CHAPTER 3

Music and Your Emotions

While music (without lyrics) is pretty useless at telling a story, it can express and evoke emotions. Before we go any further, I’d like to clarify what we mean when we talk about emotions. Emotions are not the same as moods; we are always in one mood or another, but we are not always experiencing an emotion.¹ Emotions are relatively brief and intense, and they are often linked to unconscious physical reactions such as changes in skin temperature. If an emotional response is music-related, it will be synchronized with the music—that is, the music will trigger the emotion.²

Perhaps the most important thing to bear in mind about emotions is that they are biologically evolved reactions which are vital to human survival.³ From a survival point of view, emotions such as disgust and fear help us to avoid, or run away from, potentially dangerous situations. Anger helps energize us to deal with threats, and happiness steers us toward rewarding situations such as eating and sex. So it’s a bit odd that something like music, which has no obvious link to survival, can generate emotional responses—but we’ll come to that point later.

Music and emotion are linked in two ways. In some cases we simply recognize which emotion the composer intended us to feel, but we don’t engage with it. The other response is when we don’t just monitor what’s going on; we get emotionally involved. When the “fear” music comes on for one of James Bond’s enemies, we don’t get fearful; we just think “Ha!—eat metal death, you unpleasant cesspit of moral turpitude!” But if the fear music comes on for James’s girlfriend, we get frightened along with her. (This is usually a perfectly justifiable fear, as the average life expectancy of a James Bond girlfriend is about thirty-five minutes.)

The early days of research into the emotional content of music involved a lot of hopeful guesswork as well as some brilliant insights. The pioneering author Deryck Cooke laid a lot of interesting groundwork, but he was over-fond of equating music with language. As we shall see, there are some links between speech and music—but they aren’t the ones he had in mind. In his book The Language of Music, published in 1959, Deryck claimed to have identified sixteen musical devices that had definite meanings. For example, the notes of a major chord, played one after another, from the bottom note upwards, were supposed to convey “an outgoing, active assertion of joy.”4 Unfortunately for the long-term sales of his book, the evidence shows that this type of direct speech–music translation simply doesn’t work.

Deryck was particularly keen on the idea that major keys bring happiness and minor keys evoke sadness.^* As we’ll see in a little while, there is some truth in this idea, but Deryck took a fairly hard line on the issue and therefore found it inconvenient that some cultures—the Spanish and Bulgarians, for example—use minor keys for quite a lot of their happy music. His response to this fact was to argue that the Spanish and Bulgarians were so used to a hard life that they hadn’t had time to acquire a “belief in the individual’s right to progress towards individual happiness.”5 In other words, they were so downtrodden and unhappy that they couldn’t be expected to create the right sort of (major key) happy music. Even in the 1960s this sort of thinking was easily identifiable as balderdash, and over the next few decades Deryck fell into the obscurity that befalls all those who don’t know how to spell their own name properly.

Faced with the complex problem of how music affects human emotions, modern researchers decided to start from scratch—so they began with the fundamental question…

Does music actually generate emotions?

Everyone agrees that music creates emotional responses, but that isn’t proof that it’s actually true. The psychologists set to work and soon established that people are pretty good at spotting the intended mood of a piece of music, even if it’s performed “deadpan” by a computer. In one test, researchers asked five composers to write six pieces of music that were individually meant to portray joy, sorrow, excitement, dullness, anger, and peace. The pieces were then played by a particularly dispassionate computer, and the different types of emotion were easily identified by listeners.⁶ It has since been confirmed that a wide range of listeners generally agree on whether or not a piece of music is intended to be happy or sad and whether it is supposed to be relaxing or arousing. These are four of the basic emotional states, and they are the easiest for humans to spot and project. You usually can tell how relaxed or aroused and how happy or sad people are by their body language—or, on the phone, by their tone of voice, even if they are speaking in a language you have never heard before. But in music, as with body language, although you can spot the basic intended emotion, it’s not possible to interpret detailed emotional information. One person might feel a certain piece of music is happy/playful, and another might hear it as happy/proud.

And being able to identify the intended emotion of a piece of music is only half the story. Can music actually produce an emotional response in the listener? In their quest to answer this question, researchers came up with the cunning ruse of simply asking a lot of people about the effect music had on them. Professor Patrik Juslin and his team asked over seven hundred Swedes to tell them about their most recent emotional musical experience.7 And the results were very encouraging for the music industry.

Everyone questioned claimed to have experienced emotions while listening to music, and more than eight out of ten of them said their most recent emotional response had been pleasurable—with the five most common emotions being happiness, melancholy, contentment, nostalgia, and arousal. But even the opinions of seven hundred Swedes are not proof of the existence of musical emotions; for proof you need to use things that don’t have opinions: you need machines.

Psychologists are never happier than when they can plug hapless victims into machines that go beep. Such devices can measure, among other things, your heart rate, skin temperature, the electrical conductance of your skin, and the amount of muscle activity in your face. When you are emotionally aroused, you use your facial muscles to smile or frown, your heartbeat changes pace, and your skin temperature can fall a little. For example, happy music produces greater skin conductance, a lower finger temperature, and, not surprisingly, more smiling than sad music does.⁸ It’s very difficult to fake these responses (except the smiling), which is why the police in some countries use the same measurements (via lie-detecting equipment) to detect signs of stress. Lots of experiments have been carried out using this type of equipment, and the results have shown that music can indeed create emotions.⁹ Interestingly, if music is calm or happy, we tend to get calmer or happier, but if music is designed to carry a negative emotion like anger, we generally just recognize that it’s angry. We don’t become angry ourselves, unless the music is playing in a movie, in which case the visual imagery and the music can work together to create a heightened response for both positive and negative emotions.10

Brain scanning equipment is the most reassuringly expensive of all the machines that go beep. These devices monitor activity levels in various parts of the brain as we change our thought patterns this way and that. We are not yet at the stage where the medics can look at brain scan images and say, “Oh, look—he’s thinking about carrots again,” but they can spot whether or not we are emotionally aroused. Ever since such equipment became available, psychologists have been examining the effects of various emotions on activity in different areas of the brain, and the results have been fascinating.

The amygdala, apart from being a very high-scoring word in Scrabble, is one of the most important parts of your brain for processing emotions. It’s particularly important in generating fear responses (although that’s not all it does), and is sometimes referred to as the “fear center” of the brain. A malfunctioning amygdala can result in mental disorders from depression to pathologic anxiety. In 2001, neuroscientists Anne Blood and Robert Zatorre used brain scanners to monitor blood flow to various areas of the brain while people listened to their favorite music. They found that there was an increase in blood flow to the areas of the brain associated with reward and positive emotions, and a decrease in blood flow to the amygdala.¹¹ So the pleasure centers were working hard and the fear center was taking the day off. Another study, by Stefan Koelsch and his team, confirmed that listening to joyful dance music increased the blood oxygen levels of various emotion-linked areas of the brain and showed a decrease in oxygen usage in the amygdala. (The amount of blood oxygen in any brain location is an indication of how hard that area is working.)12 The same project then studied the effects of unpleasant, dissonant music and found that it had the opposite effect of pleasant music: the subjects’ blood oxygen levels dropped where they had been high, and rose in the amygdala.^* Apart from proving that music actually does generate real emotions, this work indicates that music can be used to manipulate the amygdala in cases where it isn’t functioning properly. Music therapy, which we’ll be looking at in chapter five, can (and does) use pleasant music to calm down an overactive amygdala to alleviate anxiety and depression.

There have, of course, been lots of other investigations involving various types of brain scanners, and the overall conclusion is that the brain responds to music in a similar way to how it responds to other emotional stimuli.

To summarize all this in very simplistic terms:

Pleasant music stimulates the pleasure centers of the brain and calms down its fear center (the amygdala). Unpleasant music does the reverse.

Which type of music creates which emotion?

The tempo, or speed, of a piece of music is the clearest indicator of its emotional content. If the music is fast, it’s unlikely to be sad or tender/romantic. But fast music might be happy or angry, depending on factors such as how warm and consonant the harmonies are. Here is a quick summary of how some of the features of a piece of music are often linked to specific emotions.¹³

Tempo (speed)

Happiness: Fast, steady

Fear: Fast, varied

Anger: Fast, steady

Tenderness: Slow, steady

Sadness: Slow, varied

Key type

Happiness: Major

Fear: Minor

Anger: Minor

Tenderness: Major

Sadness: Minor

Average pitch

Happiness: High

Fear: High

Anger: High

Tenderness: Low

Sadness: Low

Pitch variation

Happiness: High

Fear: High

Anger: Moderate

Tenderness: Low

Sadness: Low

Harmony type

Happiness: Consonant

Fear: Dissonant

Anger: Dissonant

Tenderness: Consonant

Sadness: Dissonant

Loudness

Happiness: Medium to loud, steady

Fear: Quiet but variable

Anger: Loud, steady

Tenderness: Medium to quiet, steady

Sadness: Quiet but variable

Certain combinations of musical characteristics generally give rise to particular emotional effects. As you can see from the table, most of the happy, angry, or fear-inducing music you hear is fast, with high-pitched notes and tunes that jump around in pitch. By contrast, tender (romantic) or sad music is generally slow with low-pitched notes and far less movement in pitch.

If you’re a composer following these guidelines, you can make a happy tune more tender and romantic by slowing it down and playing it on low-pitched instruments (such as cellos). Or you can make a sad piece of music fearful by playing it on high-pitched instruments (such as violins) and increasing and varying the speed. Film score composers—who need to adjust the theme tune of a movie to fit a wide variety of situations—often use these techniques to manipulate our emotions.

But it’s important to remember that there are no firm rules in music; these are just generalizations to which there are many exceptions. The Jaws music, for example, is low-pitched and terrifying, and I’m sure you can think of plenty of songs that are high-pitched and sad.

The fact that the speed of the music gives the clearest indicator of its mood is the reason why some Spanish and Bulgarian music is cheerful even though it’s in a minor key. The music in question is moderately high-speed dance music, and the high tempo removes any sad influence that the minor key might have.

Obviously there are many more emotions than the five main ones listed in the table, and music can evoke more complicated feelings such as nostalgia, pride, and longing. On top of this, there are so many other musical techniques you can use to influence the emotional content that you can’t reliably forecast the emotional effect of a piece of music just by looking at how it’s put together.

One of the major variables involved is you, the listener. A wide range of factors can influence your emotional response to music, including your age, personality, musical preferences, familiarity with the music, and the mood you are in. My own response to a piece of music turned from moderate joy to speechless irritation in just twenty seconds a few weeks ago. My girlfriend, Kim, was driving us both back from Southampton to Nottingham the day after a friend’s annual barbeque. I was sitting in the passenger seat looking at the sky and enjoying the jolly banjo music on the stereo when suddenly a motorway junction appeared out of nowhere, and we were at the head of the queue at the roundabout. We had to join the motorway immediately—but should we be heading east or west on it? In the time-honored tradition of couples dealing with this sort of emergency, Kim raised the point that, as she was driving and I had the map open on my lap, it seemed logical to her that I might have prepared myself for this decision a little earlier, and spent less time staring gormlessly out of the window.

At least that was the gist of what she said.

I stared at the map. I stared at the road signs. I stared at the map again—and found that I just couldn’t think clearly with that dreadful jangly bloody banjo music playing. I turned it off and my head cleared immediately. “West,” I said—and we headed off into the sunset. Within a few minutes we were back in the correct frame of mind for jolly jangling and turned the music back on. If there’s one thing life has taught me it’s that banjos and map reading don’t mix. This, of course, fits in well with the research by Vladimir Konecni and Dianne Sargent-Pollock I mentioned in chapter one, which found that we don’t like complex music when we are trying to solve a problem.

Everyday musical experiences

In the Western world, about one third of the things we do every day are accompanied by music, and for around half of that time the music has some sort of emotional effect.14 In 2008 our friend Professor Juslin and his team carried out a detailed study on the effect of music on our everyday lives.¹⁵ They asked thirty-two Swedish students to carry around handheld computers that were programmed to beep seven times a day at random intervals. Whenever they heard the beep, the students had to answer a list of questions about what they were doing at that point, how they felt emotionally, and whether or not they could hear music. If they could hear music, there were further questions about the music itself and what effect they thought it might be having on their emotional state.

As I’m sure you’ll be delighted to hear, they found that, whether music was involved or not, calm contentment and happiness were the most commonly felt emotions, while negative feelings like guilt and disgust were quite rare. Some listeners experienced music-induced emotions far more often than others, but the addition of music to any situation tended to boost the number of happy or elated experiences and minimize the incidence of anger or boredom.

On top of this, whenever there was a mood change caused by the addition of music to a situation, it was almost always a change for the better, with the person becoming happier and more relaxed. This is partly, of course, because most music is designed to be pleasant and relaxing.

Now, if the subjects of this study had been from a population predisposed to giddiness—the sort of “carry me to the next bar, when does the carnival start?” party animals you might find in Brazil or Wales—we might be suspicious of the results. But this was a group of Swedes—the inventors of the ultra-safe Volvo, the vegetarian meatball, that most unhappy of all fictional detectives, Wallander,* and his even more miserable father. If music is capable of cheering up the compatriots of Wallander’s dad, we can be sure we’re on to a good thing.

So now it’s official: music is good for you. It generally cheers you up and/or relaxes you.

Which brings us to the question of why we sometimes choose to listen to sad music. We’ve all had the experience of selecting a downbeat album to match our mood after a bad day. At times like this the misery of life on earth feels a little more acceptable with an appropriate sound track: it’s you and Morrissey against the rest of the world. But there are other times when we choose to listen to sad music even when we are in a good mood. This sounds like a bad move, but here is what psychologist William Forde Thompson has to say on the matter:

Sad music, like Shakespeare’s tragedies, may be appreciated for its artful construction and for its ability to illuminate an emotion that is an essential part of being human. We can examine and appreciate the sadness conveyed in music while being reassured that there are no actual consequences to face.¹⁶

So there is a certain amount of comfortable satisfaction involved in listening to sad music when you’re not feeling sad yourself—as any country and western fan will tell you.

Background music and your emotions

As we saw in chapter one, much of the music you hear during the course of an average day isn’t chosen by you. It’s chosen by a committee of salespeople who have decided that this or that selection will optimize your willingness to buy shoes—or persuade you to linger longer over lunch in their trendy new restaurant.

A surprisingly large majority of people are quite happy to be fed a steady diet of music chosen by others, but there is a small minority who really don’t like it. These naysayers are concentrated in one social group—males over the age of forty, or, to use their more technical appellation, grumpy middle-aged men. The psychologist’s best guess as to why we grumpy middle-aged men don’t like background music is that we are used to having control over things around us.17 We don’t like it when we can’t choose, so we get tetchy and disagreeable, and we don’t like shoe shops anyway, so the irritating music gives us a good excuse to stalk off to the nearest pub.

But about three quarters of us find background music pleasant most of the time, and it generally supports, or moves us toward, a positive emotional state.¹⁸ The music involved is usually straightforward, and its emotional content can be identified very quickly. In some cases it takes only a second or so.¹⁹ Music that is produced specifically as the background to shopping or eating (as opposed to hits of the seventies) is deliberately designed to be non-distracting and utterly unmemorable—so it creates a sort of aural wallpaper, allowing you to focus on the pros and cons of the turquoise suede sandals as compared to the blue kitten-heeled winkle pickers. This need to be unobtrusive is the reason why aural wallpaper music is often made up of old, familiar songs (like “The Girl from Ipanema”) played without (attention-grabbing) lyrics on mellow-sounding instruments.

The result of listening to background music is generally a diluted version of your emotional response to listening to music you might have chosen. This could mean a minor reduction in boredom when you’re doing something tedious, or a gradual shift from fed-up to not-quite-so-fed-up. But as music psychologist John Sloboda points out, these small changes shouldn’t be dismissed as trivial. The accumulation of little positive pushes has been shown to enrich your life by improving how you perform both socially and intellectually.20

Useful music

The enjoyment of music seems to be restricted to humans. We have incontrovertible scientific evidence that marmosets don’t give a damn about music, and tamarins also have a tin ear.²¹ Of all the creatures tested so far, only one has been found to be capable of spontaneous bodily movement in time to the beat of music. Yes, you guessed it—put your hands together and let’s hear a big round of applause for the only other creature capable of getting down on the disco floor… the parrot. It’s possible that being able to move to a beat is somehow linked to the other capacity that parrots and humans share—the ability to learn how to pronounce words—but the link hasn’t been proved yet.²² The most famous of all dancing parrots can be found on YouTube if you look for “Snowball—our dancing cockatoo.” Watching Snowball pick up the beat to Queen’s “Another One Bites the Dust” is really impressive—and very funny. Neuroscientist Aniruddh Patel (author of Music, Language and the Brain) was astounded when he saw the video, and became even more impressed once he’d carried out tests which proved that Snowball really was responding to the beat.

Part of our enjoyment of music comes from the fact that we can use it to enhance our performance in certain types of activity. Music can be particularly useful in five roles in our day-to-day lives.

Apart from the obvious music-movement link in dancing, we occasionally use music to energize us and focus our attention on a physical task. In olden times sailors used sea chanteys to encourage and coordinate their team efforts—pulling on various ropes, splicing mainbraces, whittling wooden legs, and so on. Nowadays the crew are more likely to be using driving, rhythmic music to optimize their activities in the ship’s gym. Studies have shown that energizing music in the gym motivates people to stay on equipment longer and to increase their pace to match the pulse of the music.23 This can be of particular use to long-distance runners, which is why Rule 144.3b of the USA Track and Field organization bans headphones from competitions if awards or prizes are involved. In the less cutthroat world of charity marathons, headphones are often banned but for a different reason—namely the safety risks of not being able to hear what’s going on around you, particularly if the roads are not closed. As you’ll see if you visit runners’ websites, these rules are extremely unpopular with the competitors, many of whom feel that they can’t finish a marathon or half marathon without their favorite running music. I do feel sorry for the runners involved but, frankly, if I want to travel thirteen miles I take a taxi.

We can also use music to distract us from the tedium of a boring activity, and this is particularly useful if you are doing something dull but not quite mindless. If your job is simply to transfer hundreds of books from boxes to shelves, it may not be terribly boring because you don’t need to think at all, and you can daydream or chat while you are doing it. If, however, you have to put the books on the shelves in alphabetical order, then the boredom is monumental: you can’t do the task on autopilot, but you are using your brain for only about twenty seconds out of every minute. In this type of situation music is a godsend. Your attention can shift from the music to the task and back again every few seconds, so you feel less bored. To quote a marvelous phrase from psychologist John Sloboda, using music as a distraction “is a way of engaging unallocated attention and reducing boredom.”24 (I love the idea that you can have bits of spare attention hanging around on the street corners of your mind, waiting to be allocated.)

Occasionally we use music to enhance the meaning of a situation. Obvious examples of this are weddings and funerals—although it also applies to situations like choosing romantic music for a drive in the country with a new beloved.

And speaking of romance, music has been shown to aid the release of the hormone oxytocin into our blood. This hormone, which is also released during breast-feeding and sex, encourages bonding between people.²⁵ More generally, music encourages social bonding whenever we dance or sing in groups—as any soccer fan will tell you.

Another common use for music is stress reduction. We all have our favorite stress reduction pieces, and they don’t necessarily involve calm, relaxing music—just stuff we love. Stress reduction has become such a popular branch of music that you can now choose from a broad range of albums specifically designed to relax you. My dentist has one that sounds like a mash-up of whale song, harp strumming, and orchestrated hits of the eighties. I can’t even begin to tell you how annoyed and stressed it makes me feel. I must explain to him one day that my flinching and whimpering have nothing to do with the drilling or injections—it’s the thought of another twenty minutes of Lionel Richie and Culture Club played on panpipes. Surprisingly, this relationship I have with my dentist’s sound system brings us to an important point about emotional states generated by music: in most cases a strong emotional impact can be achieved only after you, the listener, have decided that you are going to encourage it to happen. The music can’t do the job on its own. You have to be a consenting partner.

Your earliest musical moment

I can almost guarantee that you won’t be able to remember the first time music had an emotional effect on you. Why am I so confident? Well, because it probably happened before you were born. Human fetuses are responsive to sounds almost two months before they are born: fast, rousing music speeds up their heart rate and slow, soothing music calms them down. A few weeks later, as they are getting ready to join the hustle and bustle of daily life, they show emotional responses to music that can be measured by both heart rate changes and differences in how much they move around in the womb.

Once the baby is born, the parents (particularly the mother) will often talk to their babies with different types of sing-songy lilt depending on whether they want to entertain them or calm them down. The soothing phrases—“There, there, sweetie, it’s OK”—usually involve low notes, descending melodies, and a very slow pace, rather like actual lullabies. Playful chitchat—“You lovely boy! Where did that noise come from?”—uses higher notes, a wider range of notes, strong rhythms, and lots of repetition. These motherly musical offerings, together with actual play songs and lullabies, have an effect on the levels of cortisol in the baby’s bloodstream,26 and your bloodstream cortisol level is associated with how aroused or relaxed you are.

Lullabies are so important to babies that they can even be used in a medical context. In 2000 Dr. Jayne Standley used lullabies to help premature babies who had difficulty feeding. Babies born before the thirty-fourth week of pregnancy haven’t developed the coordination to suck liquids without inhaling the milk, so they have to be fed by tube. Babies who can be taught to suck put on weight faster than tube-fed babies. With this in mind, Dr. Standley invented the Pacifier Activated Lullaby system, or PAL.27 When the baby sucks on the pacifier (or dummy, as they are called in the UK), the PAL plays a lullaby, and if the baby stops sucking, the music stops. The babies clearly enjoy the music, because after only a few minutes of practice they learn how to keep the music going by continuing to suck.

As babies approach their first birthday, they start to develop the ability to distinguish between happy and sad music, and by age four, kids can point to a picture of a facial expression that matches the mood of the music they are listening to.²⁸ From that age on they are also capable of singing the same song in a sad or happy way—if bribed with enough chocolate. But at this stage in their lives, “happy” simply means quick and loud, whereas “sad” means quiet and slow. (As adults we pick up several more sophisticated clues about the mood of the music, but we still retain this basic link between happiness, speed, and loudness.)

Between the ages of five and twelve we start to accumulate a library of expectations about music, and we can be amused or annoyed by stuff that violates those expectations. For example, once we’ve been exposed to a lot of tunes and songs, we acquire a subconscious ability to understand what key the music is in. We probably don’t even know what a key is, but we can hum the group of notes being used to produce the music—and could, for example, make a good guess at the note the tune will end on and what chords are likely to be used in the harmonies. If unusual or out-of-key notes are used, even kids as young as nine will find it laughable or weird.²⁹ During these primary school years we also start to identify minor keys with sad emotions and major keys with happiness³⁰—which, by happy coincidence, is the next subject we are going to look at.

Why are major keys happy and minor keys sad?

As I said earlier, you are an expert listener to music even if you have never been taught anything about it. Whenever you listen to a piece of music, you are subconsciously analyzing all sorts of things, like what genre it is, what rhythms are involved, and, as I just mentioned, what key it’s in—even if you don’t know what a key is.

So—what is a key, and how on earth do we identify them if we don’t even know what they are?

In Western music we have twelve different notes we can choose from, but we tend to use a selection of about seven of them at any one time (although the composer might change from one group of seven to another during the course of the piece). In any Western music the particular group of about seven notes involved at any one time is called the key.

As the music plays, you build up a clear picture of which seven notes are being used, so that if someone plays a note that isn’t in the same key it will sound wrong to you, out of place—out of key.

As you subconsciously gather your collection of seven notes, you will also identify the most important member of the team. This is the keynote. The keynote of C major is C, and its team is C, D, E, F, G, A, and B. The keynote of D major is D, and the full team is D, E, F sharp, G, A, B, and C sharp. Don’t worry; you don’t have to remember these rows of note names. I’m just putting them in here to show that the groups of notes—the keys—have a fixed identity.

In Western music we use two types of keys, minor and major, and our subconscious music analysis equipment doesn’t find it particularly difficult to work out which type we are listening to. As I said at the end of the last section, we associate minor keys with sadness and major keys with happiness—but why?

One of the main reasons is cultural. In northern Europe and the United States, most of the sad lyrics you hear are set in minor keys (e.g., the jazz song “Cry Me a River”) and happy lyrics tend to be presented in major keys (e.g., the Beatles’ “Here Comes the Sun”). So people who grow up in these societies expect these links. But as usual with music, it’s not a hard-and-fast rule even in these societies. It’s possible to write happy music in minor keys (like Purcell’s Round O in D Minor*) and sad music in major keys (like Leonard Cohen’s “Hallelujah”).

And as we saw earlier, some societies (Spain, the Balkan countries, and India, for example) use minor keys for joyful and sad music. So is there any objective reason why minor keys should sound sad?

Well, there are a couple of technical reasons why minor keys might be considered more suitable than major ones for expressing complex emotions like sadness and longing—but these shouldn’t be thought of as rigid laws; they are more like gentle persuaders.

The first reason why minor keys express sadder emotions so well is that they really are more complicated than major keys. To understand this idea, let’s take a quick look at how the strings on a harp are tuned to produce a major key.

A harp string produces a musical note by vibrating backwards and forwards at a certain frequency. A string might, for example, be traveling from left to right and back again 110 times every second, causing the pressure of the air near our eardrums to go up and down 110 times a second. This fluctuating pressure pushes our eardrums in and out at the same rate—so we hear a note of that frequency. (This is the frequency of the note A₂—the note produced by the A string of a guitar.)

If we make the other strings on the harp vibrate at frequencies that have a simple mathematical relationship to that first note, they will sound good together.* For example, a string vibrating 165 times a second will work well because 165 is one and a half times 110.

For a major scale, you start with a note of a certain frequency and then add notes whose frequencies equal the first note’s frequency multiplied by 1⅛, 1¼, 1⅓, 1½, 1⅔, and 1⅞. These simple relationships mean we have put together a strong, closely related “team” called a major key.

Let’s stick with the sports analogy for a minute. A minor key is basically a major key with three team members replaced by rookie players who don’t quite fit in. Our “1⅔” member, for example, is replaced by the relatively unfamiliar fraction “1.” To complicate matters further, minor key music allows some of the original major team members to substitute occasionally for these new, weaker players.

All in all our new team has less cohesion and is therefore more conducive to complex or somber musical moods than the more self-confident major team.

Music psychologists David Huron and Matthew Davies have unearthed another potential reason for our association between minor keys and melancholy. Their vast survey of minor and major key music revealed that, on average, minor key melodies involve smaller jumps between notes than major key tunes.³¹ When we are happy we tend to use relatively large jumps in pitch when we talk. The phrase “Hi, glad you could make it! How’re you doing?” might usually involve a fairly big jump down in pitch between “Hi” and “glad” and then a jump back up for “How’re you doing?” When we are unhappy, we use smaller jumps. The sentence “I’m sorry to hear about your mother. How’re you doing?” would involve only small jumps in pitch. We can often judge a person’s mood from the size of their pitch jumps, even when we can’t understand the language they’re speaking. So it seems reasonable that the smaller pitch jumps of minor key music could be one reason why we interpret it as sad.

Links between music and speech

A growing amount of evidence shows that there are similarities between the ways we perceive emotion in both speech and music.32 Even as young children we become very skilled at quickly picking up clues from the speed, loudness, pitch range, and timbre^* of sounds, and brain scans have shown that we use many of the same areas of the brain when we are judging the emotions being expressed by music and speech.

Language could only have been developed long after we evolved the vocal equipment required for speech, so there must have been a period of time when early humans used their voices, without words, to communicate emotions (anger, affection, etc.) and basic information (“Something dangerous is coming!” “There is food over here!” etc.). The linguist Dwight Bolinger has suggested that high and rising pitches would have been used to indicate interest (“Look at this!”) and incompleteness (“What’s going on?”), and, as languages were eventually developed, this resulted in an almost universal use of rising pitches for questions in modern human languages.³³ Similarly, the falling vocal pitches that we find at the end of statements in nearly every language indicate a reduction in interest and a feeling of finality.34

You might have thought that early humans developed speech first and then singing, but actually the two have probably always been intertwined. It is not too fanciful to imagine that early humans hummed and lah’d “songs” to themselves and to their babies long before languages were developed. It seems likely that these “songs” would naturally follow the characteristics of the wordless communication techniques outlined by Bolinger, with pitches that rise toward the middle of each (wordless) phrase, then fall toward the end—a pattern that is still followed by many songs to this day.

Most musical phrases follow an arch-shaped contour: they start on a certain note, then rise in pitch before descending again, often to the note they started on. There are two common types of arch, both of which are also used in speech patterns. When David Huron analyzed about six thousand European folk songs, he found a straightforward “up toward the middle—down toward the end” arch in about half of them.³⁵ And if you pay attention to the way you use your voice, you’ll notice that this gradual up-and-down curve is common to many of the statements you make. If you say, “I thought we’d go out for a meal tonight, but I don’t know where,” you’ll probably find your pitch rising up to the word “meal” and then falling away again. Another type of arch sometimes used in music and speech involves a steep jump upwards at the beginning of the phrase, followed by a series of downwards steps. This sort of contour is common in eastern European or Arabic laments and is similar to wails we produce when we are extremely distressed.³⁶

Folk songs, which are often centuries old, have given us another clue about the link between music and speech. In normal speech we raise and lower the pitch of our voice from one syllable to the next, but the range of pitches we use is obviously a lot narrower than the one we use for singing. You might therefore assume that there is no direct relationship between the size of the pitch jumps in melodies and those used in speech—but this isn’t strictly true, as neuroscientist Aniruddh Patel discovered.

Dr. Patel carried out an in-depth comparison of French and English folk songs to see if the language had an effect on the melodies. He identified rhythmic similarities between the songs and the language in each case—which is hardly surprising since the melody must fit the words, and you would therefore expect the rhythms of the language to influence the rhythm of the music. When he analyzed how the pitch went up and down between words in the two languages, he discovered that there were smaller jumps in pitch in spoken French than there are in English. This tied in with the fact that the French folk songs jumped around in pitch less than the English ones. So the language involving smaller pitch steps (French) also developed songs with smaller pitch steps.37

As we’ve seen, music and speech share a common set of emotional cues: big jumps in pitch, a fairly high degree of loudness, a fast tempo, and a pleasant timbre indicate happiness,³⁸ and more muted sounds, smaller pitch jumps, and a slow tempo indicate sadness. This is why, as I said earlier, we can often understand the emotional content of a conversation even if we can’t understand the words. We are very good at picking up emotional cues simply from how the voice is being used—and it probably all starts with those lullabies and play songs I mentioned earlier.

If you hear a mother and her six-month-old-baby “talking” to each other, it sounds as if the mum is imitating the sounds of the baby, but in fact it’s the other way round. The mother (and other adults) will start talking to the baby in what psychologists call “motherese” long before the infant is developed enough to respond with “ma-ma ga-ga.”³⁹ Motherese is different from normal speech in several ways. It is much more repetitive, has a more regular rhythm, is slower and higher pitched, and involves a lot of deliberate emphasis. Also, as music psychologist Diana Deutsch has noted, it often uses the up-down arch common to a lot of musical phrases (e.g., “Go-o-od girl!”).40 All of these features make motherese much more melodic than normal speech. When babies reach an age of about six months, they start trying to copy motherese by babbling. Babbling is very useful because it allows babies to practice using the seventy or so muscles they will need to control in order to demand expensive birthday presents later in life.

The sing-songy nature of babbling and motherese is common to a wide range of human societies, and it’s therefore not surprising that lullabies are similar all over the world. Nor is it surprising that, having had these enjoyable musical experiences as babies, most of us grow up with a natural affinity for music.

How does your brain turn music into emotions?

I said earlier that emotions are linked to survival—so why do we have emotional responses to something as apparently remote from survival as music? Well, your brain doesn’t treat music as a special case; it just treats it as audible information, a sequence of sounds to be rapidly processed like any other. Part of this processing often involves the generation of an emotional response. Music psychologist Patrik Juslin and his colleagues have suggested seven basic psychological mechanisms by which music produces emotions.⁴¹ They are:

Brain stem reflexes

When you hear a sudden noise—whether it’s the sound of a tree falling toward you or an unexpected saxophone wail in a jazz standard—a very basic part of your brain called the brain stem recognizes that something urgent is going on. Before you can say, “What the hell was that?” your brain stem has given you an adrenaline boost and you’re ready to deal with the situation: you are in an emotionally aroused state. Then other parts of your brain put a damper on things, effectively saying, “Calm down, you drama queen, it’s just music,” and you relax a bit. But you won’t completely calm down because you need time to recover from the original brain stem response. Also, luckily for you, your brain stem can’t be trained to respond more calmly next time, so the stimulation effect works again and again. This is the probable cause of our emotionally excited response to music that involves sudden, loud, and dissonant sounds, and fast or rapidly changing rhythms.

Rhythmic entrainment

Your pulse rate and breathing rate slow down if you are calm and speed up if you are emotionally excited. Under the right conditions your heart rate can rise or fall toward the beat of the music you are listening to, and this fools your brain into experiencing the emotion that is appropriate to your new heart rate. Fast dance music in a nightclub excites and energizes you for a couple of hours, slow blues in a dimly lit bar relaxes and calms you.

The same rhythmic entrainment effect can happen with your breathing rate—but because each breath takes a few seconds, your breathing synchronizes not with the beat of the song but with aspects of the music that last for several beats, such as complete musical phrases. The process of synchronization takes several minutes, and although your pulse or breathing rate might move toward the beat of the music, some beats are too slow or fast for you to match them.

Evaluative conditioning

If a piece of music is repeatedly linked to a pleasant activity such as watching your favorite TV show, then that music will make you happy even if it’s not accompanied by the show. You’ve conditioned (brainwashed) yourself into becoming cheerful whenever you hear that particular piece of music. This kind of conditioning works for other emotions as well. For example, even today, whenever I hear the music for the soccer program Match of the Day, I am filled with an almost uncontrollable urge to strangle my brother Richard. This is because as children, in those far distant days when each family had only one TV set, he and I had to take turns at ten o’clock on Saturday nights, alternating between our favorite shows. As a result I was prevented from watching the ravishing Diana Rigg in The Avengers every second week—and if you’ve seen Diana Rigg dressed in black leather, you’ll understand what a torment it was for a red-blooded Englishman in his early teens to have to watch a soccer match instead.

Emotional contagion

Emotional contagion involves identifying what emotion the music is trying to portray and then allowing yourself to be caught up in that emotion. You become infected by the emotion in the same way that you might become cheered up just by being near a bunch of cheerful people. As I mentioned earlier, emotional contagion happens far more often for cheerful or relaxing music than it does for unhappy or angry music.

Visual imagery

Though not everyone has this experience, music is pretty good at encouraging you to daydream in pictures. Many people conjure up images of landscapes and other visual images when they listen to music, which enhance their connection to the music and generate deeper emotional responses. In the 1960s, music therapist Helen Bonny developed a music therapy technique known as Guided Imagery and Music (GIM), in which patients are encouraged to experience (and talk about) visual images while music is playing. This kind of guided musical visualization can be very effective at reducing stress levels and alleviating depression.

Episodic memory

This is the “Darling, they’re playing our tune” effect. A piece of music triggers a specific memory, and you then return to the emotional state you were in when the memory was created—the joy of your wedding day, for example. Besides regenerating the original emotional state, this type of stimulus can also arouse emotions such as nostalgia and, in the most acute cases, attacks of revolting sentimentality.

Musical expectancy

As an experienced listener you build up expectations of what the music is likely to do next. Even two or three notes of a piece you’ve never heard before will prompt a subconscious expectation of what notes might come next. Whether your expectations are confirmed, delayed, or violated, you’ll have an emotional reaction, ranging from the smug glow of satisfaction when you get it right, to the thrills you experience if the composer or songwriter delivers something unexpected and beautiful.

So what has all this to do with survival?

Music is not necessary for survival, but emotional responses are, and music creates emotional responses. Let’s have a look at a few examples of how the seven psychological mechanisms just listed are useful in keeping you alive in non-musical contexts:

Brain stem reflexes make you jump out of the way when you need to.

Rhythmic entrainment helps you to do repetitive physical tasks more efficiently (no music is necessary—you can choose your own rhythm).

Evaluative conditioning helps you learn what’s good for you.

Emotional contagion can be thought of as a version of empathy: it helps you bond with people around you (e.g., mother and child).

Visual imagery allows us to “dry run” things we are thinking of doing to imagine how safe or unsafe they are: “If I climb that tree to steal some bird’s eggs, will that branch hold my weight?”

Episodic memory helps you figure out what sort of situations you find rewarding, threatening, tiring, etc.

Expectancy helps you to estimate what’s about to happen (in language and in life) and prepare accordingly. This is a crucial skill. Even casual chitchat involves you getting your response ready while your co-chatterer is still finishing her sentence; if we couldn’t do this, every conversation would be full of those long, pregnant pauses you get in brainy French films. Those of us who don’t live in French films have to forecast how the other person will end each comment, so we can chip in immediately and demonstrate how attentive and clever we are.

All of these mechanisms help you to live. They happen whether you want them to or not, and, by happy coincidence, they can be triggered by music. The mechanisms are independent of one another, so one or more of them can be activated by a certain piece of music, and each one leads toward different emotions, which explains why different people have different emotional responses to the same song. A dance tune might stimulate rhythmic entrainment and emotionally arouse one person, but trigger an episodic memory in others, leading them to feel nostalgic for their school days. The fact that several mechanisms can be activated at the same time could explain the mixed emotions we sometimes experience when we listen to music. For example, if we combine the emotions I just mentioned, we could end up with aroused, nostalgic dancing—whatever that looks like.

This research by Juslin and Co. is only a few years old and may have holes picked in it in the future, but for the moment it’s one of the few theories that tries to answer the question of how music generates emotions.

As you can see, the study of music and its effect on our emotions has come a long way since Deryck Cooke’s assertion that minor keys are sad, unless you’re Bulgarian… or Spanish.