Mix methodology
Recently I have re-acquired an old habit: the assembly of songs to form personalized albums and sequences. Back in the day -- this was about 1985 -- such an endeavor involved sitting on the floor in front of my stereo surrounded by piles of cassette tapes (we didn't have CDs back then, kids) with a piece of paper and a stopwatch, trying desperately to figure out how to squeeze 0:15 more from the blank tape in deck 2 because the last song was running a little longer than I had estimated before beginning to dub, and if it didn't fit the WHOLE DAMN SIDE OF THE TAPE was ruined and had to be redone…
These days we have better technology for such things:
iTunes. The handy "total time" indicator at the bottom of the playlist screen ensures that things will all fit (a single 700 mb CDR can hold 1.3 hours of music under normal circumstances); SoundCheck will harmonize the levels of the tracks so that they are not wildly different from one another in loudness (but this comes at a cost, in that quiet tracks sometimes get distorted; manual adjustments of the volume level work better, but take more time); and, perhaps best of all, it's all real-time and non-linear so you can put a group of songs together, listen to it, see if it "works," and then make on-the-fly adjustments. Almost nothing quite as depressing as recording a mix tape, listening to it, and then deciding that song 2 and song 7 need to be swapped, and song 5 doesn't really belong on the tape at all -- but iTunes makes this a very simple process.
Yes, I am aware that there are other programs out there that do all of these things. But iTunes is, IMHO, the gold standard.
There are several types of personalized albums or song sequences (I'm grasping for a neutral term to describe a whole class of objects; "mix" doesn't work for reasons that will become apparent in a moment). The simplest is a "playlist," which is just a collection of stuff selected pretty much at random. iTunes allows you to create "smart playlists" that automagically update as you add to and subtract songs from your library; I have one called "(live)" that collects every song that I label as a live track by appending "(live)" to the end of the title of the song when I first add it to my collection. Indeed, you can produce playlists that incorporate a variety of criteria by tweaking the selection principles a bit: a little alt.country, some 80s pop, one song with a length shorter than 20 minutes from one of
Spock's Beard's studio albums other than
Snow, and something off of the "not recently played" playlist -- i.e. something I haven't listened to in a while. And keep the total length under two hours.
(20 points for your House if you can see where I'm going with this yet.)
Now, a playlist is just a collection of songs selected more or less at random, or at any rate selected for "external" reasons -- not because the songs go together in any more profound sense. If attention is paid to how the songs cohere, we move from a playlist to a "mix," regardless of the criteria used to evaluate the coherence of those songs. It could be how the whole thing sounds; it could be the theme that the songs are all playing with; it could be whether the songs produce a good sampler that nonetheless sounds good on its own. The trick here is that a mix is an
entity in its own right in a way that a playlist isn't yet.
Mixes can start out life as playlists. I generally produce playlists to run to, and then modify them after a run to get them to flow better. The playlists are on their way to becoming mixes. Also, a good DJ strives to produce something other than a mere playlist when she or he is spinning.
(15 points for your House if you get the point yet. Look at the blog entry title again.)
Mixes cohere to some extent, but some mixes are more coherent than others. At a certain point we pass beyond a series of songs that simply work together in an indefinite sense (generally sonically, I think: mixes have to
sound good, but nothing more) and enter a realm where the songs in the mix start to work off of one another, echoing themes and concepts and compositional styles, complementing one another and producing, ever so subtly, something emergent and novel. Call these mixes "blends." Like a good mixed drink or a fine meal, the variants among the component pieces are harnessed so as to produce an effect that the individual pieces wouldn't exercise individually. A song evokes a certain sentiment or implies a particular train of thought; when paired with another one its flavor alters in minor but important ways. In a way, a blend is like a fine wine or a good shot of
Becherovka: complex, stretched out over a temporal axis such that it has to
unfold rather than simply being quaffed (and if you do simply quaff it, it finishes out in unexpected ways -- part of the fun of it).
I would hazard a guess that a musical blend needs liner notes of some kind, since "coherence" is not an intrinsic property of the songs themselves. Listeners need to be shown how the songs cohere, so that they can (metaphorically, if not literally) borrow someone else's ears to listen to the sequence. Not that the coherence of a blend is somehow traceable to the subjective motivations of an author; once a blend has been produced, coherence can be discovered by listeners in all sorts of surprising ways. But if I am making a blend for someone I generally want to communicate a point, and to aid in that endeavor -- and because of the intensely personal character of the meaning that people attribute to songs, which makes it more than likely that they will go spinning off in their own direction when I accidently put a song in the blend that reminds them of their first love and the fiery break-up of that relationship, or something like that -- I generally find liner notes useful. These can be written, or you can simply listen to the blend with the intended recipient, explaining as you go, and perhaps discussing the themes expressed or whatever is raised by the communal listening experience.
(10 points for your House if you get it now. What goes on when two or more people listen to a blend and wrestle with it?)
And then there are the superior blends, the ones that work so well that one can no longer listen to the individual songs in the blend again without thinking of the blend and its specific and unique favor. The song has been constitutively altered, and no longer stands alone in the same respect at all. Call these blends "alloys." I'm thinking of
white gold here: something that is immeasurably more precious than the sum of its parts, but that remains in a fundamental sense a union of initially diverse things. Alloys aren't "naturally occurring," which doesn't mean that they aren't
natural, but does mean that they require deliberate social effort to (re)produce. I prefer to think of alloys as
fictional rather than as "artificial," since the former term means "something made" but doesn't need to carry the strong normative connotation of the latter term. (And Durkheim's opposition of organic and mechanical solidarity doesn't work for me for the same reason; both
Gemeinschaft and
Gesellschaft are equally "fictional" in this sense. Or, as Clifford Geertz put it, the real is just as imagined as the imaginary. But I digress -- or do I?)
A good concept album is an alloy. I mean, really, the songs on Pink Floyd's
The Dark Side of the Moon are good individually, but when you put them all together, how much better do they sound? Ditto for Genesis'
The Lamb Lies Down on Broadway or Spock's Beard's
Snow. Each piece implies the whole of which it is a part, even if you listen to it in isolation (which is why the running playlist that I mentioned above excludes
Snow). And if you've only heard "Time" or "Money" in isolation, and then first hear them in their correct album sequence, and understand how they proceed from "Speak to Me / Breathe" and lead up to "Any Colour You Like / Brain Damage," the whole thing changes, the whole
world of which these songs are a part changes. (If you don't know
The Dark Side of the Moon, all I can say is, you should; it's one of the best albums ever recorded by anyone, I'd argue.)
Indeed, a good alloy works at the level of world-constitution, changing the sense of the whole (like a Heideggerian "mood," or Wittgenstein's comment that the world of the happy man is different than the world of a sad man, even if all of the facts remain unaltered) in ways that cannot be captured by rational language. Individual songs do this too -- that's part of the mysterious appeal of music -- but a good alloy does it even better, in a way that goes beyond the haphazard lurching characteristic of, say, unformatted radio.
(5 points for your House if you get it now. Hang on, the punchline's coming.)
So this morning it occurred to me that the process of assembling sequences of songs is very much like the process of engaging in methodological self-reflection. In both cases, one uses bits and pieces that are floating around in the social environment and puts them together so as to open the world in a specific way. And one can do this in a variety of ways, depending on the kind of effort and forethought that one puts into the process. We all know of works in our various disciplines that have playlist-methodologies, especially those whose coherence is only explicable in the following way: "well, persons X, Y, and Z were on her or his committee, so this is there for X, and this is there for Y…" Result: incoherent, incomprehensible mess. Mixes also exist, in which the author/researcher makes at least
some effort to fit the pieces together to achieve a pleasing effect as opposed to basic cacophonic discord. "Mixed methodology" work in the social sciences reads like this, although the combinations are often strange and jarring -- and generally betray a dominant theme with various subordinate helper-techniques. So it's a mix, yes, but that's all.
Better, but rarer, are blends and alloys. Methodological alloys are what we should be aiming for in our theory chapters and sections; if we fall short we still end up with blends, which are a damn sight better than mere mixes and playlists. Alloys and blends do take longer, more nuanced and subtle discussions. but they are worth it in terms of the overall effect (aesthetic, intellectual, moral -- all the same thing, in the end, aren't they?) that they produce. Which would you rather listen to? What kind of world would you rather live in?
Good research is implementing a set of methodological principles. Producing those principles -- forging the tools -- is all about creating a coherent blend or, at best, a transformative alloy. And that's what science as a vocation is all about.
[Posted with ecto]
