In comparison to the age of the universe (or of human civilization), it wasn’t that long ago that recording music was a “perfect take” affair: set up the microphone, record to the masterdisk, and hope for the perfect take.
Of course, things have changed technology-wise, and so had the approach to recording, editing and mixing music, with all the steps possible in today’s DAWs in between the sound coming from the performer and the sound going out to the audience listening to the recorded release.
And with those advances in technology and possibilities, the opinions on “what is right” for a specific genre and recording situation has changed. For studio recordings of pop/rock music, lots of multi-tracking, overdubs, alternate takes and sample-accurate editing are the norm, and are well-accepted by the audience of those genres. The same is not necessarily true for Classical or Jazz music on one hand, and for live recordings on the other. And finally, improvised music plays a wholly different role here (especially if it’s from that more radical approach where the improvisation does happen without any (conscious) preconceptions).
The public acceptance for those production techniques also has grown, so it seems, over the years. Even twenty years ago, it was already well-accepted that a Classical symphony (say, Bruckner’s Eight) was recorded in front of a live audience. Then producer and conductor got together and found the “not-so-good” parts, to return to the same venue (only without audience) to go into overdubbing, sometimes for several days. During that time, also the identification of a recording as a “live recording” vanished, and the mode d’empoli seems to be well-accepted by the most narrow-minded audience of all, the Classical traditionalists (althought I’m sure a lot of them don’t even know about this).
For studio work in the earlier days of Classical music, it was mostly due to the most tech-savy (and at the same time, chops-ladden) artists such as Glenn Gould, who was once quoted as saying to an enervated producer that he needed at least one more day in the studio to overdub one or two trills…go figure.
For rock music (in the widest sense), the modern production techniques were adopted much earlier, if only for studio albums. Incidentally, one of the most forward-thinking artists here was someone at home in the more radical/improvised genres: Frank Zappa started in the 70s to take recordings of (not only) guitar solos from live performances and reassemble them with studio tracks from the rhythm section, other tracks, or even compose songs around them. This approach was dubbed the Ampex guitar by Dick Kunc, his then engineer. Prime examples for this technique are contained on Sheik Yerbouti and Joe’s Garage, as well as on various tracks on Zappa’s guitar solo compilations. But this had been done earlier on at least a few occasions as part of Miles Davis‘ work with Teo Marcero, here A Tribute To Jack Johnson being a good example.
On the other hand, there’s musicians and theorists who believe that improvised music must not be recorded to begin with. And that a live album must only be presented unedited and without overdubs. So what is the twenty-first-century truth on that?
Why even think about this?
For me, as an artist and as a producer, this has of course a lot of relevance, especially when we’re talking about live improvisations – with the definition of “live” ranging from the live audience over the virtual audience back to the improvisation alone in the studio but with an “as if” mindset.
In fact, everything I have released under my name (and also, with the exception of maybe a dozen or so tunes, performed) ever since my 2000 groxis tafelbilder album up to 2012 was radically and freely improvised. This includes my Eclectic Blah project back in the day, and this is also the reason why I bring this up right now, as important production choices were and are made based on my current stream of consciousness on the topic.
So Eclectic Blah is freely improvised, and is live, so it has to live up to the toughest standard in non-use of contemporary editing techniques – or does it?
Is it still improvised if you record it?
The answer to me is yes – as simple as that. And I know that I have a few well-respected artists on my side on this, including (but not limited to) John Zorn, Derek Bailey and the aforementioned Zappa.
Is it acceptable to overdub/edit on a record called “live”?
This one is more tricky. Does the audience expect this to be a 1:1 recreation of the actual performance. Even if that one was flawed in one of many possible ways? This has, of course, a lot to do with how you perceive yourself as an artist, how you define your product, and consequently, what the customer expectations are. Before we go into the details of what is possible (and what the impact is), I generally like the approach of some bands who might release “officialized bootlegs” (such as King Crimson‘s Thrakattak or, once again, Zappa‘s Beat the Boots) with that unchanged and unbeautified experience, and on the other hand, professionally recorded and produced live albums.
But what can we do with regard to editing on a live album? Ordered in ascending order of radicalism (to a degree)…
Basic mixing techniques are, so I think, accepted by about anyone. Adjust levels, adjust EQs, mute channels etc. – after all, you’re only making fine adjustments to the sonic picture for higher listening pleasure but don’t change the musical message, right? It gets more interesting once you use creative effects: putting a filter delay on a (otherwise boring) snare drum and activating it every fourth beat can change an uninspired drum part into a really capturing experience. And even adjusting the EQ on the bassdrum can turn a long-haired rock drummer into an Eighties’ mainstream pop act.
At the “basic” level here, I’d like to quote techniques that are related to what, in video editing, is called linear editing, although the terminology does not 100% match. Essentially, you can cut out a passage of the entire piece (like another boring chorus in a solo break), but you can’t wildly rearrange order on a per-track basis with the multitrack stems. If we do that, to e.g. take a great drum fill from that boring chorus and paste it into the head out, we’re in non-linear editing territory. And finally, we also can take parts from entirely different performances (see the Ampex guitar reference before), and by that time we reach inter-song editing.
As with the mixing techniques before, acceptance of those techniques drops the more wicked they get. Interestingly, while an edit is by definition a more surgical approach than a simple EQ adjustments, the effects are often much less striking from a sonic perspective.
With Overdubs, we’re finally deeply in no-go territory: having a musician sit in after the recorded performance (maybe even someone who wasn’t at the concert to begin with) can’t be “live” by any standard – and this is independent of whether the overdub is “corrective” (i.e. fix a few wrong notes), “enhancing” (doubling parts, or again do The Zappa and replace a distorted bass drum channel with a sampled bass drum, as on the YCDTOSA2 album), or is a completely new part.
Putting it together
The above diagram tries to summarize the results from the last paragraph: what is possible, how it affects the sonic picture, and how does it affect the artistic integrity. As you can see, while basic mixing techniques are considered quite integer (close to the bottom), they can affect the sonic picture a lot, if used properly. On the other hand, while a corrective overdub has but a small effect (unless your band is crap), the artistic integrity is hugely compromised by public opinion. And finally, you can do quite a lot with mixing alone, or with mixing plus nonlinear editing.
In the above article, we’ve found what some artists did in the past, what is possible, and what those possibilities result in. On the other hand, we’ve definitely not found out what is “ok” for our specific application, piece of art and business case (whatever that is or means).
In my case, it means to a large degree to make trades, especially when we’re moving upwards in the diagram (e.g. everything at or above the corrective overdub). To give two examples: the bass drum from the The Ebersberg Enigma concert was so unuseable audio-quality-wise I decided to extract the timing from the audio track, sequence the dynamics by hand and then use that to drive a pristine drum sample, because the big impact on the sonic picture (in an all-positive way) justified that. For a slightly unbalanced track containing a Wavestation and an ample multitap from the TC D2, I am currently undecided if I should completely rerecord that (but will most probably decide against it, because the sonic impact is rather low); same goes for replacing the artifical-sounding Proteus percussion samples on Dreams of Hesse with some from the Battery library.
By no means can this article aid you to make the decisions you need to make – as an artist, as a producer, or as an artist/producer. If anything, this article might help you to make more use of the wonders of standard mixing techniques…