Author: Ethan Hein

Jazz Remixes

Jazz Remix

Photo by Jimmy Baikovicius, via Flickr

I don’t place a lot of value on originality in music. My tastes lie mostly in blues, jazz, R&B, and hip-hop. While there’s plenty of creativity in all of these forms, it’s built around shared musical materials: stock licks and phrases, standard song structures and schemas, frequently borrowed beats and samples. Hearing a familiar blues riff or funk break is like encountering an old friend, and the intertextuality created by all of the shared musical DNA enriches the listening experience.

The title of this post could be read a couple of different ways. You could take it to mean “people who electronically rework jazz recordings.” Jazz has certainly been a bottomless source of inspiration for hip-hop producers. A significant portion of my own creative output is based on samples of my favorite jazz recordings.

By the way, if you’re looking for a good break, let me recommend the drumming of Sam Woodyard in the late-period Duke Ellington Orchestra; he’s a gold mine.

Really, though, the title of this post refers to jazz musicians themselves. Jazz is all about repurposing pop and folk material for new expressive ends, and the greats were remix artists before the term existed. Even the most prolific and brilliant jazz composers, such as Miles Davis and Thelonious Monk, devoted album after album to arrangements of standards. Nobody arranged standards more radically and personally than John Coltrane.

Some of Coltrane’s most compelling statements of musical truth are renditions of extremely corny pop songs. The best known one is “My Favorite Things,” from his 1961 album of the same name.

Coltrane’s arrangement of this tune bears the same relationship to The Sound Of Music as “Hard Knock Life” by Jay-Z bears to Annie. Jazz uses different technology than hip-hop, but it makes the same musical statement: putting a stamp of personal ownership on a piece of public musical property. The lawyers among you will probably now want to jump in and point out that neither The Sound Of Music nor Annie are public property. Even though both function the way that folk songs do, they’re both very much under copyright. Music has always consisted of endlessly reinterpreted and recombined folk memes, but now most of the really good memes are privately owned. It makes for some legal and cultural awkwardness.

I bought My Favorite Things when I was eighteen or nineteen after reading a Jerry Garcia interview in which he raved about it. (I have never been steered wrong on a music recommendation by Jerry Garcia.) On my first listen, I wasn’t impressed. A show tune I sang in middle school chorus played on soprano sax, whee! Now I experience Coltrane’s version of “My Favorite Things” as the mind-expanding flight of imagination I was promised, but I had to grow up a little to appreciate it. And appreciate it I did, to a point of near-obsession. When I had a jazz band, I insisted that we perform it regularly, and that we include it on our one album.

Coltrane had a way of anticipating what music would sound like in the future. He was particularly prescient about the importance of looped bass lines. Jazz bass is usually a complex semi-improvised stream of quarter notes. But Coltrane liked to have his bassists play strictly unvarying two-bar loops. On “My Favorite Things,” Jimmy Garrison plays a few simple octave patterns on the root and fifth of the key with no variation for the entire duration of the song. This kind of bass line anticipated the looped, sequenced, and sampled bass parts in hip-hop and other electronic music.

Coltrane was also prescient in his liking for open-ended loops on a single chord, or a few repeating chords from a single scale. This is the basic structure of nearly all forms of electronic music, and most contemporary pop too, but in 1961 it was a radical departure from most of the music in the air. Like James Brown and the hip-hop artists he inspired, Coltrane relied a lot on the “repeat until cue” instruction. In between the phrases of the “Favorite Things” melody, he inserts open-ended grooves, first on E minor, then on E major. He and McCoy Tyner play each groove as long as they want, signaling the band that it’s time to continue to the next section by playing the “raindrops on roses and whiskers on kittens” melody. Coltrane was a great admirer of Ravi Shankar, so much so that he named his son after him, and you can hear the influence of Indian classical music on this recording.

The album My Favorite Things is most famous for its title track, but it also includes three other startling reinterpretations of standards. “Every Time We Say Goodbye” is played double-time at an extremely slow baseline tempo, stretching the melody like Silly Putty. “Summertime” is played fast, with an angry feel and crunchy, dissonant chords built from the melodic minor scale. Finally, “But Not For Me” is transformed almost as radically as the title track. Here’s a conventional version of the tune by Judy Garland:

And here’s Coltrane’s version:

The most obvious change is the first four bars. In the Gershwin tune, the line “They’re writing songs of love but not for me” runs over a simple ii-V-I progression in E♭. Coltrane’s first four bars are a sprint through the keys of E♭, B, and G via those keys’ respective dominant chords. The bass line spells out the descending E♭ whole-tone scale: E♭, F♯7/C♯, B, D7/A, G, B♭7/F, E♭. Coltrane rewrites the melody completely to fit this new chord progression. Coltrane also inserts some new structural elements of his own. He adds a long tag section where he lifts unexpectedly up to several distant minor keys for eight bars each. There’s also the extremely extended open-ended tag on the ii-V-iii-VI turnaround. Should we consider Coltrane’s arrangement to be the same piece of music as the Gershwin original? Certainly, if you want to play the Coltrane version at a jam session or a gig, you’d better come prepared with charts and a lot of explanation.

Radical jazz adaptations of standards raise the same questions about authorship and ownership that sample-based compositions and remixes do. Where do you draw the line between an arrangement, a new melody written to existing chord changes, and an improvised solo? Bandleaders like Duke Ellington and Benny Goodman routinely used improvised phrases by their band members as the basis for new tunes (and were not overly concerned with crediting those band members). The borders between arrangement, interpretation, improvisation, and composition are blurry at best. Should we consider “Whispering” and “Groovin’ High” to be the same song? How about “I’m In The Mood For Love” and “Moody’s Mood For Love”? Or “I Got Rhythm” and the uncountable bebop heads it inspired?

Jazz was largely built on a scaffolding of show tunes and other pop songs. The ones that have emerged as standards share certain musical characteristics that make them more amenable to jazz adaptation. They have singable melodies with rhyming lyrics accompanied by a simple chord progression (or sometimes not so simple, but always intelligible to the ordinary person’s ear). They’re repetitive and predictable. They follow a small set of conventions in their structure: four-, eight-, and sixteen-bar phrases, repeated two or three or four times, with the larger grouping of phrases repeating more or less intact for the entire duration of the tune. There are some recurrent harmonic tropes involving counter-clockwise trips around the circle of fifths. The modular structure of standards makes them amenable to disassembly and reassembly. These jazz compositions and improvisations are constructed from a giant box of shared musical Legos, rearrangeable at will on paper or in the improviser’s head.

As with harmony and form, there’s a finite toolbox of riffs, patterns, and scale runs you can use to build your jazz melodies and solos. Blues is particularly reliant on Lego-like modular riffs. Jazz and blues intros and endings are few, highly standardized, and easily interchangeable. One much-recycled ending is the one Count Basie uses in his performance of “Fly Me To The Moon” with Frank Sinatra.

Another basic Lego is the Duke Ellington ending, as in “Take The A Train.”

Miles Davis turned this ending into an entire bebop head called “The Theme.”

You can hear a stretched-out but still recognizable rendition of “The Theme” at the end of each set on Miles’s Fillmore East performances from March 7, 1970.

The distance between a jazz module like the Duke Ellington ending and a sample like the Funky Drummer break is short. In my own experience, the creative process of writing a jazz tune based on licks and progressions from existing songs feels much the same as building tracks from samples. It can’t be an accident that the most creative jazz musicians are the ones who borrow the most heavily from one another, from pop culture, and from themselves. Coltrane’s tune “Impressions” is a mashup of “Pavanne” by Morton Gould and “So What” by Miles Davis (which is itself partially inspired by “Pavanne.”) The 6/8 single-chord grooves in “My Favorite Things” also appear in Coltrane’s versions of “Greensleeves,” “Spiritual,” and “Afro Blue.” And why not? That groove never gets old. If the most creative artist in the history of jazz is doing so much sampling, I think everyone should feel emboldened to do the same.

Brahms’s Third Racket


Photo by Susanne Nilsson, via Flickr

A few weeks ago, my research group at NYU hosted an event with called Musical Shares. Everybody starts a session in some digital audio workstation. (We used GarageBand.) You spend ten minutes creating a track. Then you share it with the person to your left, and you receive a track in progress from the person on your right. You work on your neighbor’s track for ten minutes, and then you pass left again. You repeat until you run out of time. Finally, you listen to your original track and experience the requisite surprise, delight, or horror. Musical Shares is a combination of Exquisite Corpse and Telephone, and it’s a lot of fun. I subsequently tried it with my students, and it was an instant success. If you’re a music teacher, I recommend you try it, too.

A colleague of mine, Dafna Naphtali, happened to pass through the conference area where we were doing the event. Dafna is an experimental electronic musician and vocalist who runs the NYU laptop orchestra. She thought the Musical Shares concept was cool and wondered if we could dream up something similar for her ensemble. (Many interesting opportunities have come my way just from loitering around that NYU conference area.) We couldn’t think of any obvious way to adapt Musical Shares for a live performance situation, but Dafna asked me if I would try to think up something else. I’ve never written anything for a laptop orchestra, so I figured, why not.

My idea of a laptop performance is a DJ set, or some other dance-oriented form of controllerism.

Hip-hop and EDM are not what Dafna had in mind, however. In fact, her only stylistic guideline was: no beats. This was a tough one, since she was asking me to do without the sole unifying characteristic of all of the music I make and enjoy. It would have been easy for me to come up with some arbitrary set of instructions that the laptop orchestra could turn into chaotic digital screeching, but I wanted to write something that would sound “musical,” by my relatively narrow-minded notion of the word.

I had checked out the laptop orchestra during my first week of grad school. The professor who led it that semester (not Dafna) set the mood by playing us HPSCHD by John Cage. I don’t know if you’ve had the pleasure, but it sounds to me more or less like seven harpsichords being thrown down a flight of stairs. I was so horrified that I fled the ensemble and never looked back.

I believe that the job of a musician is to include people and give them pleasure, not to shock them out of their bourgeois complacency. I recognize that there are people out there who get the same pleasure from John Cage that I get from Michael Jackson, but I have no idea why. How can Cage’s work be so revered by my fellow NYU professors, while causing me such distress? Music critics and scholars have expended a lot of mental effort trying to define what “objectively” good music is. I don’t believe that anyone will ever succeed. We can never form a complete consensus, and often we can’t even come close. On Fawlty Towers, Sybil accuses Basil of “skulking in there listening to that racket.” Basil replies, aghast, “Racket? That’s Brahms! Brahms’s Third Racket!”

So if quality is all subjective but you still want to write good music, what do you do? In art music, you can follow your idiosyncratic whims, or find a system and adhere to it, or imitate your teachers, or try to guess what will impress the tenure committee. In pop, you try to figure out what’s cool, or what’s going to appeal to the largest number of other people (not always the same thing). I align myself with Team Pop, because I want my music to be enjoyed by anyone—at least potentially.

Since I couldn’t use dance beats in my laptop orchestra piece, I needed to find some other way to organize it and make it listener accessible. Here’s what I arrived at. It’s called Divergence/Convergence. (Hip composers always use a slash in their titles.)

Each performer loads a short, shared sample. It should have a distinct attack and decay, for example a bell or gong. It can be pitched or unpitched, musical or unmusical.

Each performer triggers the sample repeatedly, either as a steady loop or at any arbitrary time interval.

After a few repetitions, each performer manipulates the sample as they see fit, via pitch shifting, time stretching, filtering, or other effects. Transformations should be gradual and clearly perceptible.

Once the entire ensemble is playing altered versions of the sample, the performers begin to undo their manipulations, preferably in the reverse order that they were originally applied.

When all performers have resumed playing back the original sample, the piece ends.

Hein's Divergence/Convergence

The shared-sample project is a mainstay of the Disquiet Junto, the internet’s most happening electronic music collective, for good reason: there’s no clearer way to see the creative thinking of a diverse group of musicians than to give them the same set of raw materials. My hope is that the shared-sample structure and the transition from uniform to disparate and back to uniform will be self-explanatory. I want the audience to be able to follow what’s happening in the music and why it’s happening, without the support of program notes or other inside knowledge. I want people to walk out of the performance saying, “I really liked that one piece where they were all playing the same bell sound.” Maybe they’ll walk out complaining about all the racket instead, but at least I can say I tried.

Mad Fresh

The most sampled recording in history is probably “Change Le Beat” by Beside and Fab Five Freddy, which was produced in 1982 by Bill Laswell.

If you’re listening to this song for the first time right now, you might be wondering what’s so special about it. You’d be right to wonder. Even Fab Five Freddy was reluctant to have his name on it. The special part of the track comes at the very end, when there’s a beep, followed by a vocoded voice saying, “Ahhh, this stuff is really fresssshhhh.” Hip-hop folklore has it that it’s Fab Five Freddy speaking the line, but it’s actually Bill Laswell’s manager Roger Trilling—he was playfully imitating Elektra Records head Bruce Lundvall. As Dave Tompkins puts it in his book How To Wreck A Nice Beach, “One of the most cloned hip-hop noises was but an imitation itself, mistaken for someone else in disguise, imitating the imitator on the A-side but replicated by a machine.”

The words “ahhh” and “fresh” are for hip-hop turntablists what the twelve-bar blues is for guitarists: both an entry point for beginners, and a bottomless resource for master practitioners. “Ahhh” and “fresh” are comprised of filtered white noise, which always scratches well. The “ahhh” has a distinctive attack and decay, so it’s easy for turntablists to keep track of where in the sample they are. And “fresh” is, well, fresh.

There are several different definitions of “fresh.” As a synonym for new or different, it can refer to food that isn’t canned, frozen, or otherwise preserved; a well-rested, energetic, healthy-looking person; an inexperienced noob; someone recently arrived, as in “fresh off the boat”; water that’s good to drink and not salty; or air that smells clean, pure, and cool. Fresh is also a dated slang term for impudence or impertinence. In hip-hop culture, fresh is on the endless string of synonyms for cool. Neither Roger Trilling nor the record executive he was mocking knew the hip-hop sense of the word when “Change Le Beat” was recorded, but that’s naturally the sense that turntablists intend when they scratch it. Thus you get Doug E Fresh rapping, “You’ve got to be [fresh], to rock with [fresh], and I’m D-O-U-G-I-E [fresh]!”

The wonderful thing about the hip-hop usage of fresh is that it could be referencing any of the various original senses of the word: new, refreshing, appetizing, attractive, or sassy.

fresh yogurt

We in Western culture have a habit of reflexively using “original” as a synonym for “good,” especially in music. I’m going to argue that originality is not actually a virtue, but rather, that freshness is. The concepts are related, but not identical.

In the strictest sense, we can understand originality to be a measure of information entropy, the information in a system that’s novel or unexpected. A song’s information entropy is high the first time you hear it, and then drops precipitously on each subsequent listen. Once you’ve thoroughly memorized and analyzed the song, its information entropy approaches zero. More generally, the music with the highest information entropy will be the most dissimilar to music you’ve heard before.

Producing original music in the information-theoretic sense of the word is trivially easy. Pull note names and durations out of a hat, or get a toddler to bang on a MIDI keyboard, or consult the I Ching. If you want to be really novel, you can generate audio files by randomly filling an array with ones and zeroes. The result is likely to be either tedious or annoying, or both. You’ve generated a lot of new information, but without a pattern or structure, it’s just noise. Now of course, some people like noise, and good for them. But even noise music is more structured than complete randomness. Most of us don’t want total originality in music; we want small variations and hybrids of known ideas, a delicate balance between novelty and familiarity. That balance will tilt one way or the other, depending on the listener.

Very often when we praise music for being “original,” we mean that it’s new or surprising to us personally. Surprise is entirely a function of your expectations. Recently, a student of mine presented a song by a self-described “experimental” Korean band called Clazziquai. I was expecting some kind of skronked-out punk, and instead was greeted by tame electronic pop with some occasional audio manipulations and stutters. Within the formulaic confines of K-Pop, no doubt these effects are startling. If you listen to a lot of Aphex Twin or Squarepusher, however, Clazziquai will hold no surprises for you.

Rather than evaluating music in terms of its originality, we need a criterion that gets at more meaningful aspects of musical quality: emotional truth-telling, recursive patterns of symmetries and asymmetries, intellectual depth, danceability, and so on. We should be judging music by its freshness. We can use exactly the same standards for music that we’d use for produce. A carrot doesn’t have to be unlike all other carrots that came before it; it just has to be crunchy, tasty, and nutritious. Unlike vegetables, music can retain its freshness over long time spans, and can even get fresher over time. Herbie Hancock’s “Rockit” epitomized freshness when it first came out—the album it appears on was named Future Shock for a reason.

This song exemplifies the ’80s so perfectly that it was inevitably going to become dated and lame after a while. But then, like many artifacts of early hip-hop culture, “Rockit” attained a retro freshness that will never wear out. The documentary Scratch features a series of turntablists who cite “Rockit” as their first inspiration. I incorporated some of them into my own remix:

“Change Le Beat” itself was never all that fresh, and it probably never will be. But just like a rotted log feeds a whole new miniature ecosystem, the “Ahh” and “Fresh” samples are inexhaustible sources of new music. No track that includes the samples can be original by definition, but they can most certainly be fresh.

Biting Breaks: Sampling and Ownership

sample break

This is my first post on NewMusicBox, and I’m delighted to be here. Over the next four weeks, I’m going to be looking at music composition through the lens of electronic music production, specifically the kind based on sampling. This music raises some tough and intriguing philosophical questions: Who is the composer of a sample-based track? Are tracks and notated works equivalent? Are producers “composers”? What even is a composer?

All of these questions are brought into stark relief by “They Reminisce Over You (T.R.O.Y.)” by Pete Rock and CL Smooth, a classic of ’90s hip-hop.

The track was inspired by the life and early death of Pete Rock’s cousin and friend “Trouble” T Roy of Heavy D & the Boyz. In a 2007 Village Voice interview, Pete Rock gave the backstory:

I had a friend of mine that passed away, and it was a shock to the community. I was kind of depressed when I made it. And to this day, I can’t believe I made it through, the way I was feeling. I guess it was for my boy. When I found the record by Tom Scott, basically I just heard something incredible that touched me and made me cry. It had such a beautiful bassline, and I started with that first. I found some other sounds and then heard some sax in there and used that. Next thing you know, I have a beautiful beat made. When I mixed the song down, I had Charlie Brown from Leaders of the New School in the session with me, and we all just started crying.

The Tom Scott record in question is his rendition of “Today” by Jefferson Airplane. The great sax riff comes at 1:37.

Here’s a transcription:

“They Reminisce Over You" sax riff

“They Reminisce Over You” sax riff

And here’s the original Jefferson Airplane song at the head of this memetic family tree:

The chain of musical inheritance doesn’t end with Pete Rock and CL Smooth. Their song has been sampled and quoted many times. Hear my mashup of some of them here:

I’ve debated the musical merits of sampling endlessly with my friends and students, musicians and non-musicians alike. “T.R.O.Y.” is a perfect example of why sampling is so valuable. There would have been no other way for Pete Rock to have arrived at his sound, not even if he had hired Tom Scott to come in and play his sax riff live in the studio. They could, in theory, have painstakingly recreated the instrumentation and ambiance from Scott’s original recording, but the result would still not have had the effortless, tossed-off feel of the samples. Playing a riff from a chart sounds very different from discovering it in the heat of the moment. Pete Rock’s looping transformed unprominent pieces of Tom Scott’s shaggy improvisation into laser-beam-focused funk.

In hip-hop terms, a “break” is a short segment of recorded music that can be sampled and looped. The term originally referred to drums and percussion, but it was later generalized to mean any kind of sound. In his book Making Beats: The Art of Sample-Based Hip-Hop, Joseph Schloss argues that Pete Rock created the Tom Scott sax break by sampling it:

On a conceptual level, this means that the break in the original jazz record was brought into existence retroactively by Pete Rock’s use of it. In other words, for the twenty-four years between its release and the day Pete Rock sampled it, the original song contained no break. From that day on, it contained the break from “They Reminisce Over You.” Producers deal with this apparent breaching of the time-space continuum with typically philosophical detachment. Conventionally, they take the position that the break had always been there, it just took a great producer to hear and exploit it. Record collecting is approached as if potential breaks have been unlooped and hidden randomly throughout the world’s music. It is the producer’s job to find them.

For a hip-hop fan, listening to ’60s and ’70s soul albums means regularly encountering familiar breaks. When I first heard “Are You My Woman (Tell Me So)” by the Chi-Lites, I immediately recognized the horns and drums from Beyoncé’s “Crazy In Love.” While I understand that, logically, the breaks in the Beyoncé song are really from the Chi-Lites, I still hear them as “belonging” to Beyoncé’s producer Rich Harrison.

Among sampling musicians, discovery has the same creative status as invention. DJs always want to play something that listeners don’t already know but that they will immediately like, and hip-hop producers have inherited this attitude. In a world saturated with recordings, creating more music ex nihilo is not the valuable service to humanity that it once was. I make sample-based music because I feel like it’s more worthwhile to identify existing sounds that have been overlooked, to bring them to fresh ears, and to give them fresh meaning in new contexts.

Theft is frowned upon in the hip-hop community, but the concept means something different from its traditional sense. If I were to use the Tom Scott break in a new track, without intending it as an homage or reference to Pete Rock, I would be “biting” his idea. However I would not be biting Tom Scott, or Jefferson Airplane for that matter. Copyright law disagrees on this matter completely, but sampling artists have never been overly concerned with copyright law, unless they’re forced to be.

Pete Rock’s moral ownership of the Tom Scott break is complicated by the fact that he wasn’t the first hip-hop producer to have noticed it. Slick Rick used it a year earlier on “It’s A Boy.”

Did Pete Rock bite Slick Rick? Is it a case of convergent discovery? Or is Pete Rock’s track just so much better that his ownership overrides Slick Rick’s? I don’t know the answer, but I suspect it’s the latter.

Even after 30-plus years of hip-hop, a lot of people continue to feel moral discomfort about sampling, especially when it happens without permission. Samplers themselves wryly acknowledge the moral ambiguities—see the Beastie Boys’ “Rhymin’ and Stealin’” or Ice Cube’s “Jackin’ for Beats.” Why does sampling need so much defending, when everyone long ago made peace with collage in other media? Maybe it’s because sampling amplifies the unreal qualities that all recorded music shares. Simon Reynolds observes:

Recording is pretty freaky, then, if you think about it. But sampling doubles its inherent supernaturalism. Woven out of looped moments that are like portals to far-flung times and places, the sample collage creates a musical event that never happened; a mixture of time-travel and séance.

Maybe our anxiety about sampling isn’t about ownership at all. Maybe we just don’t like being confronted so directly with the uncanniness of recorded music. While we might like to pretend that recordings are essentially documents of a performance that actually took place, sample-based music reminds us that this is totally untrue. Our discomfort with sampling is probably also the basis for the often-repeated statement that producers aren’t “real” musicians, that sampling is “just pushing buttons.” Having created music both with instruments and software, I can tell you that making good music is not any easier with the latter than with the former.

In the past, it made sense to conflate musicality with technique, because instruments are hard, and music is hard, and by the time you’ve learned to play, you’ve probably spent a ton of time learning the other. Music editing software is comparatively easy to learn, but you still have to master the music. Consider Microsoft Word: any reasonably bright person can quickly learn how it works, but learning how to write well is another ball of wax entirely. So it is with digital audio production. I can take any motivated student and have them chopping up samples in an hour. But are the results going to sound good? That’s where the musicianship comes in, and it takes as many dedicated hours of practice to attain it as with traditional instruments. Hip-hop has made any attentive listener into a potential composer. Now it’s up to us to use our ears.


Ethan Hein

Ethan Hein

Ethan Hein is an adjunct professor of music technology at NYU and Montclair State University. As a founding member of the NYU Music Experience Design Lab, he researches and designs beginner-accessible interfaces for music learning and creation. He maintains an active and widely followed music blog at