to bring attention to the needs of diversity in technology with regard to language variation, so that AI speech systems such as Amazon’s Alexa or Apple’s Siri are more accessible to all members of society, as well as to help destigmatize a variety of American English that has carried social, cultural, and historical stigma for centuries.

[expander_maker id=”1″ more=”Read more” less=”Read less”]

Introduction and Background

2020—American society finds itself at the thrilling forefront of technological innovation, yet is still plagued by racial inequality and systematic racism. Our study aimed to tackle a portion of this American duality from a sociolinguistic standpoint. We investigated AI speech recognition systems and the identification differences while processing speech of the standard American English dialect (SAE) versus African American Vernacular English (AAVE). SAE is most broadly defined as the most uniform, accepted, and understood language in the US. On the other hand, AAVE is not solely slang or a lesser form of SAE; rather, it is a language variety that has systematic linguistic patterns and carries social, cultural, and historical stigmas. We believe AI speech recognition systems will reflect our country’s racial biases by misinterpreting AAVE far more often than SAE. As these technologies become more commonplace and embedded in society, our study’s goal is to shed light on whether AI speech technology is inclusive of the AAVE dialect and its speakers.

Methods

An overview of AAVE linguistic features can be found in a paper discussing the matter by Erik Thomas (2007), though we narrowed our focus on the most defining features of AAVE in this study. These included G-dropping, the th sound (as in bath) becoming the f sound (as in fast), reduction of consonant clusters at the ends of words (wes side versus west side), and the use of the verb be (Singler, 1998).

The analysis of the main facets of the verb be included the dropping of be, the habitual be (Collins, 2006), the use of BIN and be done. The heavy focus on be was due to the fact that it is one of the biggest syntactic differences between SAE and AAVE (Lanehart, 2015). As a result, the variants of be  were most likely to affect the transcription performance of voice recognition software when processing AAVE. 

This study used the abundant linguistic resources available from YouTube to exhibit the use of SAE and AAVE dialects. We compiled audio clips of authentic, normal speech in both of these varieties and fed the audio clips into an AI transcription software, specifically dictation.io/speech, as well as examined YouTube’s AI-generated closed captions. It was important to ensure that the captions we used on YouTube were the auto-generated ones, and not captions that had been manually entered by the uploader.

To determine what percentage of speech and what types of features of the dialects were misidentified by the AI speech recognition systems, we quantified the data by categorizing the accuracy of the AI transcriptions into four groups: “All,” meaning All of the words were picked up, “Most,” “Few,” or “None.”  If every single word in the audio recording was transcribed correctly by the AI speech recognition system, then that audio clip was placed in the “All words picked up” group. If most words or only a few words were transcribed correctly by the AI speech recognition system, then that audio clip was placed in “Most” and “Few” groups, respectively. And if nothing in the audio clip was transcribed correctly by the AI speech recognition system, then that audio clip was placed in the “None” group.

Our hypothesis was that the percentage of interpretation inaccuracy in AI speech recognition systems while interpreting AAVE would be significantly higher than the inaccuracy recorded by speech recognition software while interpreting SAE (i.e. more AAVE audio clips would land in the “Few” and “None” transcription accuracy groups). This would lead to an expectation that voice recognition software is globally less effective for speakers of AAVE.

Results/Analysis

The results of this study were overall in line with the hypothesis—the transcription software picked up more SAE words than AAVE words. The vast majority of AAVE video samples collected had “Few” words transcribed correctly. Figure 1 shows the amount of AAVE words picked up by the transcription software. Our study found that for all of the clips of AAVE:

    • 16.7% “All” words were picked up
    • 16.7% “Most” words were picked up
    • 56.7% “Few” words were picked up
    • 10.0% “None” of the words were picked up

It is noteworthy that the transcription software was unable to pick up the majority of AAVE speech when the video clips were similar to the SAE video clips with regard to background noise, speaker volume, etc. We believe this is largely due to AAVE phonetics rather than AAVE syntax and word choice, which is what we were focusing on in this study.

Example 1: This clip was categorized under “Few” words being picked up; in the beginning of the segment, the speaker actually says it’s a whole buncha people in New York that we know from. (“African-American English in North Carolina”, The Language & Life Project)

On the other hand, all video samples of SAE fell under “All” or “Most” transcription accuracy. The SAE video clips had the majority of their words picked up and transcribed correctly. According to Figure 2:

    • 31.6% of the clips had all their words picked up
    • 68.4% had the majority of their words picked up

The results of transcribing SAE versus the results of AAVE are drastically different; we believe this is largely due to the phonetic differences between the two English varieties. The only time SAE was not picked up was when the speaker was speaking faster than normal. All of the words could have blended together, and as a result, the transcription software was not able to distinguish what was being said.

Another unexpected observation for both AAVE and SAE was that women speakers were not picked up as often as male speakers; AAVE female speakers made up 66.7% of the “None” category, despite comprising less than 20% of our AAVE samples.

Limitations

We had some limitations in our project as well. Due to COVID and time restrictions, our sample size was regrettably small at only 49 clips. We also had a few clips become unusable between the time of our finding them and analyzing them. As a result of the small sample size, we ended up noticing a huge, unexpected difference between the transcription accuracy of male versus female speakers of AAVE, but we weren’t able to draw any solid conclusions as to whether this difference is representative of a larger population. However, the difference was quite apparent within the random pool we gathered, so a study with equal amounts of male and female AAVE speakers could address this issue more equitably.

Speakers that were present in the room would also have enabled a more accurate representation of AAVE and voice recognition software’s real-world interaction. Our computers and phones were limited due to the differing microphone quality, which could have influenced the amount of words the voice recognition software picked up. In addition, we had planned to use our phones’ dictation softwares as the primary method of data acquisition, which ended up being wholly impossible, so we only had the time to find and use one transcription site.

Many of the video clips were filmed in neighborhoods, with background noise included, such as cars going by, wind, and people talking. This could also have limited our transcription site and created another layer of ambiguity. However, SAE clips with similar amounts of background noise that were played were transcribed (mostly) correctly. So it is unclear if the background noise is fully a limitation or an example of voice recognition software not picking up AAVE.

Lastly, we began to wonder if the phonetics of AAVE had more of an effect on our voice recognition software than the syntax did—that is, we wondered whether the sound of AAVE was more impactful than the sentence structure. A few of us used the software to speak sentences that were syntactically AAVE, but not phonetically, as none of the researchers of this study were AAVE speakers. Those sentences were transcribed correctly, including AAVE syntax. However, we did not have enough time to pursue that avenue of research, but it would be a promising starting point for any future projects.

Discussion and Conclusions

This investigation was intended solely to pursue the question of whether there is a racially-based difference in accuracy of voice recognition software. Despite the unexpectedly small scope of our study, we believe our results are sufficient to prompt further investigation into the reasons as to why there is such a stark difference in accuracy.

To that end, a question that arises naturally is whether closing the gap in accuracy is as simple as writing and including a few more lines of code. If it is, then what’s preventing this from happening now? And if the voice recognition software instead needs to be re-constructed from the ground up, then that, in turn, spawns a whole host of follow-up questions (e.g. Who’s paying for this development? Who’s working on it? How long will it take?).

It is, as with any discussion of racially-based imbalances, also worth considering systemic variables at play. Another question worth pursuing is whether there is a higher average level of inaccessibility to the technology sector for speakers of AAVE compared to speakers of SAE, which would contribute to a broad range of consequences—one of which might be the discrepancy in accuracy of voice transcription softwares demonstrated in this study.

In 2016, Rickford and King cited Schneider (1996) to describe AAVE as “the US English dialect most examined by linguists for quite some time.” In the academic world, it’s common knowledge that the prejudices held against AAVE  — and its speakers — have no basis in fact. The drastic difference in voice recognition software transcription accuracy between SAE and AAVE is not only one more imbalance to correct in pursuit of a fully equitable society, but also a symptom of the systemic racism that influences all aspects of daily life. Studies like this one which stop at identifying the problem are only the first step; the next is to examine why the problem exists in the first place, so that work to resolve the inequity can begin.

References

Collins, C. (2006). A fresh look at habitual be in AAVE. CREOLE LANGUAGE LIBRARY, 29, 203.

Koenecke, A., Nam, A., Lake, E., Nudell, J., Quartey, M., Mengesha, Z., Toups, C., Rickford, J., Jurafsky, D., & Goel, S. (2020). Racial disparities in automated speech recognition. Proceedings of the National Academy of Sciences, 117(14), 7684-7689.

Lanehart, S. (Ed.). (2015). The Oxford Handbook of African American Language. Oxford University Press.

Lippi, R., Donati, S., Lippi-Green, R., & Donati, R. (1997). English with an accent: Language, ideology, and discrimination in the United States. Psychology Press.

Rickford, J. R., & King, S. (2016). Language and linguistics on trial: Hearing Rachel Jeantel (and other vernacular speakers) in the courtroom and beyond. Language, 92(4), 948-988.

Singler, John Victor. “What’s not new in AAVE.” American Speech 73.3 (1998): 227-256.

Thomas, E. R. (2007). Phonological and phonetic characteristics of African American vernacular English. Language and Linguistics Compass, 1(5), 450-475.

[/expander_maker]

, ,

They and Them: Gender Inclusivity Across Languages

Alexander Gonzalez, Maeneka Grewal, Nico Hy, Zoe Perrin, Vivian San Gabriel

The relevance of gender-neutral language has surged due to growing acceptance towards nonbinary and gender non-conforming people as well as the dissolution of the gender binary. Through comparative analysis of native English and Spanish speakers, we investigated the impact of grammatical gender on the methods speakers employ to express gender neutrality. Since Spanish sentences require full gender and number agreement, expressing gender neutrality in Spanish presents more challenges than in English. We asked participants to describe images of individuals and observed that the English speakers used gender-neutral language at higher rates than the Spanish speakers did. Their methods differed as well. Spanish speakers were more likely to mix feminine or masculine forms, alongside neutral descriptions, which we interpreted as attempts to use gender-neutral language. We can infer that even when Spanish speakers are looking to express something gender-neutrally, they may be limited by the lack of gender-neutral lexical items that can be used throughout an entire utterance. Our experiment was limited to written responses and as a result may not be representative of these speakers’ language use overall. More experiments dealing with oral speech and analyses of other gendered languages would contribute to the knowledge and understanding of this field.

[expander_maker id=”1″ more=”Read more” less=”Read less”]

Introduction

The language of gender inclusivity is constantly shifting and becomes increasingly relevant as our understanding of gender changes and the voices of nonbinary individuals are amplified. The recent surge in nonbinary visibility has drawn attention to the grammar of gender-neutral pronouns, especially in languages that have grammatical gender marking. We wanted to explore how speakers navigate using gender identity-related pronouns and terms to express gender neutrality, particularly in English, a language that does not use grammatical gender, and Spanish, a language that does use grammatical gender. 

In English, the pronoun “they” is often used as a gender-neutral pronoun. The Spanish equivalent would be the novel pronoun “elle/ellx.” However, Spanish’s grammatical gender makes this pronoun difficult to use in spontaneous speech. In Spanish, all nouns and everything associated with them must be modified to fit gender and number agreement, while in English, nothing needs to be modified in order to use “they” in a sentence.

Through this experiment, we were looking to explore how grammatical gender may impact the ways speakers’ expresses gender neutrality when referring to a subject. This experiment focused specifically on individuals’ use of pronouns and other gender markers in their writing. We collected responses from native English and Spanish speakers of varying gender identities, looking to highlight the different ways gender neutrality is encoded in languages with grammatical gender compared to languages without it. As English already possesses a gender-neutral pronoun and does not use gender agreement, we predicted that English speakers will be more likely to use gender-neutral terms than Spanish speakers.

Methods

To test out this hypothesis, we conducted an experiment using Google Forms surveys to track the usage of pronouns and gender marked words in English and Spanish written communication. We collected our data through 2 separate surveys, each written in English and Spanish respectively. Each survey contained the same 10 computer generated images of androgynous individuals paired with a prompt requesting that participants describe the individuals using full sentences. It was important for us to disclose that the individuals in the photos were computer generated so as to avoid participants manipulating answers due to fears of misgendering real people. It was also important that we asked participants to describe the people in full sentences to increase the likelihood of participants using pronouns and gender marked terms.

Figure 1: A computer-generated image of a person (from www.thispersondoesnotexist.com)

The methodology used in this experiment was inspired by a prior study done by Bradley, Salkind, Moore & Teitsort (2019) which examines English L1 cisgender subjects’ perception of singular “they” as a non-gendered pronoun. In this study, the researchers analyzed English recordings of participants’ verbal reactions to image stimuli. However, our experiment will be analyzing how gender neutrality is expressed in writing for both English and Spanish. We chose to analyze written communication because of possible difficulties in verbally expressing gender neutrality in Spanish due to the language’s grammatical gender. It was found in the study by Slemp (2020) that verbally expressing gender neutrality in Spanish takes conscious effort. This is not only because of Spanish’s grammatical gender agreement, but also because there is no verbal standard gender neutral morpheme. Slemp found that, in order to express gender neutrality, participants alternated between the morphemes -e and -x in written language as replacements for -a and -o. These findings guided our decision to analyze written language as opposed to verbal responses.

Results

We received 13 responses to our English survey, and since each responder was asked to describe 10 images, we received a total of 130 descriptions in English. We found that 51.5% of these descriptions used gender-neutral language. Examples of gender-neutral language found in our English speakers include the explicit use of the gender-neutral pronoun “they” in sentences like “they have dark colored eyes with crow’s feet,” as well as the total avoidance of pronouns in favor of gender-neutral terms like “person” in sentences like “this person has dimples.”

The remaining 48.5% of descriptions used gendered language, with 25.4% of the responses being feminine descriptions and 23.1% being masculine descriptions. Of our 13 responders, 10 people (76.9%) used a neutral description at least once, while 3 people (23.1%) did not use a neutral description at all, meaning that they gendered every single image.

Figure 2: Percentage of feminine, masculine, and gender-neutral descriptions used by English speakers, from a total of 130 descriptions.

We received 11 responses to our Spanish survey for a total of 110 descriptions. While the use of gender-neutral language was a majority in the English survey, only 21.8% of the Spanish descriptions used gender-neutral language, and 5 of these gender-neutral responses used feminine or masculine pronouns or adjectives combined with neutral descriptions. We interpreted these responses as attempts to use gender-neutral language. Other ways in which Spanish speakers expressed gender neutrality include the avoidance of pronouns similar to the avoidance practiced by English speakers, the use of question marks to signal uncertainty about gender, as in “el señor?” and the use of a dual marker “-o/a” for gender-neutral adjectives.

Within the remaining 78.2% of gendered descriptions, 40.9% were feminine and 37.3% were masculine. Of our 11 responders, 7 people (63.6%) used a neutral description at least once, while 4 people (36.4%) did not use neutral descriptions at all.

Figure 3: Percentage of feminine, masculine, and gender-neutral descriptions used by Spanish speakers, from a total of 110 descriptions. Five of the neutral responses combined feminine or masculine pronouns or adjectives with neutral descriptions.

Discussion and Conclusions

Our results show that our English-speaking participants used more gender-neutral language than our Spanish-speaking participants. We can most likely attribute this to the fact that using the pronoun “they” was the most common way English speakers chose to convey gender neutrality: as we hypothesized, it appears that the availability of the gender-neutral “they” is what allowed them to do so. Gender-neutral language also seems to be more easily accessible in English, as shown in the way some of the English speakers fluidly switched between the pronouns “he,” “she,” and “they,” both between and within sentences.

            On the other hand, none of the Spanish speakers used the novel pronouns “elle/ellx,” which suggests that these pronouns are less widely accepted and less readily available than the English “they.” This highlights an obstacle to introducing a new pronoun into a language: it is not likely to be understood and used in casual language if it is not well-known by speakers. It is most likely because the Spanish speakers didn’t have this gender-neutral pronoun available that they used various other methods to convey gender neutrality, such as mixing the gender agreements of articles, adjectives, and nouns. Mixing masculine and feminine forms suggests that they were aiming to construct gender-neutral sentences using the resources available to them.

            In both languages, there were speakers who avoided pronouns altogether and used the word “person” or “individual” instead of gendered terms like “man” or “woman.” Some speakers expressed uncertainty over their use of gendered language as well as the gender of the person in the image, either explicitly through words like “I think,” or implicitly through the use of question marks. These uncertainties suggest that participants would have been more confident if there were more gender-neutral options in circulation—not only existent, but well-known and commonly used, as to allow a mutual understanding of the word between both speaker and listener.

            While our data was collected in the form of written responses and may not accurately reflect the use of gender-neutral language in English and Spanish speakers, especially because written language lacks the spontaneity of spoken language, our results suggest that English speakers use gender-neutral language at a higher rate than Spanish speakers do. We think it would be worthwhile to conduct a similar study with a focus on speech rather than writing, as it would not only allow more insight into the use of gender-neutral language in general, but also investigate the feasibility of introducing new phonemes into languages for the sake of gender inclusivity, such as the -x marker in “ellx.” Ultimately, though, we have reached a better understanding of the various ways speakers can incorporate gender-inclusive language in their casual speech.

Bibliography

Balhorn, M. (2004). The Rise of Epicene They. Journal of English Linguistics, 32(2), 79–104. https://doi.org/10.1177/0075424204265824

Bradley, E. D., Salkind, J., Moore, A., & Teitsort, S. (2019). Singular ‘they’and novel pronouns: gender-neutral, nonbinary, or both?. Proceedings of the Linguistic Society of America, 4(1), 36-1. https://journals.linguisticsociety.org/proceedings/index.php/PLSA/article/viewFile/4542/4148

Lew-Williams, C., & Fernald, A. (2007). Young children learning Spanish make rapid use of grammatical gender in spoken word recognition. Psychological science, 18(3), 193–198. https://doi.org/10.1111/j.1467-9280.2007.01871.x

Schriefers, H., & Jescheniak, J. (1999). Representation and Processing of Grammatical Gender in Language Production: A Review. Journal of Psycholinguistic Research, 28, 575-600.

Slemp, K. (2020). Latino, Latina, Latin@, Latine, and Latinx: Gender Inclusive Oral Expression in Spanish.

https://ir.lib.uwo.ca/etd/7297

[/expander_maker]

, , , , ,

“I’m Sorry”: A comparative study of gender and individual differences in applying apology strategies in YouTube videos

Kristin Nguyen, Luxuan Huang, Vanessa Zhu, Andrea Mata, Shiyun Zhou

In recent years apology videos have become a very popular tactic used by social media influencers in efforts to help restore their online image. This study will compare and contrast the apology strategies used in 3 male and 3 female YouTuber apology videos by investigating the types of linguistic features that are found in both genders.  Moreover, we will further explore how the specific apology strategies being used influence the perception that their audiences/supporters have towards these specific Youtubers based on the comment section. The results showed that male Youtubers are more likely to use the “acknowledgement of responsibility” and “promise of forbearance” approach when apologizing while females are more likely to use the “explicit expression of apology” and explanation or account” strategy. Interestingly enough, the videos with the most positive responses came from 2 male and 1 female YouTuber which suggests that, according to their data set, there is a pattern in certain apology strategies that are more effective than others.

[expander_maker id=”1″ more=”Read more” less=”Read less”]

Introduction & Background

An increasing number of influencers apologized by publishing YouTube videos as an act to save faces and restore their images in the public’s mind. Previous studies about gender differences in using interpersonal apology strategies categorized apology strategies into 4 categories — explicit expression of apology, explanation or account, acknowledgement of responsibility, and a promise of forbearance, and concluded that women used more explicit apologies than men in interpersonal apologies (Holmes, 1989). However, apology videos are a fairly new phenomenon, and we wonder whether the pattern would also appear in our studies. As a result, we form our topic as a comparison study of gender difference and individual differences in Youtube apology videos in which we would explore the individual YouTuber’s choices of apology strategies and also gender differences based on the categorization mentioned above. 

Project Design

To identify how different apology strategies are used by individual YouTubers as well as by each gender, we chose a sample of 6 Youtube apology videos, 3 by females and 3 by males (see Table 1). The criteria for sample choice were based on Karlsson (2020)’s standards, which are 1) made by an independent Youtuber who runs and owns the channel, 2) made originally for Youtube and conducted in English, 3) belong to Beauty, Lifestyle, or Vlogging genre, 4) apologized for racist comments. By choosing videos addressing the same controversy over past racist comments, we minimized the influences of social and cultural contexts.

Following Holmes (1989)’s categorization for apology strategy, we collected the instances from each apology video that fit into the classification. Both qualitative and quantitative methods were used to analyze each individual video as well as all videos produced by each gender. We mainly employed discourse analysis to conduct the qualitative research when examining each individual Youtuber’s utterances as categorized by the four apology strategies and we also calculated the proportions of each strategy usage as divided by the total instances of apology in individual videos as well as in all videos produced by each gender. Besides, we also took the top 10 comments into account to evaluate how these apology videos were perceived (negative or positive), which then gave us implications of the effects of different combinations of apology strategies.

Results & analysis

A. Individual Differences

A.1. Jenna Marbles – apology video

Jenna Marbles’s apology video (which has been taken down along with her channel by Marbles herself) consisted of four different occurrences that she addressed. Three were ones that people criticized and questioned her about, and one was an issue that she felt she needed to apologize for, despite her claim that no one said anything negative about this issue. Her most used strategy was “explanation or account,” which was 11 times out of the total 22 strategies in her video. She showed regret in her old content and emphasized that when she made those videos, her intention was never to hurt anyone. Although the comments we analyzed were not pulled from the original video, we still believe they reflect the general opinion of her apology. Viewers almost unanimously agreed that her apology for all four instances were genuine, and many users actually displayed sympathy for her. Her apology generated many conversations and discourse about the existence of “cancel culture.” Her least used strategy was “promise of forbearance”; however, this is arguably her most effective strategy. Marbles claimed that she wanted to “be accountable for myself” and that she could not continue being on Youtube. Since the release of the apology, Marbles has not returned to Youtube.

A.2. Tana Mongeau – apology video

Tana Mongeau is a “storytime-centric creator. To preface this particular analysis, she was already well-known for embellishing some of her stories. It is interesting to note that her credibility was already questionable before the apology was released. This apology was in regard to her younger self using the N-word and to her reaction to another creator (iDubbbz) heavily criticizing her for this. Looking at the numerical data of her strategy usage, “explicit expression of apology” and “explanation or account” are nearly equal, with 11 times for the former and 13 times for the latter. However, 17 out of the 22 minutes in run time consisted of her explaining her logic of why she used to believe the Nword was acceptable to use and why she reacted aggressively towards iDubbbz. It was important to take into account the runtime and to consider that the number of times she used “explanation or account” alone does not fully reflect the implications of this strategy. She often repeated her explanations with slight variation in syntax but the overall lexical meaning was retained; she explained she was always “running away” or “hiding from my problems.” When there were “explicit expression of apology,” Mongeau also frequently berated her own image and character along with the explicit expression, such as “I’m sorry I was so fucking stupid.” Her apology had a significant negative reaction from her the audience, and many did not find her apology to be sincere or authentic.

A.3 Laura Lee – apology video

Laura Lee apologized for retweeting with racist comments in the year 2012. In her apology video, she applied explicit expression of apology 10 times, explanation or account 12 times, acknowledgement of responsibility 9 times, and a promise of forbearance 3 times. Laura used explanation or account most (p=35%) and explicit expression of apology (p=29%). Laura’s accounts or explanations were supposed to express remorse or clearly present the context of the event, but she ended up shifting blame. Laura’s strategy was to shift the responsibility to the younger her by reiterating the time when the retweet event took place was “six years ago” when she was “stupid and ignorant”. In terms of explicit expression of apology, Laura expressed her apology 10 times to different target audiences. She used “sorry” 8 times out of 10 and only said “apology” 2 times in the video which set an informal tone to her video. Overall, Laura’s apology video was like an interpersonal talk to her subscribers. Thus, even after editing, the apology video was not logical and seemed that she did not plan ahead and the use of sorry instead of a more formal term “apology” fit with this general tone.

A.4. Pewdiepie – apology video

From our pool of samples, Pewdiepie (Felix Kjellberg) had the shortest video where the run time was under two minutes. He has over 100 million subscribers, and he made the apology video in order to address and apologize for his use of the N-word during a livestream. He used all strategies a total of only six times. He only explicitly apologized for hurting and offending viewers once, and his most used strategy was “promises of forbearance.” He explained he had used the slur in the heat of the moment but that it was ultimately an inexcusable action. He placed focus on his own need to be accountable for his character and what he planned to do moving forward. His video was extremely concise compared to our other samples, and viewers seemed to react positively towards his apology. In general, most viewers commented and judged that his apology was genuine. It is also interesting to note that his comment section had many users comparing aspects about his apology to other apologies, especially that of Laura Lee’s and Tana Mongeau’s.

A.5. Shane Dawson – apology video

Shane Dawson made an apology video for doing blackface and saying the n-word in past racist YouTube videos. In his video, 80 instances counted as apology strategies, Shane mostly used “explicit expression of apology” (p = 37.5%) and “explanation or account” (p = 37.5%). “Acknowledge of responsibility” made up 15% of his apology, followed by 10% of “a promise of forbearance”. Specifically, the way Shane used “explicit expression of apology” mainly focused on expressing his regret (N = 26) through informal offers of apology “I’m sorry” which indicates Shane’s intention to resonate with his audiences. By repeating “I’m sorry” with the lowering of pitch, Shane reinforces his remorse and desperation. In terms of how Shane explained his wrong-doing, he frequently used –“funny” and “joke[s]” — to define his past mistakes. Moreover, he shifted to a higher pitch to imitate what the young Shane thought. However, this voice-shifting may not contribute to the apology but create the impression that Shane was trying to disassociate from the then-self and shift the blame. Shane also addressed himself from a third-person perspective, such as “I can’t even…see this white fucking guy do blackface.” The way he referred to himself may strengthen the impression that he was shifting the blame. The Youtuber did not frequently employ “acknowledgement of responsibility” and “a promise of forbearance”. Shane accepted the blame mostly and hardly offered a repair. Shane’s promises are also quite general and vague, such as his use of the demonstrative pronoun “that” in “I would never talk about that now”, which does not specify the action or mindset he would change.

A.6. Gabriel Zamora – apology video

Gabriel Zamora, an influencer with over 800 thousand subscribers, made an apology video for racist tweets he has posted in the past. By analyzing his apology video the data set above reveals that his most used strategies were “acknowledgement of responsibility” (p = 45%) and “explicit expression of apology” (32%). That being said, conclusions could be made towards the fact that those specific strategies are what contributed to bringing a positive light to his image because he is not just ignoring what he did, rather he is owning up to his mistakes and not making any excuses for himself. He recognizes his ignorance and explicitly states taking full accountability for it repeatedly throughout his video. Interestingly, he consistently used the phrase “truly sorry” 4/7 times he used the word “sorry”…“and for that i’m truly sorry…“the fact that i wasn’t [a positive creator] im truly sorry…” I believe his way of using it allowed for a greater expression of the extent to how regretful he actually was. Gabriel occasionally combined his strategies with “explanation or account,” his third most used (p = 13%) in efforts to disclose his true intentions…“i’m not a malicious person, i’ve never gone out of my way to try to bash someone in a racial way or in just a petty way in that sense.” Lastly, Gabriels usage in “promise of forbearance,” he promises his supporters/audience that he has educated himself and continues to do so. He also takes it upon himself to spread more awareness on the history behind the n-word by linking two educational videos about it in the description of the video. Although this seemed to be his least used strategy (p = 9%), the videos he linked helped boost his reputation by showing his audience that actions speak louder than words and he is moving towards the right direction to prove his growth.

B. Gender Differences

In Table 8, we calculated the ratio of each apology strategy usage by gender and came to the following realizations: 

  • The Frequency of Apologies: The result seems to buttress the theory that females were more likely to present explicit expressions of apology (Holmes,1989). However, as an evitable part of an apology video, the percentage of male and female using this approach is relatively close. From this point of view, gender seems to merely exert an influence on the linguistic differences of these stances (Stubbs, 2001).
  • Explanation or Account: Based on the data collections, females seem undoubtedly offer more verbal explanations or accounts than males. While most of the explanation or account for females is recounting the emotions, their apology video became longer and more complex. Females seem to include more verbal statements of concluding the whole controversial incident than males frequently may be less willing to use affirmative words to go over the account (Bennet, 2008). Overall, explanation or account drew huge linguistic differences between how males and females approach their apology.
  • Acknowledge of responsibility: Males present more acknowledgment of responsibility than females from the final data set. Gender differences in this stance are evident that males seem more likely to recognize their faults of actions. The mythological consideration for some females is vague. In Laura Lee’s video, she uses the denial strategies to shift the responsibility (Benoit, 2008), hence leading to an adverse audience reaction. In comparison, males seem more willing to acknowledge responsibility and re-evaluate the prior self who committed the transgression.
  • Promise of forbearance: Males perform more promise of forbearance than females. As usually the last part of an apology video, the promise of forbearance is crucial to provide the major idea of remedy to the transgression. Males seem more likely to provide corrective action and repair in the collecting data (Benoit, 2008). However, the overall proportion of promise of forbearance seems to appear identical for two genders without looking at the data of Shane Dawson who used this strategy the least. Hence a concrete conclusion is hard to draw from these 6 limited cases.

Discussion & Conclusion

The discussion on individual differences and gender differences are presented above. This section dedicated to looking at the audience’s overall comments reaction as outlined in Table 9, only Pewdiepie, Gabriel Zamora, and Jenna Marbles received a positive response. They all have different focuses on their strategies. Pewdiepie addressed most in the promise of forbearance, Gabriel Zamora focused on Acknowledge of responsibility and Jenna Marbles spent most of her video approaching explanation and account. Which might indicate promise of forbearance, acknowledgement of responsibility and explanation and account are more effective strategies. The overall conclusion is females are more likely to approach explanation and the frequency of apologies while males approach acknowledgement of responsibility and promise of forbearance strategies. However, all these conclusions just came from the limited 6 data sets and in a broader view, after excluding the extreme cases, most of the data appear to be identical between both genders (Holmes 1989). A key implication for this research is the importance of not only considering the difference in gender while looking at the apology videos, but also by looking at the approaches taken by different people on a larger view. The analysis and data for this paper might provide a good sample for further future studies of linguistics features in gender differences and apologies approaches.

References

Benoit, W. L. (2008). Image restoration theory. The International Encyclopedia of Communication.

Bennet, S. (2008). Gender and apologies: Exploring offended females’ perceptions of apologies from males and females. https://ro.ecu.edu.au/theses_hons/127

Cheng, M. (n.d.). The Stance of Personal Public Apology. Retrieved November 17, 2020, from https://scholar.uwindsor.ca/ossaarchive/OSSA11/papersandcommentaries/96/?utm_source=scholar.uwindsor.ca%2Fossaarchive%2FOSSA11%2Fpapersandcommentaries%2F96

Gabriel Zamora. (2018, Aug. 21). My Truth. [Video]. YouTube. https://www.youtube.com/watch?v=QWnmPEHzRrk&list=PLsuhXm2zs07IwijVL8Mkm4EnPab78F7TC&index=8

Holmes, J. (1989). Sex Differences and Apologies: One Aspect of Communicative Competence1. Applied Linguistics, 10(2), 194-213. doi:10.1093/applin/10.2.194

Jenna Marbles. (2020, July 2). Jenna Marbles Apology. [Video]. YouTube. https://www.youtube.com/watch?v=679d-SQfWLk

Karlsson, G. (2020). The YouTube Apology: analysing the image repair strategies and emotional labour of saying sorry online. Retrieved 2020, from https://www.diva-portal.org/smash/get/diva2:1483089/FULLTEXT01.pdf

Laura Lee. (2018, Aug. 20). Laura Lee apology video with original captions. [Video]. YouTube. https://www.youtube.com/watch?v=NYVmWxitVSQ&t=182s

Leppänen, S., Møller, J., Nørreby, T., Stæhrc, A., & Kytölä, S. (2015). Authenticity, normativity and social media. Discourse, Context and Media, 8, p. 1-5

Maclachlan, A. (2013). Gender and Public Apology. Transitional Justice Review, 1-21. doi:10.5206/tjr.2013.1.2.6

Pewdiepie. (2017, Sep. 12). My Response. [Video]. YouTube. My Response (Pewdiepie)

Shane Dawson. (2020, June 26). Taking Accountability. [Video]. YouTube. https://www.youtube.com/watch?v=ardRp2x0D_E

Smith, N. (2008). I was wrong: The meanings of apologies. New York: Cambridge University Press.

Tana Mongeau. (2017, Feb. 17). An Apology. [Video]. YouTube. https:/ /youtu.be/Fazh9Lm1kDE

[/expander_maker]

, , , ,

Gendered Use of Compliments and Insults in Professional Video Game Streaming

Kavi Dalal

This study examines male to male power hierarchy in online multiplayer video games. Using screen recorded footage of a professional gamer’s live broadcast as data in addition to transcription based conversation analysis, I present some observations on how compliments and insults are used in male socialized environments. The analysis sheds light on actual tactics employed by men in order to build solidarity and/or establish power amongst themselves. In conclusion I discuss the importance of continuing linguistic analysis at the intersection of gender and hierarchy in emerging online and male dominated environments.

[expander_maker id=”1″ more=”Read more” less=”Read less”]

Introduction

Gender inequality is a growing concern in the professional video gaming industry. Esports and professional gaming channels are growing more rapidly than ever as a form of globally reaching entertainment. Twitch, the most popular video game channel streaming platform, is presently ranked as the 32nd most traffic-heavy site in the world, ranking ahead of Twitter.com with millions of daily viewers and subscribers. Undoubtedly the market for professional gaming has grown to include a larger and more diverse following than ever. Still, male professional gamers continue to outnumber females by far in their field. According to the 2019 ESA annual report, female gamers represent roughly 46% of all video game players, yet only represent about 5% of the tactical shooter genre that is most commonplace amongst Esports competitions and professional competitive play. For this reason, sites like Twitch that broadcast professional gameplay videos are dominated by male-to-male dialogue between members of all-male gaming teams. These videos offer a unique window into the linguistic patterns of a highly gendered industry that is only growing in popularity and size.

Various studies have been conducted locating the meaning of compliments relative to gender and hierarchy in professional environments. Few however have analyzed male-male utilization of compliments and insults in a professional setting and none have used professional gaming as the sample for researching the operation of evaluative speech acts. Deborah Tannen and Janet Holmes are the loudest voices in academia when it comes to the gendered nature of complimenting. Both have proposed that women tend to perceive complimenting as an expression of positive affect or solidarity while men tend to view compliments referentially or with more emphasis on their objective informational content  (Holmes, 2008, p. 11). In You Just Don’t Understand: Women and Men in Conversation, Tannen (1990) argues that for men, complimenting is primarily about asserting one’s authority over the other through evaluation. Even when evaluating  someone positively, a person who gives a compliment is asserting that they have the authority to pass judgement on someone else. In return this causes men to occasionally perceive compliments as a face threatening act. Insults, another form of evaluation, are face threatening acts by nature. In an insult, the speaker gives a negative evaluation of some trait, possession, or behavior of their addressee, thereby attacking their positive face (Eckert & McConnell-Ginet, 2013, p. 187). This, coupled with knowledge that even positive evaluations can be used to assert dominance over an addressee, helps to explain why in interaction research, insults are viewed as a way to establish hierarchy and power. Perhaps surprisingly, however, many scholars have also theorized about how insults can be used to strengthen community bonds and establish solidarity. In The Hidden Life of Girls: Games of Stance, Status, and Exclusion, Goodwin (2006) describes both boys and girls trading mock insults as a way to practice verbal skills through play. Kochman (1972) has observed similar mock ritual insult exchanges between boys and theorizes that, while ritual insult can be used as a way to build bonds between addressors and addressees, even if an insult is intended as play, it may be taken seriously and seen as a face-threatening act. This risk is especially high when the insult just exaggerates an actual characteristic of the addressee. This research seeks to determine whether evaluative speech acts are used to build solidarity or enforce power differentials in an all-male professional game setting. Taking into account that there already exist observable power asymmetries between the owners of video game streaming channels and the other players they invite to play with them on their channel, research methods were designed to answer the following question. How are complimenting and insulting behaviors affected by the dominance status? Who pays more compliments and insults? Who is the typical addressee? Based on the prevailing theory that men typically use evaluation to assert their own authority to judge others, I hypothesized that both compliments and insults would flow down the power differential more freely than they flowed up it, and that ownership of a channel would contribute to the hierarchical power distribution.

Methods

Target Population

Research for this study was conducted by recording and analyzing gameplay dialogue between professional male gamers and their male teammates in multiplayer, first-person shooter games. Twitch is a live video streaming website specializing in E-sports broadcasting and personal streams of individual players known as “streamers”. The website operates on a channel and subscriber model in which a streamer runs a channel and amasses followers through streaming content and participating in tournaments. A streamer is able to monetize their channel through endorsing sponsors as well as being gifted small money contributions from subscribers. A typical stream session consists of broadcasted live game footage either solo or with teammates invited to play in a game broadcasted onto the channel. The video game that was chosen for this study was a first person shooter (FPS), battle royale style game titled Call of Duty: Warzone. Gameplay in Warzone is multiplayer, consisting typically of four teams fighting against each other to be the last one standing. Teammates communicate verbally through audio chat to strategize, but the audio communication is often used for socializing in less strategy demanding situations of gameplay. In the context of the gaming platform, the owner of the channel is superordinate to the guest players, and streamers with large followings hold particular status. As of June 16, 2020 TimTheTatman was the 8th most followed Twitch channel, boasting roughly 4.9 million followers and making him one of the most successful professional streamers. In this professional gaming environment, TimTheTatman’s role was analogous to a boss to his guests, some of which were professional streamers themselves but with smaller followings. Guest players were privileged to be on TimTheTatman’s stream and have exposure to his fanbase with no guarantee that they would be invited back again. In this context, the channel owner was the dominant player, and his guests were subordinates, or occupying a position of lower status.

Data Collection & Linguistic Units

This study analyzed six hours worth of gameplay dialogue between TimTheTatman and his channel guests. To collect data, instances of compliments and insults were recorded and tallied noting the speaker and the addressee. Addressors were broken into two categories: the dominant player (TimTheTatman) and non-dominant players (Tim’s three teammates). Addresses were broken into three categories: the dominant player, the non-dominant teammates, and the opponents (players on other teams that were encountered during the game). For this research a compliment was defined as a speech act that attributed credit from a speaker to an addressee, be it explicitly or implicitly, for some trait, action or possession valued positively by both interlocutors (Holmes, 1986, p. 485). An insult, on the other hand, was defined as “a negative appraisal and attack on the addressee’s positive face through implicit blame for what is being criticized” (Eckert & McConnell-Ginet, 2013, p. 187). Once the data was collected, certain calculations were required to accurately compare the data. The number of compliments/insult speech acts made collectively by all three guest players were subsequently divided by three to arrive at the mean number of compliments/insults made per guest player. Guests were not counted individually because there was no conclusive way to distinguish the voices of the three guest players on the audio chat. For that reason, the average number of compliments and insults per guest was calculated instead. The total number of speech acts by each type of speaker was also tallied, as well as the ratio of compliments to insults given by each type of player.

Results

Overall, the data from this study indicated that while the non-dominant player complimented others with more frequency than the dominant player did, the dominant player insulted his teammates more than non-dominant players did. Both dominant and non-dominant players complimented and insulted their opponents at roughly the same rate. As indicated in Figure 1, there was a significant difference in the frequency of compliments given out by the dominant player versus non-dominant players. On average, non-dominant players complimented their teammates and the dominant player twice as often (4 times) as he complimented them (twice). Interestingly, both dominant and non-dominant players complimented their opponents at exactly the same rate (twice). There was no difference between the rate at which the dominant player complimented his teammates and his opponents. However, non-dominant players averaged slightly more than twice as many compliments for their teammates as for their opponents.

By contrast, Figure 2 shows that the dominant player insulted his teammates at a much higher rate (7 times) than the average non-dominant player insulted him (2.3 times) or other non-dominant teammates. Both the dominant player and the non-dominant players insulted their opponents at roughly the same rate (2 and 2.6 times respectively), and interestingly, this was similar to the rate at which both speakers complimented their opponents. The dominant player insulted his teammates at more than three times the rate that he insulted his opponents. The average non-dominant player insulted the dominant player slightly less than he insulted his opponent, and insulted other non-dominant players even less than that.

In total, dominant and non-dominant players engaged in a similar number of evaluative speech acts. As is visible in Figure 3,  the dominant player engaged in a total of 13 evaluative speech acts, and the average non-dominant player engaged in an average of 16.7 evaluative speech acts over the course of six hours of gameplay. The preferred type of evaluation differed by addressor, however. 

Figure 4 shows that 69.2%, or slightly more than two thirds, of the dominant player’s evaluative comments were insults and only 30.8% were compliments. Conversely, only 36%, or slightly more than a third, of the average non-dominant player’s evaluative comments were insults, while 64% were compliments.

Discussion

In setting out to conduct this research, I hypothesized that evaluative speech acts would be used in all-male gaming settings to assert power and reinforce hierarchy. I expected that the dominant player would engage in more evaluative speech acts than non-dominant players did. The data suggests, however, that the overall frequency of evaluative speech acts does not reflect the hierarchy within this setting as much as the types of evaluations and whom they were directed to do. Both the dominant and non-dominant players had similar numbers of evaluative comments, and, in fact, the non-dominant players made ever so slightly more evaluative comments.  The fact that dominant and non-dominant players both complimented and insulted their opponents at a similar rate suggests that evaluating individuals outside of a group is a low-risk way for all players, regardless of hierarchical status, to build solidarity amongst individuals within the group. Goodwin (2006, p. 232) reinforces how insults can function to unite those laughing with the insulter while othering the target. Contrary to my hypothesis, the data showed that while the dominant player had a higher tendency to insult teammates, the average non-dominant player had a higher tendency to compliment. One explanation is that,  “implicit in any evaluation is a claim on the part of the evaluator that he or she is in a position to judge the evaluatee. And taking an evaluation seriously attributes this position to the evaluator.” (Eckert & Sally McConnell-Ginet, 2013, p. 180). In other words, the dominant player’s frequent use of insult seems to support the interpretation of evaluations as speech acts used to assert power.

However, the frequency of compliments from the non-dominant player directed toward the dominant player raises questions about this interpretation. One potential explanation is that non-dominant players used compliments to facilitate interaction and create solidarity within the gameplay. This type of compliment use has been frequently observed within groups of women, as well as in co-ed groups where women take on the role of the ‘interactional shitworker’, instigating and facilitating communication between the parties present (Fishman, 1978, p. 398). Given the inferior status of non-dominant players within the Twitch power hierarchy, it seems likely that these players use of compliments in this setting is evidence that they were attempting to deliver positive affect compliments, which have been typically gendered as a more feminine use of complimenting, (Holmes, 2003, p. 143). The difference between the use of compliments as a solidarity building linguistic act as opposed to an evaluative linguistic act is illustrated in the excerpt below.

Excerpt 1

3:59:43-4:00:10

TIM=TimTheTatman        PL1=guest player 1 
PL2=guest player 2      PL3=guest player 3

01  PL1:     Tim the Tatman’s~cookin ~now~uh-

02           ((Tim’s character dies))

03  TIM:     I got sniped at the same fucking time I just want to 

                                                      fuck myself

04           baby, YEAH:::=

05  PL1:     =(h):::m (h)m (h)m? 

            ((laughter followed by 6x slow claps))

06  TIM:     Put it right in my f(h)ucking a:::ss.

07  PL2:     Alright calm down for two seconds I’m coming.=

08  PL3:     =I’m stayin here cause they’re hunting me

09  PL1:     hhhhu hhh (1) hhhhhe:: ((laughter))

10           ((Tim gets revived by PL2))

11  TIM:     Hey thank you Matt you’re a good friend.

Excerpt 1 opens with a compliment from Player 1, a non-dominant player, about Tim, the dominant player. Player 1 observes that Tim is “cookin,” a metaphor implying that Tim is playing well. Although the compliment is about the dominant player, it is not addressed directly to him. Rather, it is addressed to the group and names Tim in the third person. This, coupled with the fact that the compliment evaluates Tim’s playing generally without describing any specific feature of his gameplay, suggests that Player 1 is using flattery to create solidarity with Tim rather than to evaluate him objectively. “Giving praise is inherently asymmetrical,” and compliments given from a high hierarchical position to someone lower are called “praise” while a compliment from a lower position upwards is called “flattery”, (Tannen, 1990, p. 69). Immediately after Player 1 flatters Tim, Tim makes a mistake and his avatar dies. Tim acknowledges the mistake and then adds “I want to FUCK myself baby,” followed by “put it right in my fucking ass”.  Often, men use sports metaphors to describe sex, but in this example the inverse is true (Eckert & McConnell-Ginet, 2013, p. 250).  Sex, specifically the act of being penetrated, is a metaphor for losing, or dying, in the game. Tim uses misogyny to liken himself to a woman or other passive participant in sex. His use of profanity and hyperbole detracts from the sincerity of the admission that he made a mistake, and therefore diverts blame away from himself. Ironically, by linguistically equating himself with a powerless participant in a graphic sexual act, he is able to save face by avoiding a sincere apology or acknowledgement of his mistake. This outburst spurs Player 2 to put his own avatar at risk to revive Tim’s player, after which Tim says, “Hey thank you Matt, you’re a good friend.” In contrast to Player 1’s compliment, this statement is directed at its subject and directly acknowledges a specific helpful behavior from Player 2. This is a rare instance of compliment from the dominant player, and is in keeping with the observation that when compliments are less frequent, they are more likely to be referentially oriented or genuine expressions of admiration (Herbert, 1990). This compliment garnered no response from the addressee or the other players, which is typical of most compliments in this setting apart from the occasional expression of gratitude. This evaluation allows Tim to assert his authority to evaluate Player 2. Perhaps he does this in part to recover his face after having lost agency in the course of gameplay.

Excerpt 2 illustrates the ways that insults are used to assert power and establish solidarity. It is an outlier situation in which we get to see both dominant and non-dominant players insult each other.

Excerpt 2

4:19:55-4:20:21

TIM=TimTheTatman     PL1=guestplayer 1

01  PL1:     Tim you’re always nowhere near us [fighting people

02  TIM:                                       [ºsh:::: I got this 

                                                          shit bro

03  PL1:     ((sarcastic)) Oh here we go

04           ((tim kills opponent))

05  TIM:     wha what did you say Matt,

06  TIM:     ((mocking)) Oh here we go. Yeah look at that shit bro

07  PL1:     Tim I gotta be honest with you man 

08           like know your truth. You die a lot=

09  TIM:     =No I do not.

10  PL1:     Tim there’s another guy there’s another guy below 

11           you. I mean you are deaf as a fucking.

12  TIM:     under me::?

13           (3)

14  Tim:     ºI’m so confused bro

This excerpt begins with a non-dominant player implying that Tim is too far away from his teammates, to which Tim tries to reassure Player 1 that he “got this shit,” and is therefore in control, not a liability. Player 1 responds in an exasperated tone, implying that he doesn’t trust Tim not to mess up. His use of sarcastic tone suggests this is an instance of an off-record request using irony (Brown and Levinson, 2014). The implied request is that Tim not enter into combat by himself. Rather than accepting the request, Tim quickly questions and repeats Player 1, in effect insulting and mocking him. Even though non-dominant players rarely insult Tim, overtly or implicitly, in this instance, Tim responds to Player 1’s suggestion as if it were an insult. This is a reasonable reaction  considering Eckert & McConnell-Ginet posit that “comments can be taken as serious insults even if not so intended” (Eckert & McConnell-Ginet, 2013, p.188) . He sees it as a face threatening request and he questions player 1’s right to challenge Tim. He then mocks Player 1’s indirect, less confrontational, and more stereotypically feminine speech style by repeating the phrase “Oh here we go,” in a mocking tone. This indirect insult reaffirms Player 1’s subordinate status. What follows in lines 6 to 11 is an escalated series of insults from Player 1 and more deflections from Tim. Given the tone of the insults, this seems to be an example of “mock ritual insult” (Kochman, 1972, p. 314). The teammates seem to be verbally jousting more than they are giving serious insults. Tim holds his face throughout the whole altercation. He didn’t give legitimacy to any of the insults by evaluating them as false. In this way he was able to maintain face by resisting imposition (Brown and Levinson, 1987).

Conclusion

The total amount of evaluative speech acts had no bearing on enforcing power differentials in an all-male professional game setting. Insults flowed down the power differential more frequently as expected, but contrary to my hypothesis, compliments flowed up it more frequently. This was ultimately attributed to non-dominant players’ assumption of a more typically feminine speech style that utilized compliments effectively to boost solidarity. This exposes how gendered hierarchies are present in language between men, even when no women are present.

One limitation of this study was difficulty distinguishing the voices of guest players. In a future study a stream in which the voices of players could be distinguished via timbre and pitch would be preferable. This study also only analyzed one small slice of the gaming world. Future studies could benefit from analyzing a wider spectrum of games and streamers that might reflect different power hierarchies. In addition there are more nuanced speech acts such as declaratives which were far more frequently occurring than compliments and insults. I would encourage subsequent studies to analyze other evaluative speech acts in male-male gameplay and how they operate to assert a hierarchy. This research unsurprisingly shows that language between males in professional gaming ascribes to strict patriarchal tendencies. This is important to understand in the growing field of professional gaming, and this language must be challenged if women are to have a more representative presence in the profession. Certain Twitch streamers such as KittyPlays are paving the way for the next generation of female pro gamers by challenging sexist language as it is encountered real time during her stream. Nonetheless, given that hidden biases are likely to perpetuate in this domain through language even if there are more professional female players,  more studies should look into gender hierarchy’s implications on the gaming world given its influence over language in popular culture.

 

See also:

HALO 3: Negative comments by gender

SEXISM IN VIDEO GAMING: Online harassers are literally losers?

 

Bibliography

Berger, P. L. and T. Luckmann (1966). The Social Construction of Reality: A Treatise in the Sociology of Knowledge. Garden City, NY: Anchor Books.

Brown, P. & Levinson, S. C. (1987). Politeness: Some universals in language usage. Cambridge: Cambridge University Press.

Eckert, P. & McConnell-Ginet, S. (2013). Language and Gender. Cambridge: Cambridge University Press.

Fishman, P. (1978). Interaction: The Work Women Do. Sociolinguistics by N. Coupland and A. Jaworski. London: Palgrave.

Goodwin, M. H. (2006). The Hidden Life of Girls: Games of Stance, Status, and Exclusion. Oxford: Blackwell.

Herbert, R. K. (1990). Sex-based Differences in Compliment Behavior. Language in Society, vol. 19. Cambridge: Cambridge University Press.

Holmes, J. (2003). Complimenting: A positive politeness strategy. Sociolinguistics: The essential readings. ed. by Christina Bratt Paulston and G. Richard Tucker. Malden, MA: Wiley-Blackwell.

Holmes, J. (2008). An introduction to Sociolinguistics. London: Pearson Education Limited.

Kochman, T. (1972). Rappin’ and Stylin’ Out: Communication in Urban Black America. Chicago: University of Illinois Press.

Tannen, D. (1990). You just don’t understand: Women and men in conversation. New York: William Morrow.

twitch.tv Competitive Analysis, Marketing Mix and Traffic – Alexa”. www.alexa.com. Retrieved June 16, 2020.

[/expander_maker]

 

 

Scroll to Top