02. Part II 18:12
03. Part III 11:13
04. Part IV 9:47
Drums [Tuned], Marimba, Glockenspiel – Bob Harms, Gary Kvistad, Gary Schall, Glen Velez, James Preiss, Russ Hartenberger, Steve Reich, Thad Wheeler
Piccolo Flute – Mort Silver
Voice – Jay Clayton, Pamela Wood Ambush
Recorded May 1987 at RCA Studio A, New York City.
For one year, between the fall of 1970 and the fall of 1971, I worked on what turned out to be the longest piece I have ever composed. Drumming lasts from 55 to 75 minutes (depending on the number of repeats played) and is divided into four parts that are performed without pause. The first part is for four parts that are performed without pause. The first part is for four pairs of tuned bongo drums, stand-mounted and played with sticks; the second, for three marimbas played by nine players together with two women’s voices; the third, or three glockenspiels played by four players together with whistling and piccolo; and the fourth section is for all these instruments and voices combined.
While first player the drums during the process of composition, I found myself sometimes singing with them, using my voice to imitate the sounds they made. I began to understand that this might also be possible with the marimbas and glockenspiels as well. Thus the basic assumption about the voices in Drumming was that they would not sing words, but would precisely imitate the sound of the instruments. The women’s voices sing patterns resulting from the combination of two or more marimbas playing the identical repeating pattern one of more quarter notes out of phase with each other. By exactly imitating the sound of the instruments, and by gradually fading the patterns in and out, the singers cause them to slowly rise to the surface of the music and then fade back into it, allowing the listener to hear these patterns, along with many others, actually sounding in the instruments. For the marimbas, the female voice was needed, using consonants like "b" and "d" with a more or less "u" (as in "you") vowel sound. In the case of the glockenspiels, the extremely high range of the instrument precluded any use of the voice and necessitated whistling. Even this form of vocal production proved impossible when the instrument was played in its higher ranges, and this created the need for a more sophisticated form of whistle: the piccolo. In the last section of the piece these techniques are combined simultaneously with each imitating its particular instrument.
The sections are joined together by the new instruments doubling the exact pattern of the instruments already playing. At the end of the drum section three drummers play the same pattern two quarter notes out of phase with each other. Three marimba players enter softly with the same pattern also played two quarter notes out of phase. The drummers gradually fade out so that the same rhythm and pitches are maintained with a gradual change of timbre. At the end of the marimba section, three marimbas played in their highest range are doubled by three glockenspiels in their lowest range so that the process of maintaining rhythm and pitch while gradually changing timbre is repeated. The sections are not set off from each other by changes in key, the traditional means of gaining extended length in Western music. Drumming shows that it is possible to keep going in the same key for quite a while if there are instead considerable rhythmic developments together with occasional, but complete, changes of timbre to supply variety.
I am often asked what influence my visit of Africa in summer of 1970 had on Drumming. The answer is confirmation. It confirmed my intuition that acoustic instruments could be used to produce music that was genuinely richer in sound than that produced with electronic instruments, as well as confirming my natural inclination towards percussion (I became a drummer at the age of 14).
The transition from glockenspiels to the last section of the piece, for all instruments and voices combined, is made by a new musical process I call build-up and reduction. Drumming begins with two drummers building up the basic rhythmic pattern of the entire piece from a single drum beat, played in a cycle of twelve beats with rests on all the other beats. Gradually additional drumbeats are substituted for the rests, one at a time, until the pattern is completed. The reduction process is simply the reverse where rests are gradually substituted for the beats, one at a time, until only a section leads to a build-up for the drums, marimbas, and glockenspiels simultaneously.
There is, then, only one basic rhythmic pattern for all of Drumming. This pattern undergoes changes of phase position, pitch, and timbre, but all the performers play this pattern, or some part of it, throughout the entire piece.
— Steve Reich
Reich made the pioneering recording of "Drumming" back in 1974, for DG: not only did he supervise the recording, he also played the marimba in Parts II and IV, and whistled in Part III. One of the drawbacks of that recording was that it lasted circa 85 minutes - Drumming, completed in 1971, was back then and remains the longest piece ever composed by Reich: instead of being played continuously (as it was conceived), each part had to be broken down to one LP side, and even when it was reissued on CD (in 1989) on DG's 20th-Century collection (Reich: Drumming; Six Pianos; Music for Mallet Instruments, Voices and Organ, later reissued in the 20/21 collection, still on two CDs but with no complement, Drumming), Part IV was rejected on CD II. In the latter case, the damage wasn't so bad, since Part III in fact tapers off to a single repeated note on the glockenspiel, then picked up by other instruments in Part IV, but still, DG's decision to fade out then fade back in wasn't the most clever one: it might have been better to simply cut off on the last glockenspiel note and start with the next one.
So the interest of this 1987 recording on Elektra/Nonesuch, whose duration is 56:40 minutes, is to enable you to hear the piece continuous. There is a certain degree of freedom afforded to the performers in playing "Drumming", based on how many repeats are made. Funny that, in the liner notes to this Elektra/Nonesuch release, Reich should comment that "Drumming lasts from 55 to 75 minutes". Hey man, haven't you counted how long your own 1974 recording lasted???? At 75-minutes, it would have fitted on a single CD no problem, and there would have been no need for this remake.
Another nice feature of the re-recording is that it involves many of the original performers. But after all, it was hardly more than a decade later. Stupidly, the booklet's credits list them in alphabetical order, so you don't know who's playing what. In Drumming, Reich (to quote his own presentation in the booklet of the DG release) brought to the "final refinement [his] phrasing technique in which two or three identical instruments playing the same repeating melodic pattern gradually move out of synchronization with each other, but the work also introduce[d] several new techniques: (1) gradual changes of timbre while pitch and rhythm remain constant, (2) the use of the human voice in an instrumental ensemble imitating the exact sound of the instruments, (3) the process of gradually substituting beats for rests (or rests for beats) within a constantly repeating rhythmic cycle". So Drumming is made of four parts, the first scored for four pairs of tuned bongo drums, the second for three marimbas and two soprano voices, the third for three glockenspiels, whistling and piccolo flute, the fourth combining all these instrumental groups. Each part develops the same melodic patterns, and at the beginning of each next section the new instruments enter doubling exactly the pattern of the instruments already playing, and these then taper off. So, here, all you are told is that Reich whistles, Mort Silver plays the piccolo and Pamela Wood Ambush and Jay Clayton (the latter was already part of the 1974 recording) sing. But from the members of the 1974 recording, you can make the educated inference that Russ Hartenberger and Bob Becker play the bongos and that Glen Velez, Bob Becker, Russ Hartenberger and James Preiss play the glockenspiels (they all did so in 1974). Among the nine performers who played the marimbas in 1974, five are still there in 1987: Reich, Hartenberger, Becker, Ben Harms and Glen Velez, and three new names appear whom I'll suppose play the marimbas as well: Gary Kvistad, Gary Schall and Thad Wheeler. Who's the 9th performer and whether there is a 9th performer, maybe we'll never know.
So, it all seems nice and well, but there is also a drawback to doing a "short" version: it doesn't always have time to unfold. Just one telling illustration. The piece starts with one bongo playing a single pitch in a regular beat, soon joined by a second playing the same pitch, and then followed by a second pitch. In 1974 second bongo entered after 13 seconds and second pitch at 0:23. In 1987 the second bongo enters, if my ears don't betrray me (it is almost indetectable) at 0:05, and the second pitch at 0:08. No wonder then that the new recording should be about as short as, according to Reich himself, the piece can be. Just compare section by section (the side breaks aren't exactly the same; add 30 seconds to tracks 1 & 2 of the 1987 recording (and substract as much from track 3) to make things comparable).
Part (1974) 24:35 (1987) 17:31
Part II (1974) 25:19 (1987) 18:11
Part III (1974) 15:40 (1987) 11:11
Part IV (1974) 18:57 (1987) 9:50
This may be not so much a drawback as an advantage for some people, likely to find the 85 minutes of repetition of the 1974 recording a little too much to swallow. I do find the 24+ minutes of Part I somewhat too long and repetitive, because it involves the relatively bland colors of the skin-struck bongos. But I am not sure that the shorter version will sound less bland and boring to those inclined to find it bland and boring, because what is lost even in that movement is the slow building of the patterns, the ritualistic aspect of the music (Reich acknowledge the influence on the composition of his voyage to Africa in 1970). What's left, in the two outer parts, is only the energetic and tension-filled sense of rhythm, and, in Part IV, of an inexorable crescendo and rise in tension. Part IV is more hypnotic in 1974. Despite the significant timing discrepancy I didn't find that the difference was as perceptible in Parts II and III. What is immediately perceptible though is that it is more dynamically and propulsively played in 1987, which accounts at least for some of the timing difference. No judgment here: the sweeter and more dreamy approach of 1974 also as its value. I like both in fact. Still, Part III, with its tinkling bells, is so exquisitely fairy-tale like, I am happy with the 4 additional minutes I get in 1971.
Is there at least a significant sonic improvement that would make the 1987 preferable, if only on that ground? Not really. If anything, there is slightly more vividness in the earlier recording of the bongos. The 1987 sonic perspective is slightly more distant and airy. Strangely, it is the opposite in the other tracks, but there differences are marginal.