How Musicians Push Our Emotional Buttons - Why You Love Music: From Mozart to Metallica-The Emotional Power of Beautiful Sounds - John Powell

Why You Love Music: From Mozart to Metallica-The Emotional Power of Beautiful Sounds - John Powell (2016)

Chapter 14. How Musicians Push Our Emotional Buttons

image

Whether they are performing a written piece or improvising, musicians have an array of techniques they can use to maximize the emotional effect of the music. In this chapter we are going to look at musical performance over a wide range of styles to see how musicians cast their spells over us.

Performing musicians in different genres have differing levels of control over what you hear. A classical keyboard player performing a Bach fugue on a harpsichord must play all the notes in the right order and has control only over the speed of her playing. She can slow down and speed up at various points and introduce dramatic pauses, but that’s about all the flexibility she has. When the same musician is playing a romantic piano piece by Chopin later in the week, she has slightly more influence over the sound because on a piano you can vary the loudness of the music. But once again, all the notes must be produced in the correct order.

On Saturdays our busy musician plays in a pub rock and blues band. When they’re performing Led Zeppelin’s iconic “Stairway to Heaven,” they know that they need to be as faithful as possible to the original recording (because a lot of people are familiar with it), but no one is going to complain about a few missing harmonies or a simplified guitar solo. By midnight the band members are meandering their way through a twenty-seven-minute version of “The House of the Rising Sun.” Audience expectations are completely different in this case. It’s a traditional folk song and the famous 1964 cover of it by the Animals is a well-known launchpad for endless blues improvisation—so, apart from the last two minutes and the first two minutes, the band will be improvising and free to produce whatever notes they want, as long as they stay within the standard rules of blues improvisation.

In all these musical environments our keyboard player will be using various techniques to give the music its maximum emotional impact. The same is of course true of musicians from traditional non-Western genres, who improvise entire performances without reference to songs or tunes the audience might have heard before.

We’ll start our discussion of these techniques by looking at Western classical musicians who are trying to reproduce a piece of music written as dots on a page several hundred years ago. How do you inject any human emotion into a performance like that?

It is, as I have said in earlier chapters, possible for a computer to evoke an emotional response in a listener simply by producing the right notes in the right order, but real musicians do the job much more effectively. As the authors of Psychology for Musicians put it:

The crux of expressive performance is nuance. Nuance is the subtle, sometimes almost imperceptible, manipulation of sound parameters; attack, timing, pitch, loudness and timbre, that makes music sound alive and human rather than dead and mechanical.1

Phrasing

One of the main differences between a dry performance by a computer and a more emotional one by a musician is that the musician will automatically (and usually unconsciously) divide the music up into phrases. Phrasing is as important to music as it is to speech because it allows the listener to experience the music as bite-sized chunks rather than a continuous stream of notes. The core reason behind our preference for phrased musical sounds is linked to our predilection for grouping items or events together,2 which we covered in chapter eleven.

In some cases the beginning and end of a phrase is obvious. The tune to the line “Baa, baa, black sheep, have you any wool?” is a musical phrase that clearly begins on the first “Baa” and ends on the long note on “wool” (tunes often end on long notes). But lots of pieces of music consist of an unbroken stream of equal-length notes, and the beginning and end of each phrase is not at all clear if you just look at the printed score. If a simple computer program performs this sort of piece, it will produce a sequence of equal-length, equal-loudness notes, but we listeners will automatically use our previous musical experience to divide the music up into phrases that are usually only a few seconds long. Human musicians will make things a lot easier by phrasing the music for us, but what are they actually doing?

Modern recording techniques have made it possible for musicologists to make a detailed study of the loudness and duration of every note in a performance. This type of analysis has revealed the fact that musicians use subtle (and sometimes not so subtle) variations in loudness and speed during each phrase. In general, the middle part of each phrase tends to be slightly louder and faster than the beginning or the end.* Because most phrases last for only a few seconds the performer is constantly varying the volume and speed, but the changes are usually so subtle that we don’t consciously notice them.

Obviously, musicians are also listeners, and they pick up the concept of phrasing from all the music they have heard and then naturally apply it to the music they produce, in the same way that we learn to phrase our spoken sentences from listening to others. But some of our musical phrasing might be instinctive rather than learned. Psychologists Amandine Penel and Carolyn Drake have looked into the possibility that slowing down at the end of phrases is partly involuntary.3 They suggest that our hearing system tends to make the final two notes of any group sound closer together than they actually are, so musicians automatically stretch out the timing at this point to compensate for the perceived rush. They recorded eight professional pianists playing a romantic piece in a normal, musical way and then asked them to play the same piece with no emphasis, as a machine would. On average the musicians, when they were playing normally, slowed down by about 30 percent toward the end of a phrase. Later, when they thought they were playing like an accurate robot, they were in fact still slowing down at the same places, but only by about 10 percent. The musicians couldn’t hear themselves slowing down; it just happened automatically.

Another piece of research into phrasing by a team led by Finnish psychologist Eva Istok involved listening to a series of four recordings with a range of phrasing styles:

1. extreme phrasing (unusually high levels of acceleration toward the middle of a phrase and equally unusual slowing down at the end)

2. normal “expressive” levels

3. deadpan (no phrasing)

4. inverted phrasing (slower in the middle of phrases and accelerating toward the end)

They found that melodies played with inverted phrasing were rated as sounding less pleasant than any of the other phrasing strategies, including the deadpan performance. I’ve never heard any music with inverted phrasing because it’s extremely counterintuitive and you need a computer program to manipulate the speed of the notes, but I’m sure Eva and her colleagues are right when they say that “inverted phrasing simply sounds wrong.”4

Emphasis

As their training develops, musicians learn that some types of music benefit from more exaggerated phrasing than others. A romantic prelude by Rachmaninoff can survive overdose levels of phrasing, but a piece by Bach can sound forced or amateurish if the phrases are emphasized too heavily.

Whatever the type of music involved, we can think of each phrase as having its own arc of loudness and speed: increasing toward a maximum somewhere near the middle, and diminishing again at the end. Pieces of music generally last several minutes, however, and if the musician just kept on repeating the same loudness/speed arc for each phrase, the piece as a whole would sound lifeless. Most pieces of music have a main climax and several minor climaxes on the way, and the musician needs to build the music up to these climax points and then relax it after them to give the whole experience an ebb and flow of emphasis. During the course of the piece a sort of architecture is built up of arches within arches.5 Like this:

image

Written music for a solo performer is like a written speech, and the musicians need to do the same job as a good orator or actor: they have to breathe life into it all by using a hierarchy of emphasis. Subtle differences in emphasis can have a big effect on the emotional content of a piece of music just as they can in speech.6 For example, if I emphasize different words in the simple sentence “Fred loves Pete’s banjo recordings,” we get a variety of meanings. In each of these examples I’ve put the emphasized word in italics, and the statements in parentheses show the alterations in meaning:

Fred loves Pete’s banjo recordings. (But no one else does.)

Fred loves Pete’s banjo recordings. (He’s surprisingly keen on them.)

Fred loves Pete’s banjo recordings. (But he doesn’t like recordings by other banjoists.)

Fred loves Pete’s banjo recordings. (But dislikes recordings of Pete playing the flute.)

Fred loves Pete’s banjo recordings. (But doesn’t like Pete’s live performances.)

Musicians can use a number of techniques to emphasize the different levels of climax within a piece of music. As we’ve just seen, loudness and speed are very important tools, but deliberate timing adjustments, dramatic pauses, and changes in timbre can also have a big effect.

Rubato

Classical musicians are trained to obey the composer’s instructions, which are simply represented as dots on a page together with a few, often vague written instructions such as “almost walking speed” or “majestically.” (Traditionally these instructions are written in German, French, or Italian: andantino,* maestoso.)

One instruction that occasionally crops up is the Italian word rubato, which means “stolen.” If this word occurs in the score, it means that the musician should play some notes and passages quicker than they are written, and others should be played more slowly, to bring out the maximum emotional effect. Composers use the word “stolen” because some of the time available for the whole piece has been stolen from certain sequences of notes and given to others. This effect is usually associated with romantic music like pop ballads and composers such as Rachmaninoff and Chopin, rather than music that requires a steadier beat like techno or Bach.* Musicians use their judgment in choosing where to hurry and where to dawdle, and the most common technique is simply to make the short notes a bit shorter and long notes a bit longer.

If the score doesn’t include the instruction “rubato,” the written notes have, in theory, precise pitches and lengths, but the musicians, particularly the experts, don’t obey the instructions exactly. Expert solo musicians often take far more liberties with the written rhythm than competent amateurs do.7 But of course there is nothing arbitrary about these timing changes: the experts are using rhythm adjustments in an attempt to maximize the emotional effect of the piece. Careful observations of virtuoso classical pianists have revealed that, although they make considerable changes to the lengths of the notes involved, they do so repeatably. Asked to play the same piece months later, they play the work with almost identical stretching and shortening of particular notes and phrases unless they have specifically decided to rethink how they play the piece. And this ability to reproduce the nuances of a performance doesn’t just apply to classical musicians. Musicologist Richard Ashley has found this type of reliability in separate recordings of “Yesterday” by Paul McCartney.8 You might assume that this type of performance stability comes from memorizing individual note values and emphases, but that’s not the case. Instead, the performer is only remembering a general performance strategy for that particular piece which has worked in the past.

Deliberate mis-timing

Lots of pop singers add extra emotion, life, and movement to their performance by singing occasional notes a little early or a little late. I was going to spend hours listening for a good, clear example of this, but fortunately Howard Goodall, author of The Story of Music, has already found one for us.9 Here’s the first verse of Bob Dylan’s song “Make You Feel My Love”:

When the rain is blowing in your face

And the whole world is on your case

I could offer you a warm embrace

To make you feel my love

If you listen to Adele’s 2008 cover of the song, you can hear that she sings the last two notes of the first three lines early. Listen for the words “your face,” “your case,” and “embrace.” If you tap along with a regular rhythm, you’ll hear that these words begin just before your tap. The written music would place the beginning of each word on a tap—but in fact your tap will hit the middle of the word instead. Singing words early is very effective and it’s also quite difficult to do. I’ve been trying for almost ten minutes now and can’t get anywhere near it. So there’s no reason to feel anxious, Adele. I won’t be applying for your job anytime soon.

Loudness and timbre

Even if a piece is played with inhuman precision by a computer, there are parts of the music that are more significant than others—the culmination of a climb up to a long, high note, for example. One way for a (human) musician to add extra emphasis at these moments is to pause just before the climax, or to play the climax unexpectedly loudly, or even quietly. It’s the unexpected nature of the emphasis technique that matters, so a sudden reduction in loudness can be just as effective as an abrupt increase.

Changes in timbre can also be used as an emotion-enhancing tool. (The timbre of an instrument is its distinctive sound or voice.* It’s why a saxophone sounds different from a banjo even if they are playing the same note.) The timbre of some instruments can be quickly changed from smooth to harsh and back during the course of a piece. This is another way to emphasize climaxes or to create emotional lulls within the music.

And I think this is the point at which we must take a few minutes out to feel sorry for harpsichord players…

Harpsichords were the precursors to pianos, but for reasons that will shortly become apparent, they fell rapidly out of favor as soon as the early piano designers finally got pianos working properly. The British conductor Sir Thomas Beecham described the harpsichord as sounding like “two skeletons copulating on a tin roof in a thunderstorm,” which is a bit harsh, although it is rather jangly and clunky-sounding. But the big problem with harpsichords isn’t their jangly sound; it’s the fact that the player has no control over the loudness or timbre of the notes.

A note on a harpsichord is produced by pressing down on a piano-like key, but that’s where the similarity between the two instruments ends. On a piano, the key you press is attached to a series of ingenious levers that eventually fling a felt-covered wooden hammer at strings (there are two or three strings for each note). The quicker you depress the key, the faster the hammer is traveling when it hits the strings. Quicker hammers give louder notes with a slightly harsher timbre—so the pianist can control the level of emphasis.

In the case of a harpsichord, the pressing of a key moves a lever that has a small spike (made from the quill of a crow’s feather) sticking out of it. When the key is pressed, the spike plucks the string to produce a note. And no matter how fast or hard you hit the key, the plucking action doesn’t change; the string can’t be plucked harder or softer to produce a louder or quieter note. So on a harpsichord, neither the volume nor the timbre of the notes produced can be altered,* which means the only emphasis control the player has is to speed up or slow down. This limitation is the prime reason why the solo harpsichord has a very low tear-jerk rating and is rarely used as an accompaniment to kissing scenes in films… unless the couple doing the snogging are vampires.

Pianos, as I just mentioned, do offer control of the loudness, but although the timbre changes slightly, it is not actually controlled. All that happens is that an increase in loudness is accompanied by a slight increase in harshness of tone because the louder note contains a slightly greater proportion of its higher harmonic frequencies, which makes the sound a little more aggressive.

Timbre control is better on stringed instruments because, as you pluck or bow the string closer to its end (i.e., nearer the bridge), the timbre gets more metallic and tense-sounding.* Saxophones and other wind instruments are also pretty good for timbre manipulation, but arguably the best instrument for timbre control is the human voice. For the voice, as for other instruments, our perception of different timbres having different emotional content is probably based on our experience of hearing people talk in different emotional situations. When we are calm, our vocal cords are relaxed and we don’t use much breath to speak. The passage of low-pressure air over relaxed vocal cords means that the “notes” we produce as we speak contain a very low proportion of high harmonic frequencies. But if we get angry or frightened, our vocal cords get tighter and we use more air to speak. This combination introduces a greater proportion of higher harmonics into the vocal mix, and we therefore associate this type of timbre, even in musical contexts, with tension, harshness, and excitement.10

Legato and staccato

On top of all these variables there is also the choice between producing the notes as a sequence of separate entities (as you might say the words “What do you mean?”) or running the sounds together in a joined up slur (“Whaddyamean?”). The “running notes together” technique is called legato, and the “distinct separation of notes” is called staccato. Legato is more common than staccato and generally doesn’t just involve making the notes join end to end in a continuous stream; there’s often an overlap between them. On many instruments it’s possible to let the previous note keep on sounding for a while after the next note has started. This is a very common technique in guitar playing and a central feature of harp music. Pianos can do it, and if the right-hand pedal on a piano is pressed down, the overlaps between notes become much more pronounced. This pedal is often mistakenly called the loudness pedal, but it doesn’t make anything louder. It just allows the notes to ring on for a while so the sounds overlap.

The king of note overlap is the organ. On pianos, harps, and guitars, the earlier note is always dying away as the subsequent notes are played, but on an organ, the old notes retain their power as the new notes are introduced. You can start off with one note and add others one at a time until you’ve got a huge powerful chord with up to ten notes in it. J. S. Bach uses this technique a few seconds into his Toccata and Fugue in D Minor just after the famous opening flourish.

It’s impossible to produce overlapping notes on a wind instrument such as a saxophone or flute because they are designed to produce only one note at a time. But wind instrument players can at least make the notes join end to end by producing a stream of notes in one breath. For the “separate note” staccato effect, a wind instrument player will provide a separate burst of air for each note, and a piano player who wants to play staccato will use the left-hand pedal, which damps the sound, making the notes die away more quickly than usual.

Playing music staccato tends to increase the tension and energy in the sound, and legato makes everything sound smoother and more relaxed.

So there we have it. There are lots of ways of adding emotional content to a musical performance—even if the starting point is a couple of thousand dots on a page written down three hundred years ago.

Of course all the techniques I’ve discussed so far are available to musicians across all genres, and they use them all the time—whether the music is composed by a jazz musician, written by a garage band, or improvised by a sitar player. Nearly all musicians use phrasing, loudness, and speed variations, emphasis, timbre, and detached or slurred-together notes. But there are some exceptions to this free-for-all. While many musicians use rubato to some extent, there is one group of musicians who never do: African drum ensembles. One of the oldest African musical genres is drums-only music, which often employs polyrhythms. A polyrhythm is produced when two or more drummers follow different but linked agendas at the same time, creating complex and interesting effects from a combination of rhythms. For example, if one drummer taps with a completely steady beat and emphasizes every fourth beat, you would hear the equivalent of this pattern:

image

Boring, eh?

Now, if you ask that drummer to stop playing and get a second drummer to tap along at the same speed, emphasizing every third beat this time, you get another uninspiring rhythm:

image

But if you ask both drummers to play their dull rhythms at the same time, you get something quite complicated:

image

(The biggest I’s show when the two drummers both emphasize a beat at the same time.)

With this simplest of polyrhythm rules (try it with a friend, tapping on a table), we have created an emphasis pattern that repeats only every twelfth tap and would be considered very jazzy and advanced in Western music. It would, for example, fit very well into a complicated classical piece like Stravinsky’s Rite of Spring. But to a couple of African drummers, this rhythm would be laughably simple. Members of an African percussion group who had never heard Western music before might find tunes and harmonies by Beethoven or the Beatles interesting, but they would consider the rhythms simplistic to the point of being childish. Using polyrhythms, they might be working in patterns of sevens and nines with double and triple taps rather than the single ones in our simple example. Because polyrhythmic music requires a strict underlying beat, rubato must be avoided because it would muddy the effect of the drumming patterns coming in and out of synch with one another.

And I see we’ve drifted into the realm of world music—which brings us nicely into a discussion about improvisation.

Improvisation is simply making music up as you go along—and my use of the word “simply” here is completely misleading, because there’s nothing simple about it.

Western classical music composers writing between about 1800 and 1900 frowned on improvisation as the unworthy pastime of ne’er-do-wells. That attitude has, in most cases, persisted in the genre to this day, and apart from minor deviations in timing for emotional effect, Western classical musicians are generally taught to follow the instructions of (usually dead) composers precisely. Most other Western music genres also concentrate on faithful re-creations of composed pieces, with only a smattering of improvisation.

As far as most of the rest of the world’s traditional musicians are concerned, this approach is deeply weird. Professional musicians from non-Western traditional musical cultures train just as hard as Western professionals, but much more of their training concentrates on the ability to improvise.

At this point I need to explain a bit more about improvisation. When you improvise, you don’t allow yourself total freedom. You choose your notes carefully, according to the rules appropriate to whatever genre of music you are playing—and there are always rules. Rule-free improvisation would be like rule-free conversation: a frustrating, unrewarding mess. This comparison with conversation is quite useful because conversations are improvised. You piece together your conversation by following a theme and expanding upon it while carefully monitoring what your friend is saying. You are (subconsciously) juggling dozens of rules and guidelines that prevent you from making mistakes. For example, if your friend says her new haircut makes her look stupid, do not enthusiastically agree. Also, we don’t keep inventing entirely new sentences. We use stock phrases such as “I’m looking forward to it” and “See you later.” Improvising musicians often use standard musical phrases in exactly the same way.

In Western music, jazz is the most improvisatory genre. Many jazz fans assume that there is almost unlimited freedom in modern jazz improvisation—but the musicians know otherwise. It’s true that some jazz ensembles strenuously avoid straightforward tunes, like the mythical band led by saxophonist Progress Hornsby, alter ego of TV comedian Sid Caesar:

We’ve got a nine piece band where the ninth member plays radar—to let us know if we get too close to a melody.11

But usually, beneath all the seeming spontaneity and complexity, there is a musical agenda understood by all the musicians whether there is a tune involved or not. Here’s what the world-renowned jazz pianist Dave Brubeck had to say about the rules of improvisation in jazz circles:

Interviewer: Are there any rules for improvisation?

Dave Brubeck: You bet your life there are, and the rules in jazz would just scare you to death. They’re so strict it’s fearful. Just break one of the rules and you’ll never end up in another jam session with the same guys again.12

We’ll come back to modern Western music later, but I want to return to what all those non-Western musicians were doing. How do they train to improvise?

Let’s take the example of Iranian classical music (“classical” here meaning long-established music that requires a lot of training). To become a full-fledged Iranian classical musician, you must first memorize the Radif, about three hundred short pieces of music, each between thirty seconds and four minutes long.13 These short tunes aren’t considered complete pieces in their own right; they are the building blocks from which you will produce your improvisations. It takes about four years of hard study and practice to learn your three hundred pieces, and it would take eight to ten hours to perform them all end to end. But of course they are not supposed to be played that way. The idea is that a performance will be built up of variations on a few of the pieces.

In Arabic, Turkish, or Indian music the rules are rather looser. As a musician starts a piece, there are four basic ingredients involved:

1. An idea of the mood the musician is trying to produce.

2. The chosen selection of notes he or she will be using.

3. A selection of short tunes that are typical for the chosen mood and note combination.

4. Many thousands of hours of practice and training.

One of the most common forms of improvised Indian classical music is the raga, often performed by a sitar player together with a percussionist and a third musician playing a tanpura, an instrument that provides a continuous drone. Traditionally ragas are in four sections. For the first few minutes the sitar player plays slowly without a clear rhythm. To Western ears this sounds like he is just warming up. But what he is doing is introducing the audience to the family of notes he will be using during the piece.

This family of notes is actually the raga. A raga is in some ways similar to a Western key in that it generally has between five and seven notes in it. But a raga is more than just a collection of notes. Each raga comes with a set of rules governing how the notes are to be used in certain situations. Some ragas might use only five of the notes on the way up but seven on the way down. Others have a rule that, having climbed to a certain note, you must double back on yourself a note or two before you can continue upwards (and the same type of rule might apply to a different note on the way down). It’s also common for certain notes to be emphasized more than others. As a result of such rules, different ragas have individual “catchphrases” that help the audience identify them.14 Identifying the raga can be difficult for the untutored Western ear, partly because sitar players bend the strings of their instrument even more often than blues guitarists do, and notes are rarely played without some embellishment or other. But of course you don’t have to know anything about the raga to be able to enjoy the music.

Once the slow, meandering introduction is finished and the group of notes has become familiar to the audience, the sitar player starts to give the music a definite pulse for the second section. The improvisation then often gets fast and furious in the third section before, finally, the percussionist joins in for the finale.

The rhythms used in ragas are also subject to strict rules. The rhythms are not generally written down but are learned with the help of mnemonics such as “TakitaTakitaTaka,” which, if you say it without gaps (emphasizing the three “Ta” syllables), gives you a 1,2,3,1,2,3,1,2 rhythm. (It’s best to repeat the mnemonic two or three times to hear the rhythm properly.) One of the most important rules of raga improvisation is that you must never lose your place in the rhythm, which can be very tricky when you are dealing with complex rhythmic patterns. Whereas Western rhythms can be identified after half a dozen notes, one of the Indian rhythms takes 108 beats to complete its full cycle.15 The percussionist (who usually plays the tabla, a pair of hand drums of different sizes) and the sitar player often take turns with short rhythmic phrases, chasing each other around at incredible speeds—but their years of training ensure that they are always completely in control, synchronizing exactly whenever they want to, particularly on the final note of any climax.

Improvisation in modern Western music

Jazz from the 1930s onwards and rock in the 1960s and 1970s often involved improvised solos and duets in the middle of performances of prewritten pieces. The general idea was that the harmony of the original piece would continue and the soloists would devise new melodies over it. The new tunes would follow the standard melody rules we saw in chapter ten, for example:

✵ Lots of small jumps in pitch with few large jumps.

✵ Big jumps upwards are usually followed by a smaller jump downwards.

✵ Emphasis of notes when the tune changes direction.

✵ Use of notes 1 and 5 at points of rhythmic emphasis.

And one extra rule that we haven’t come across so far:

✵ The emphasized notes in tunes will usually be found in the harmony that is playing at that moment. (So, for example, a rock, jazz, or blues band will often be going through a predictable cycle of chords while the lead guitarist performs a solo. If the chord at any point includes the notes A, C, and E, then the guitarist will tend to use one or more of these three notes as the emphasized notes in his solo. As the chord changes to another group of notes, one or more of these new notes will now be emphasized. Written or composed music also tends to follow this rule, but in that case the chords are chosen to follow the emphasized notes in the tune, rather than the improvised tune following the chords. In both cases the chords and the tune usually have some notes in common.)*

Obviously the musician wouldn’t be running through these rules consciously; he may not even know about them. The whole process is automatic and a consequence of all those hours of practice. This type of improvisation is still a feature of many rock and blues concerts: you play half a song, then improvise over the song’s harmony, and finally you finish off with the second half of the song.

Although some jazz musicians still follow this tried-and-tested recipe, the innovators have experimented with different approaches. Bebop, the new jazz of the 1940s, was an extension of the standard technique and involved chords that had five or six or more different notes in them instead of the standard three or four. This gave the soloists more freedom to choose whatever notes they emphasized in their melodies—but the improvisation was still based on the harmony. In the late 1950s, trumpeter Miles Davis started to experiment with the idea of melodic improvisations that involved choosing a certain collection of notes, to be played over simpler chords that didn’t change much (a similar concept to the Indian raga). This technique, called modal jazz, is the basis for the most popular jazz album ever produced—Kind of Blue.

An improvising member of a band is involved in a musical conversation with the other band members, and this involves a lot of fast thinking. In real conversations you will no doubt be familiar with the irritating phenomenon of thinking of exactly the appropriate response to a comment approximately fifteen minutes too late. But we can’t all be Winston Churchill:

Lady Astor: Winston, if you were my husband I’d poison your tea.

Churchill: Madam, if you were my wife I’d drink it.

During any improvisation the musicians (like Churchill) have to make split-second decisions about how to respond to what’s going on. They’ll often play it safe and stick to note sequences they know well, which explains why so many 1970s lead guitar solos have similar bits in them. I suppose someone must have been the first to play diddle-doddle-diddle… diddle-doddle-diddle… diddle-twang on the electric guitar… but his name was lost in the clouds of smoke from patchouli-flavored joss sticks back in 1968. Individuals have their own favorite building blocks from which to construct solos (rather like the Radif mini-tunes I mentioned earlier). You might think that your favorite guitarist is far too good an improviser to be relying on a collection of favorite clichés, but the reason you can recognize a solo by him within a few seconds, even if you’ve never heard it before, is that he has a distinctive improvising style, and distinctive styles depend upon a collection of personal clichés.

But you’re less likely to chance upon a seam of beauty if you just stick to the obvious choices and clichés, so highly skilled musicians occasionally opt for more unusual notes or chords. Sometimes this pays off and something sublime happens as two or three musicians suddenly shift direction together. More often than not it just sounds OK. And, of course, sometimes the adventurous choice turns out to be something everyone regrets.

Although some improvisation mistakes are the result of an error in judgment, it’s far more common that a wrong note is simply the result of misplacing a finger on the instrument or blowing too hard. The important thing, as far as the musician is concerned, is what to do after you’ve played a wrong note. How do you recover? The answer is as simple as it is surprising: the best thing to do is to repeat the error. Play the same passage again and make sure you include the wrong note. Doing this has two effects. First of all, because the unexpected note has just been heard, when it comes around again it sounds less unexpected and fits in better.16 Second, repeating your mistake misleadingly confirms to the listener that you intended to play that note in the first place. You’re basically trying to pull off a con trick by saying to your audience, “Oh, no, I didn’t make a mistake. I did exactly what I intended to do. It’s not my fault if you are too unsophisticated to appreciate my adventurous choice. Anyway, I’m prepared to forget the whole misunderstanding if you are.”

This is quite a complicated message to get across in one bum note—but it often works.

When improvising musicians aren’t making mistakes and callously blaming the listener for them, they often use the characteristics of the human voice to create emotional responses. I mentioned earlier the effective technique of making the timbre harsher during the emotional climaxes because our voices do the same thing, but improvising musicians can take this one step further because they can also choose the notes being played. Using computer-generated note sequences, psychologists Klaus Scherer and James Oshinsky have demonstrated that we associate high pitch levels, rising pitch contours, and a fast pace with anger and fear, probably because angry, fearful, or excited people talk quickly at a high pitch that rises toward the end of statements.17 Improvising musicians tend to use all these features when they want to reach an exciting climax. If you listen to any recordings of improvised music, you’ll find very often that emotional climaxes involve harsher timbres, increased loudness, increased pitch levels, and rising pitch contours.

Whatever the genre, from raga to reggae, composed or improvised, we’ll occasionally go to an exceptional concert and hear a truly great, deeply moving performance. At these times it’s easy to think that the greatness is an intuitive and effortless outpouring of superhuman musicianship, but the unglamorous truth is that great performances are the result of a lot of repetitious, detailed hard work.18

Since the 1980s there has been an increasing trend toward collaboration between musicians from different genres and cultures. In the UK the WOMAD (World of Music, Arts and Dance) festivals, which began in 1982 under the leadership of Peter Gabriel, have grown in strength, and what used to be a minor interest has become the vast genre known as world music.* As far as music is concerned, we’ve never had it so good.