There’s no way to separate out the different tracks from a mixed song. If you want the vocals isolated, you need to get your hands on the acapella version of the song. DJ versions of pop and hip-hop singles often include the acapella and instrumental. Every so often a band will sell or give away “stems” — tracks with each instrument isolated. Stems also sometimes leak onto the web. Google is your friend here.

If you can’t get your hands on acapellas, instrumentals or stems, you can still do a lot of creative mashing up. Look for sections that are “in the clear,” where one instrument plays in isolation. Intros, endings and breakdown sections are good places to look for samples. I’m especially fond of laying a funky rhythm section break under a jazz or folk song, for example the breakdown of “1999″ by Prince under “A Hard Rain’s Gonna Fall” by Bob Dylan.

As for how to learn: just do it, do it, do it. The web is loaded with useful tutorials. Find some other DJs and remixers and ask for tips, or collaborate. The beauty of the digital music world is that busting out tracks and sharing them for comments and criticism costs nothing but your time. Don’t be too precious about your ideas. Get your tracks finished and play them for other people as often you can. After you do a couple dozen, they’ll start sounding musical, and after a hundred you’ll have some stuff that you’re proud of. Most importantly: have fun.

Here my recent remixes and mashups, enjoy:

Original post on Quora

Marcel Duchamp

In ancient times and more traditional societies, there was never much importance attached to the concept of sole authorship or ownership of creative works. Widespread belief in the lone Byronic genius didn’t take hold until the eighteenth century in Europe. Duchamp signaled the beginning of the end of the Byronic genius with his readymades, like the infamous urinal, or this bicycle wheel:

Duchamp also remixed the Mona Lisa by painting on a mustache and goatee and writing a dirty joke on the bottom.

I appreciate Duchamp’s art-historical significance, but find him to be too much of a wiseass for my tastes. I like my visual art a little more sincere.

Joseph Cornell

As a college student, I saw my first Cornell boxes, and was knocked out of my socks by them.

I didn’t know at the time why I liked Cornell’s boxes so much, but now I do. I come from a family of congenital pack rats, and Cornell shows a way to turn that neurosis into art. By combining stuff he found in dime stores, pictures cut out of magazines, and junk he found on the street, Cornell was effectively making three-dimensional mashups. Cornell himself was a weird and creepy guy, but, man, I love the work.

Cornell boxes have spawned a genre unto themselves. I see a lot of knockoffs out there, and I tend to like them better than most of what’s happening in the contemporary art scene. My sister did a whole series of them using fruit crates; here’s the one she made about my dad:

Tom Phillips and A Humument

My favorite contemporary visual remix work is Tom Phillips’ multi-decade project, A Humument. The work consists of drawings and paintings done on pages from A Human Document by W.H. Mallock, a Victorian novel that Phillips chose at random. Read the full story on Phillips’ web site. Here are some representative pages:

And here’s a side-by-side comparison of before and after:

Web memes

The internet loves visual remixes, and has unleashed wave after wave of web memes based on the idea. Maybe they’re a little less highbrow than the above examples, but they have their own charm.

We live in a world overstuffed with meaningful objects. Making new objects seems ridiculous in such a world. Better to recontextualize existing, familiar materials, giving them new meaning by combining them in unexpected ways. That’s exactly how I feel about music, too. Music is intrinsically a collage form anyway — all “original” music is assembled from pre-existing chords, scales, rhythms, melodic motifs and so on. I think our culture would be healthier if we could bring musical collage above ground and give it the same respect we give to Duchamp and Cornell.

DJ BC also did a remarkable album-length mashup of Jay-Z and Brian Eno:

Speaking of Jay, it’s a cliche by now, but Danger Mouse’s Grey Album really is a remarkable piece of work.

For creative use of pop, you’ve got to love DJ Earworm’s United State Of Pop series.

Diplo and M.I.A. deserve props for Piracy Funds Terrorism, especially their reworking of “Walk Like An Egyptian.”

The Walsh brothers did the best Wu-Tang mashup I’ve ever heard: Wugazi!

Finally, I don’t know if I’m one of the best mashup artists, but I’m pretty darn good.

Mashups by ethanhein

Additional recommendations are welcome in the comments.

Original post on Quora

Girl Talk uses comparatively primitive software for his mashups, so his process is labor-intensive. If you want to make life easier for yourself, get a hold of Ableton Live. Once you get a hang of the user interface, it’s a wonderfully versatile sampling and remixing tool. Before I got Ableton, I used a combination of various audio editors, Recycle and Reason, which worked well enough, but was a pain. Ableton is better.

Original question on Quora

While Girl Talk tracks have way too much information in them for my tastes, I could easily imagine having a rich musical life just unpacking their possibilities. Pop-oriented hip-hop over thrash metal! Gangsta rap over buttery piano ballads! Mixing prog and teeny bopper pop and classic rock! Teasing out the ideas suggested in these pairings could launch a thousand bands. In my own life as a musician, mashups have been the richest source of inspiration imaginable. Girl Talk lights the way with his fearless transgression of all boundaries of taste and style and copyright; it’s up to older and mellower musicians like me to pick up all the loose and tangled threads and knit them into something a little more coherent and structured.

Mashups by ethanhein

Why do some people like Girl Talk?

Wanna Be Startin’ Something megamix by ethanhein

Bitter Sweet Symphony Megamix by ethanhein

Human Nature Megamix by ethanhein

Prudence Never Can Say Goodbye by ethanhein

Na Na Na Na by ethanhein

Let’s say one of the umpteen copyright holders sampled by Girl Talk decides to go after him. Greg Gillis is white, nerdy and well-educated. He’s a generally sympathetic, press-friendly defendant with a rabid online following. If the copyright holder wins, the internet will rise up as one in protest, resulting in a public relations disaster that’ll cost way more than any amount of money that could conceivably be wrung out of Gillis. If the hypothetical copyright holder loses, it opens the floodgates to unlicensed sampling under Fair Use, which right now has tenuous status at best as a legal argument. Rights holders prefer the status quo, where the law is murky and everyone is so nervous that they mostly license their samples. Girl Talk might be popular, but until unlicensed mashups dominate the pop charts, the music industry will probably just keep ignoring them.

I should add that Schwimmer himself owns and loves all of Girl Talk’s albums. I think Gillis is pretty safe.

Original question on Quora

Sampling and influence

Whenever you look posts on the Musicians Wanted section of Craigslist by people who are starting bands, they all include a list of influences. They read like wish lists of samples. Whether you end up recreating a sound live or using a sample directly makes little difference in terms of the mental creative process. Every band I’ve ever been in yearned unconsciously for sampling. We’d try for the feeling of Stevie Wonder in Talking Book, or fifties Miles, or Led Zeppelin IV.

Shared musical memes are shared DNA

The tribal associations of music operate at a more granular level than entire genres or performers. Any shared musical memes build a network of musical association that can create pathways for emotional connection. Chord progressions, melodic figures, scales, rhythmic figures, lyrical phrases — all the DNA of music draws on a finite pool shared across the world’s musicians, the way that the genomes of humans and mice and fruit flies and daisies all draw on the same basic set of genes.

When John Lennon uses the sad descending chromatic bassline in “Dear Prudence,” he’s signaling an affinity for every piece of music that uses that bassline, and everyone who’s felt the mood that the bassline evokes.

Prudence Never Can Say Goodbye by ethanhein

I’m not a big Sarah McLachlan fan, but I do like her song “Ice Cream.” It has a nice 6/8 groove with a lot of syncopation, a groove I associate more with sixties Coltrane than with unthreatening singer-songwriters.

I finally looked “Ice Cream” up on the web and learned that the drummer on the session, Guy Nadon, is a jazz musician who studied with Elvin Jones.

By sneaking a little Coltrane DNA into the unlikely host of a Sarah McLachlan song, Guy Nadon was able to reach across my general hipsterish resistance and move me.

Shared DNA creates family

Michael Jackson had been on my mind quite a bit before he died, and hasn’t been far from my thoughts much since then. I’m especially interested in the “mama se mama sa mama coo sa” chant at the end of “Wanna Be Startin Something.” By quoting Manu Dibango, MJ was throwing a sly wink to all the disco and afro-funk lovers who were hip to “Soul Makossa.” Whenever someone references or samples the chant, it’s a signal of inclusion to those of us who care about MJ.

Referencing doesn’t have to be explicit or conscious for it to work. I loved “Got Your Money” by Ol’ Dirty Bastard and Kelis on the first hearing without knowing exactly why. Later I wasn’t too surprised to find out that the beat is a slowed-down sample of “Billie Jean.”

All music evolves from previous music. Sampling makes the chain of memetic inheritance more explicit than other musical memes.

Sampling is more emotionally evocative than quotation

Sampling is an more powerful tool for emotional connection than quotation, because in addition to the melodic or rhythmic figure that’s being activated in your memory, it’s all the subtle nuances of a recording that you may have heard hundreds or thousands of times. Samples can short-circuit the analytic parts of your memory and tap directly into the deep unconscious.

Permission and ownership

The simultaneous beauty and menace of sampling is that you don’t need anyone’s permission: not the performers, not the producers, not the composers or arrangers or copyright holders. Selling your sampled works might be another ball of wax, but if you just want to make mashups, all you need is the audio and a few pieces of inexpensive software.

Be brave, let people sample you

The recording artists I admire are the ones who invite new interpretations of their work. Jay-Z puts out remix-friendly versions of his albums with the isolated vocals on one side and the instrumentals on the other, with the express purpose of making it easy for anyone to repurpose them. The electronic music world has responded enthusiastically, so now you can hear Jay’s rhymes paired with the Beatles, Radiohead, Brian Eno and other unlikely-seeming musical combinations. When I was more of a hip-hop dilettante, Jay’s music was a little too intense for me. Hearing him combined with the safely familiar White Album was the gateway for me to be able to appreciate his work in its original setting. Being able to connect to Jay opens the possibility of connecting to his many fans, which has broadened my social circle noticeably.

Not everybody likes the connections formed by sampling

A lot of people are angry about the practice of sampling, and I’m not just talking about copyright holders. Plenty of people I know find sampling enraging, from professional musicians to the most casual listeners and everyone in between. Maybe these people are attached to their emotional associations to particular recordings and don’t want them invaded. I can see that. When I hear a song based on a sample before the original, I can’t help but think of the sampling track when I eventually do hear the original.

I heard “Crazy In Love” by Beyoncé dozens of times before I ever heard “Are You My Woman (Tell Me So)” by the Chi-Lites, the source of the brass and cymbal samples. As a result, the Chi-Lites inevitably evoke Beyoncé for me. I think this is basically a good thing. You can’t keep a fence around your emotions, no matter how much you might want to. More complex associations force the possibility of more connection with other people, and you can never have too much connection.

The discussion continues

A couple of my musician friends shared some thoughts about this post on Facebook. Jeremy is a rock bassist, and Jesse is a jazz trumpet player.

Jeremy:

Nicely done! One of the best defenses / explanations of sampling in music that I’ve read. One thing worth exploring / defusing when it comes to the people who hate sampling is “the craftmanship argument.” To wit: people who say that sampling is not valid because the sampler did not play or write the sampled element. I don’t think that this is a valid argument against the practice of sampling, but haven’t exactly put together why…

Me:

Thanks dude! Your opinion matters to me. I think the craftsmanship issue is a critical one, and yeah, it’s tricky. There’s the puritanical equation of effort with quality, for one thing. For another, there’s the idea that learning an instrument is more effortful than copping a sample. On the first front, well, that’s such a deep-seated cultural assumption that people tend to not be open to debating it. I’ve spent years struggling against it. Sampling seems “easier” because the work is kind of invisible compared to woodshedding on an instrument. But the work is still real. You need to listen to a lot of music and do a lot of analyzing of it before you can identify good samples. There’s not much effort in the act of pulling and looping them, but there’s a lot of intellectual groundwork to be laid, and in the end I think it ends up being as much effort as running scales on guitar or whatever.

Jesse:

Going to weigh in here with my three cents 1) The structural information of the song: lyrics, melody, harmony, even certain iconic rhythms can be quoted and referred to by a band or a composer either indirectly as in a quotation or iteration or directly by doing a cover tune, let’s say. 2) This is a paraphrase from “This is your brain on music” – A study was done where a 2 second clip of pop tunes was played. Just from the “sound” (the sonic environment – the production elements: reverb, orchestration, warmth) participants could reliably pick the tune out of a multiple choice list. I experimented with this by the way. It’s fun. Drop your iTunes needle around. Anyway, I can spot any track of “innervisions” in under a second. 3) That feel of that sound is what sampling is doing as well. And, sorry, wether its artfully done or not, is quite different than the content of the song: its the context of the sound of the song that produces the resonance.

Me:

I guess it depends on the length of the sample we’re talking about. A one-beat stab buried in the mix is going to work differently on the listener than a complete phrase. It’s the difference between the kick drum from the Funky Drummer vs the entire four-bar loop. My post was more about the use of full-length phrases, since that’s what tends to be the most emotionally (and legally) controversial.

Jesse:

I guess my point was that actually copping the sound of a recording live or on another recording is just about impossible without sampling. There’s something that can be artful about it but ultimately there is an appropriation of some producers hard, hard work to create an iconic sound. To equate the effort of sifting through a library with taste and creativity with the effort of actually producing those iconic sounds from scratch is absolutely offensive. It’s just a straight up different art form and without accurate sourcing is enslaving in that it makes something work outside its consent and original purpose.

Oh and . . . the bomb squad was great at it Irony. So much fun.

Me:

I agree with you that the whole point of sampling is to use a recognizable recording vibe along with the recognizable melody, rhythm etc. I don’t find it offensive. You could just as easily think of it as a tribute, homage, humbling yourself before your source inspiration. That’s how I think of it. I do believe in accurate sourcing. The present brokenness of the copyright system drives samplers underground and encourages secrecy about sources. I think there’s another unspoken philosophical tension here about who owns a piece of art, the artist or the world. I side with the world. I feel like, I bought the record, I own it now.

Jesse:

I’m on both sides of it. It’s a good debate. Warhol comes to mind obviously.

Jeremy:

There’s also obvious parallels in sampling to collage art in the visual world, and found art. When you take a sample out of the original context, juxtapose it with new samples or original music, you create a new, hybrid context.

You can have bad sampling, just as you can have bad original composition. When someone hijacks an entire song and just redubs their vocals, well, that sets off a lot of alarms… I think there needs to be a new critical language to address what constitutes “artful” sampling vs. “artless” sampling.

Jesse:

This is an interview that shares Pat Matheny’s views on Kenny G. Vis artless sampling. Priceless. Good for the soul. Devastatingly brutal.

A snippet: “Not long ago, Kenny G put out a recording where he overdubbed himself on top of a 30+ year old Louis Armstrong record, the track “What a Wonderful World”. With this single move, Kenny G became one of the few people on earth I can say that I really can’t use at all – as a man, for his incredible arrogance to even consider such a thing, and as a musician, for presuming to share the stage with the single most important figure in our music.”

Jeremy:

I had read this before – a friend, knowing my love for incendiary music writing, forwarded it to me. It’s great. “But when Kenny G decided that it was appropriate for him to defile the music of the man who is probably the greatest jazz musician that has ever lived by spewing his lame-ass, jive, pseudo bluesy, out-of-tune, noodling, wimped out, f*cked up playing all over one of the great Louis’s tracks (even one of his lesser ones), he did something that I would not have imagined possible…”

So I think we have a good working definition of what qualifies “bad” sampling.

Me:

Yeah. I’m not completely opposed to the idea of overdubbing your own sax solo on another recording, but ideally you’d be someone less lame.

Here’s my story of good sampling. We did a Revival Revival show and while Barbara took a break, I spun some of my instrumentals. On the Bittersweet Melody remix, I have a sample of the opening few bars of A Love Supreme. Two tenor sax players were at the gig, so after the sample had run a few times they both jumped in on it. They played it in more or less in unison a few times and then took off on this whole interlocking thing with the phrase transposed and displaced against the original. It was pretty magical, I wish we’d recorded it.

Guys, do you mind if I include some excerpts of this conversation in the blog post itself? I think non-FB users would find it interesting. If you’d prefer I didn’t that’s cool too.

Jeremy:

I’ve got no problem with that… it seems especially appropriate to throw some samples into the writing, if you don’t mind getting all “meta” with it. Personally, I have no problems with meta-whateva.

Oh, and I wish you had recorded that Revival Revival gig too! In my head it sounds awesome! Did you know the tenor players were going to do that?

Me:

No idea! One of them had been playing bass and a little sax with us regularly and the other was just there hanging out, I had never met him before. It was both of their first time hearing the track.

Jeremy:

Good frickin’ lord… real musicians just blow me away.

Me:

You’re telling me.

Jeremy:

Yeah, well, that’s what I’ve got effect pedals for… when you can’t wow ‘em with chops, confuse ‘em with expression-controlled delay trails.

The tune has no chord changes, just a straight-ahead funk groove in B flat. The angular melody is mostly fourths jumping up and down by whole steps or half steps. It’s tough to memorize, but then once you get it down, it can stay stuck in your head for weeks.

As befits such an eccentric yet popular song, Eddie Harris was an eccentric yet popular guy.

Harris had the first gold jazz record, an arrangement of the theme from the movie Exodus. He also had a hit with Swiss Movement, led by Les McCann. Harris invented a variety of odd instruments, and in 1975 he even released a standup comedy album, The Reason Why I’m Talkin’ S**t. He also subconsciously impacted my entire generation by writing a lot of the music on the Cosby Show.

“Freedom Jazz Dance” has been performed and recorded many times, but Miles Davis basically owns the tune after the rendition he did on Miles Smiles.

Here’s a chart of Miles’ solo; click to see it bigger. Check out how much space he leaves between phrases.

Miles’ version has been sampled by a few hip-hop artists. You can hear fragments of the solos throughout “I Thought Ya Knew” by Ed OG and Da Bulldogs.

Miles’ recording was also sampled by Fu-Schnickens in their Bugs-Bunny-referencing track “Creepin’ Up On Ya.”

Miles’ estate has been fumbling with attempts to keep the man relevant. So they released a strange remix EP that includes a frenetic couple of minutes of snippets of “Freedom Jazz Dance” along with Miles’ verbal instructions to the band from the studio outtakes, and then a quick, slightly awkward rap by Nas. It’s not a totally successful experiment, but it’s pointed in the right direction. It’s especially amazing to hear Miles sing parts of the melody.

Here’s a mashup of all the tracks mentioned above.

Freedom Jazz Dance megamix by ethanhein

Any recommendations for more recordings, samples or references to “Freedom Jazz Dance” that I should know about? Hit me in the comments.

I was reminded of it walking down the street the other day, because someone in our neighborhood in Brooklyn was blasting a dancehall track from their car that sampled the “na, na na na na, na na na naaah na na na na na na” part from “Land Of A Thousand Dances.” Then I got to thinking, this cartoon is actually an inspired recipe for a mashup. So here we go:

Na Na Na Na by ethanhein

Production

I put the track together using my new favorite software in the world, Ableton Live. I’ve historically been a Reason/Recycle guy, and while I’ve had Ableton sitting on my hard drive forever, it took me until recently to motivate to check it out. I’m glad I did, it’s been wildly inspirational. Here’s how the mashup looks in Ableton’s session view.

Land Of A Thousand Dances

I used the Wilson Pickett version of “Land Of A Thousand Dances.” Here’s a live version — the hook comes at 0:46.

Hey Hey, Good Bye

The chorus of “Na Na Hey Hey Kiss Him Goodbye” is such a ubiquitous meme that it’s strange to think it was originally part of an actual song, rather than something written specifically to be chanted at sports games. It’s by an anonymous bunch of studio musicians who called themselves The Steam. Here they are in Austin Powers finery:

I also included the version by Bananarama, because who doesn’t love Bananarama?

Nas Is Like

The one song I put in that isn’t from the cartoon is “Nas Is Like” by Nas. I used it because the repeated “Nas Nas Nas is like, Nas is like” fits the sonic theme well, and because I needed a good strong beat for certain sections of my track.

Batman

The cartoon references Neal Hefti’s groovy theme song from the sixties TV Batman show. The thing is that there is no “na na na” part anywhere, it’s just the way everyone sings the guitar riff. I figured I’d just put it in rather than split hairs.

Katamari

The Japanese a capella voice with the bit of beatboxing at the end is “Sasasan Katamari” by Yu Miyake, from the classic video game Katamari Damacy.

Randall Munroe shares my love of this game and its infectious theme music:

Hey Jude

Finally, I conclude my mashup with the entire end section of The Beatles’ “Hey Jude,” a masterpiece of zen-like repetition.

Other na na songs

Songs I could have included, but didn’t:

• My Chemical Romance – “Na Na Na (Na Na Na Na Na Na Na Na Na)”
• Akon – “Right Now (Na Na Na)”
• 112 – “Na Na Na Na”
• Jim Jones – “Na Na Nana Na Na”
• Tiffany – “Na Na Na”
• A.B. Quintanilla III Y Los Kumbia Kings – “Na Na Na”

Why are all these “na na na” songs so catchy?

I think nonsense syllables are crucial to good pop music. They enable emotional expression uncluttered by narrowly literal semantic meaning. “Na na na” can mean anything, everything or nothing. It’s a blank space to project whatever feeling you want onto it. I subscribe to the belief that music is much older than language, and that it reaches much deeper into the core of the brain. Like scat singing and obligato, singing “na na na” touches something much bigger and more ancient inside us than any specific set of lyrics can. Maybe that’s why these hooks are so much more memorable than the songs they come from. Anyway, enjoy the mashup, and thanks Randall Munroe for the best comic on the internet.