Slur Frequency in Popular Music of 2005


Visualization

Recently, society has been catching up on the subject of cultural appropriation – what it means, the many forms it takes, and why it’s so damaging. Of all the conversations and debates swirling around this domain, the topic of slurs – specifically the N word – continues to be a sensitive and sometimes polarizing topic. There are groups of black people who bring up that “nigger” and all its forms is a historically derogatory and painful term, and that it should never be used. Conversely, other black people insist on reclaiming the word in order to dilute its negative connotation, and use it freely in conversation and all types of media. The problem with including a historically loaded term in popular types of media is that you can’t police who sees what. Popular music, movies, and television is viewable to all audiences, and therefore all the different types of people who consume the term in the context of these mediums feel like they are free to use the term as well. This fuels the “You Can’t Say It But We Can” argument, where black communities get upset at other people who say “nigga” because ‘it means something,’ but then sling the term around freely themselves. This creates an “imbalance” that some people find unfair, because if it’s such a serious word, then why does it crop up everywhere? It must not be that bad.
With this in mind, I wanted to take a look at popular music and see exactly how often “nigga” comes up on the charts and also take a look at any patterns related to its presence and/or staying power. I thought it would be interesting to track the appearance of the word “bitch” as well, since that’s another term that has a varied semantic history. Using Billboard’s Hot 100 charts, I sourced my data. Billboard updates its lists weekly, so I chose the first week of every month in 2005, and then looked up lyrics of the 100 songs in every month. Any song that had an occurrence of either was marked, and a tally was kept for each term.

Since my major interest was to see any notable changes in language use over time, I looked for visualizations that had loosely similar content. The first I found was from someone’s personal blog where popular slang terms were tracked over time. slangTrendsAlthough interesting, there was no other information available about this graph and also no information about how much data was collected or where it was sourced from.reddit graf

Luckily, the next viz I found was much clearer about what information was being shown. Writers from the FiveThirtyEight blog combed message boards from the social network Reddit and dissected each comment posted from 2007 to mid 2015. With over a million comments collected, there is a huge amount of data to explore, and this viz lets you do exactly that by allowing you to search any term you want. Each user can then interact with the retrieved data and see how the use of particular terms intersect or increase over time.

 

Lastly, probably the best viz I found also had the most closely relevant content to what I wanted to show: A writer from besttickets.com created a series of visualizations on profanity used in popular music. His focus was specifically on rap music, and also on a smaller amount of songs since only the “five most influential” rap albums of each year were chosen. In addition, there was a wider allowance for what was considered profane, with a range of terms chosen instead of just a focus on just one or two terms. Despite the difference in data focus, all the visualizations were super interesting; profanity amounts were sorted by album, artist, year, and also over time.

Final-GIF

 

I really liked the effect of stacking “profane” data on top of total data collected,total nterm frequency by month so I wanted to incorporate that into my graphs as well. To start, I made bar graphs that showed the total amount of occurrences of each slur per “month.” The below visualization shows the total amount of the word nigga in each month- so this counts each occurrence in each song charted. Since each month’s hundred songs are only sourced from the first week of Billboard’s data, the occurrence amount can change drastically with the inclusion of one song that might have charted low but still includes a lot of profanity.  I was surprised to see that January was the most profane month of the year with 181 total occurrences; I thought for certain the “summer jam” phenomenon would yield more profanity in summer months since rap and r&b music is usually favored during that time.

I made this same graph again to lookScreen Shot 2016-06-22 at 10.29.24 PM at occurrences of the word ‘bitch,’ and January still came out on top as the most profane month. To get a better idea of how many songs were contributing to this, I made a stacked bar graph showing how many songs overall included the word – 17 songs for the month of January. I also looked at how ‘bitch’ songs charted in each month to see if there was any correlation between content and popularity… there didn’t seem to be any. None of the number one songs included the word bitch all year.

Conversely, the same examination of top ranked songs including the word ‘nigga’ yielded two results; “Gold Digger” by Kanye West ranked number one in both October and November. Looking further into how highly ‘nigga’ songs rank overall, I made a scatter graph to show how many occurrences of the term appears in each song, and how high the chart position was.

“Gold Digger” is visible on this chart, appearing at the top with a number one ranking and a slur count of 12. This is actually the most successful profane song of the whole year, since other songs that use the term more than 12 times never rank within the top ten songs of the month. “U Don’t Know Me” by T.I. has the Screen Shot 2016-06-22 at 10.39.00 PMmost term occurrence of the whole year with 30 but only charted as high as #28. It seems like most of the top charting songs have a “modest” amount of slur usage, as many hits that have 3-8 occurrences manage to make it into the top ten and stay there for two to three months, as seen with “Disco Inferno” by 50 Cent and “Drop It Like It’s Hot” by Snoop Dogg and Pharrell.

If I were to go forward with this project I’d like to examine the ways in which each terms are used – are they positive, negative, or neutral? Popular lyric analysis archive Rap Genius includes annotations for most lines of the songs included, so listeners can see the meanings behind lyrics of popular songs. All information in the Genius universe (it has since expanded from just hip hop and rap) is crowdsourced, so whatever information you read has been agreed on by many people to be the true meaning. Screen Shot 2016-06-22 at 11.02.57 PM In addition, “verified” annotations appear on some songs, with notes from artists or producers that tell a more behind the scenes story for the making of the song. I think it could be very telling to see how people interpret (or ignore) the semantics of each term, and then see whether or not a certain type of usage appears in popular music more than others.

Another interesting thing to examine would be the appearance of slur occurrences within song structure: Are terms more pervasive if they appear only in hooks and choruses of songs? Do other occurrences within the song matter more or less, and how does all of that contribute to song popularity?