3. making sense of the data ; index ; 5. the word lists

4. discussing the results

  1. Presentation, summary, and discussion of each word list
  2. What can I apply these words to in the compositional process?
  3. Other interesting results of the word search

The word lists: summary and discussion

To make the final word lists that would be used for writing the songs, I took a subset of the classified words, added some words that I found using the coexistence scripts, and put the resulting words into five categories.

The five word lists are:

  1. Instruments
  2. Sound
  3. Structure
  4. Mood
  5. Vocals

The words in each list are sorted by wordscore. Words that are similar in meaning and usage are put on the same line with their wordscores added together. When a word appears on these lists, it means that Pitchfork critics are more likely to use that word when describing records that they like.

The Instruments word list
All of the usual rock and roll instruments appear on this list (guitar, drums, bass, and vocals, in that order), but they appear alongside some instruments that aren't usually associated with popular music (piano, strings, horns, violin, trumpet, cello, accordion, flute, and clarinet). There's also a number of electronic music words (electronics, keyboards, samples, loops, sequencing, and breakbeats), and one word that references acoustic instruments (acoustic).

There are a number of basic musical elements on this list (melody, rhythm, groove, harmony, and texture). Not surprisingly, the critics are suckers for a good melody (it's #3 on the list, only below guitar and drums). There's also a very strong emphasis on groove (beats, rhythm, groove, and danceable).

Another interesting thing is the presence of processing/effects words (ambience, effects, processing, and reverb); apparently, Pitchfork reviewers are very fond of recordings that have artificial effects on the instruments. There are also several words which point in more “experimental” directions (noise, drone, feedback, and silence). “Noise” is more than half the score of “melody”, and “drone” isn't far behind, so the Pitchfork critics definitely want to hear some experimental elements in their pop songs.

The Sound word list
“Warm” occupies the first slot, which is sort of a joke among audio engineers because it's a silly buzzword that audio companies use to describe their products. (The meaning of “warm” seems to change depending on the sound of the product being described.) Aural warmth is actually related to the second word, “distorted”, since things which are distorted in certain ways can sound “warm” to some people (the “musical” distortion made by tube electronics and certain vintage analog gear is often described as “warm”).

As in the Instrument word list, there's an emphasis on artificially effected sounds (echoes, lush, ethereal, clouds, murky, fuzzy). There are also several words that refer to the “heaviness” of the sounds (gentle, heavy, soft, fragile, understated, hushed). Most of the remaining words are adjectives used to describe either high frequencies (buzzing, shimmering, cutting, chiming, skittering) or low frequencies (pounding, pulsing, rolling, rumbling).

The word “organic” is used sort of as an opposite to “sterile” and suggests that a band playing in a room and working off of each other to produce music is favored over the “mad perfectionist genius” solo approach.

It's interesting that “backwards” shows up on this list, and it's commonly used in the same phrase with instrument words like “guitar” and “drums”.

The Structure word list
Above all, the critics want you to be subtle in your approach (subtle, subtly, subtlety). I suppose this means that I shouldn't beat people over the head with my innovative new approach to pop music. The #2 spot of the word “layers” plus the high score of "dense" might suggest that the critics are fond of recordings that feature mounds of overdubbing.

Word group #3 (builds, swells, crescendos, rising) are all words that refer to getting louder over a period of time, and group #4 (blast, blasts, crashing, explosive) refers to getting louder instantaneously. This probably means that successful pop songs have to get louder over time. This especially makes sense when taken along with #7 (tense, tension). Many great pop songs build tension over time and then release that tension, and the “release of tension” parts of songs are often found at the end of a crescendo.

“Complex” and its synonyms get a very high score, but the word “simple” is right under it. This actually happens for a few different opposites in the word lists, which would suggest that extremes are generally favorable no matter which side of the extreme you happen to be emphasizing.

There are many high-scoring words that mean “unpredictable&rdquo (unexpected, unpredictable, chaos, chaotic, and possibly abstract, shifting, and disparate), and the antonym “predictable” gets a very low score. This seems like common sense (“Predictable music is boring!”), but adding too many unexpected elements into a song can make it unlistenable, so it's hard to say how much chaos the critics really want to hear.

There are several words that mean “large,” some of which are metaphors (massive, sky, spacious, ocean). Apparently, music that sounds small is less desirable. Largeness in music usually refers to a considerable difference in loudness between soft and loud sections (as in “huge chorus”), a difference in loudness between certain instruments and other instruments (as in “huge guitars”), or a recording that represents all of the audible frequencies equally (where “tinny” is the opposite of “huge”).

There are positive values for “rough” and “primitive,” and negative values for the words “shiny” and “polished.” This points towards a preference for lo-fi recordings, which are usually associated with lower-budget independent music. This falls in line with the Pitchfork reviewers' dislike of capitalism, which I talk about a bit in the other interesting results section below.

I actually find the structure word list very inspiring; when writing the songs for this project, I studied this list as a way to find new compositional directions.

The Mood word list
The most striking thing about the mood list is the proliferation of unpleasant emotion words. I divided the moods into groups, and the "happy" group is #6 underneath sadness, dark, crazed, mysterious, and violent; it's followed by scary, anxious, angry, and emotional.

The “sadness” group is by far the highest-scoring mood, beating the next mood (“dark”) by over 1100 points. As a response to that, I've tried to make these songs as sad as possible.

One of the groups near the middle of the list is “confidence,” so it's better to sound confident rather than insecure.

The Vocals word list
In order, the most positive vocal qualities are whispering, crooning, wailing, and chanting. The word “hushed” could also be considered a part of the whispering group.

There is also a helpful collection of negative vocal words: the Pitchfork critics aren't fond of nasal vocal tone, yelling, off-key vocals, or screaming.

Male vocal words rank higher than female vocal words (tenor, baritone), and in both male and female cases, high-pitched singing (tenor, soprano) gets better ratings than lower-pitched singing (baritone, alto). These wordscores are somewhat thrown off by the usage of the same words to describe saxophones in jazz music, and the female vocal words seem to be more susceptible to this problem than the male vocal words.

What can I apply these words to in the compositional process?

Just for my own benefit, I sat around and brainstormed up a list of musical variables that could be modified using the word lists. In case you're curious, the results of that brainstorm are below.

quality of a composition examples of concepts describing this quality
tempo fast/slow, steady/variable
pitch in tune/out of tune, major/minor, mode, phrase
timbre/production frequencies, harsh/subdued, tone, polished/rough, hi-fi/lo-fi
harmony harmony/dissonance (of pitch, chords, timbres, and many other structural elements)
structure familiar/strange, intro/verse/chorus/bridge, ambient, repetitive/linear, cluttered/orderly, complex/minimal
instrumentation instrument words, e.g. guitar, snare, flute, violin, cymbals
virtuosity tight/loose, difficult/simple
lyrics/vocals are lyrics present? subject matter, mood, tone/sound, energy, meaningful/nonsense
mood both musical and lyrical mood communicated via subject matter, key/mode, tempo, structure, etc.
image attitude, personality, political affiliation, luddite/technophile, high-budget/low-budget

Other interesting results of this word search

I find myself wondering why a critic would be more likely to mention ubiquitous instrument words like “guitar” when they're talking about a record that they like, and then not mention those words when talking about a record that they don't like. This is likely connected to the phenomenon that the Pitchfork critics are 3.8 times more likely to use a meaningless value judgment word when they're describing something they don't like. The critics are much more likely to describe the music itself when they like it; if they don't like the music, they aren't nearly as likely to go into detail about it in the review.

When the Pitchfork reviewers write about a record they don't like, they're more likely to mention two things:

  1. words referencing consumerism and business, and
  2. words insulting the intelligence of the musicians or listeners.
The former expresses an anti-capitalist slant, and the 9 words that I identified as part of this group have a significant wordscore when added together (-4661.02; the word “sell” alone scores -1432.56). These anti-capitalist words mostly refer to advertising, marketing, and selling music.

The intelligence words also score pretty terribly, and only two of those words (intelligent, genius) are positive. The rest of the 12 words in that group are fairly insulting (y'all, retarded, frat, vapid, mindless, etc.).

There are many high-scoring words that refer to non-rock genres of music that are usually considered more “academic” or “high-art” (avant-garde, improvisation, classical, and jazz). I'm not quite sure what to think about this. It seems that this phenomenon sits along with the pro-intelligence words above, and that the positive usage of these words conveys support of high-art influences in popular music.

On a related note, the small number of jazz records that are reviewed on Pitchfork have a much higher average rating than the other records. The artificially high ratings of records in a certain genre points to a fear of bashing these records, perhaps due to a lack of skill when criticizing these records. Alternately, maybe Pitchfork only chooses to review known-good jazz records, which makes sense since they aren't a jazz publication and wouldn't feel obliged to review all the new material in that genre.

I don't understand why the word “lyrics” gets such a low wordscore (-3141.48). I have a few theories as to why this might be, but I really don't have it figured out. Do Pitchfork critics prefer lyrics that they don't have to talk about, i.e. lyrics that aren't a large part of the music? Looking through the reviews never gave me that impression, so I'm not sure what to think.

The words “reissue” and the symbol “r” (which is used by Pitchfork to signify a reissue date in the top of an article) both have very high wordscores. This may mean that reissued material is assumed to be good simply because it's been reissued.

Sadly, female pronouns score very poorly compared to male pronouns: 

  1. his  22048.40

  2. he   2950.50

  3. she  315.38

  4. her  276.60
There aren't as many women playing music as there are men, and it looks like Pitchfork haven't gone out of their way to cover female artists in a positive light. (The word “her” appears 23.3% as much as “his”, but only has 1.25% of the wordscore, which means that male pronouns are associated with 18 times more rating points than female pronouns.)

3. making sense of the data ; index ; 5. the word lists