What conductors are doing when they wave their hands around -- and what we get out of it

As a child (and like most children, I imagine) I used to think conducting an orchestra entailed something like what Bugs Bunny does in this video:

ResearchBlogging.orgWaving the hands, as conductors frequently do, seemed largely for show. The conductor appeared to me to be more dancing along with the music than actually leading the musicians in any meaningful way. It wasn't until I married an amateur musician that I actually learned that the conductor could have an important influence on the way an orchestra sounds. But as Greta and I moved from place to place and she joined a variety of different ensembles, I got to hear her explain why some conductors were better than others.

One of the most basic functions of the conductor, I learned, was simply keeping the whole orchestra coordinated, so that all the different musicians started and stopped playing at the right times. One key to this was to express the beat of the music through the movement of the conductor's baton (or hands). Musicians could then count the number of beats when they were silent (which might number in the hundreds), and begin playing at the right time. Different songs have different rhythms, and so the conductor uses different movements to express the beat depending on the rhythm. This chart shows some of the simpler patterns a conductor might use:


Good conductors, Greta told me, use something pretty close to these patterns, so it's easy for musicians to follow and start playing at the right time. Bad conductors (often described in more colorful terms than this) are more like Bugs Bunny in the video -- so concerned with the "emotions" of the piece that they neglect the basics.

But what exactly are musicians paying attention to as they try to follow a conductor? What is it about the conductor's motion that makes it easier or more difficult to follow? As you can see from even this simple diagram (or this series of videos), even the basic rhythms aren't always easy to follow. To make things even more confusing, many conductor's manuals suggest that the beat is indicated when the conductor's motion changes direction, while the hand is moving at the highest speed. But simple physics would suggest that a direction-change is accompanied by low speed. The only way to reconcile these two notions is if a conductor's motion actually traces a wide loop, so that it doesn't have to change speed as it changes direction.

Geoff Luck and John Sloboda taped a conductor performing the simplest pattern -- the one-beat -- with a tracking device on his finger (this conductor did not use a baton). They asked him to conduct in three different ways, with progressively wider-radius loops at the bottom of the movement where the beat is supposed to be indicated.

They showed a point-light representation of these movements to volunteers (some musicians, some not), who were supposed to press a button in time with the beat indicated by the conductor. The viewers had some time to practice before the real study began, but they weren't given a lot of instruction on where the beat was supposed to occur in the conductor's movement.

So what characteristics of the conductor's movement matched up with the beats indicated by the viewers? Perhaps what's most interesting is what didn't match up: the radius of curvature of the movement. The width of the arc of the conductor's movement had no impact on the perceived beats. What did have an impact was the velocity and acceleration along the motion trajectory of the conductor's hand: both were significantly correlated with the viewer's perception of a beat. Viewers tended to indicate a beat when the conductor's velocity was high, but acceleration was negative (i.e. slowing down).

So if velocity and acceleration are all that matters, then perhaps simply showing viewers this motion, without any curves, will also produce the same beat. In a second experiment, Luck and Sloboda did just that: the conductor's velocity and acceleration were replicated along a horizontal line, as if the curve of the hand movement was simply stretched out across the computer display. Now were viewers able to pick up the beat indicated by the conductor? Yes, to a certain extent. Velocity and acceleration were both associated with the beats perceived by views. However, viewers did tend to feel that two beats were produced with the conductor's motions, instead of the intended single beat.

Finally, the conductor's motions were shown along the original paths, but with constant speed and acceleration. Once again, the radius of the conductor's motions had no impact on the perceived beat. So it appears that the velocity and acceleration of the conductor's motions are the key factors viewers use to decide when the beat should be.

The researchers say that acceleration is the more important of the two factors -- that beats are indicated when the conductor is slowing down or speeding up. But they also point out that other aspects of motion were not studied, and that in other, more complex conducting motions, different factors might be involved.

Geoff Luck, John Sloboda (2008). Exploring the Spatio-Temporal Properties of Simple Conducting Gestures using a Synchronization Task Music Perception, 25 (3), 225-239 DOI: 10.1525/mp.2008.25.3.225


More like this

This is a guest post by David Kerns, one of Greta's top student writers for Spring 2007. As movie special effects technology improves, more and more live-action shots are being replaced with computer animation. Harry Potter flies across the Quidditch field; Spider-Man swings from web to web…
Listen to this short recording: It's a sequence that repeats every sixth beat. But when we're listening to music, we usually prefer to divide rhythm into two- or three-beat patterns (duple or triple rhythm). In this case, the sequence doesn't make it obvious which pattern is correct. A traditional…
Take a look at this video of a professional drummer playing the conga: It's easy to see that the sound coming from the drum is perfectly synchronized with the motion of the drummer's hands. Or is it? When a sound enters your ear, it takes less than 1 millisecond for the signal to be transported…
Point-light displays can tell us an amazing amount about other people. Looking only at a few glowing spots corresponding to joints and set in motion, we can tell what people are doing, whether they are over- or underweight, and even identify a friend among strangers. We can also identify animals or…

As a former professional musician, I can say with a fair degree of certainty that it is actually positional. A beat, in my experience, is indicated at the lowest point of any given path of the conductor's baton. So I guess you could say the beat lies where the vertical component of velocity is 0. At least, this is where the "perceived" beat is, but then you must factor in a degree of anticipation on the part of the musician.

But, to make matters more confusing, orchestral musicians generally don't rely on the conductor for beat. In fact, in my experience, it is the mark of a novice conductor and/or a poor orchestra if the conductor is visibly "marking time", as it were. Rather, it is the manner in which that beat is expressed: Rounder, more flowing conducting indicates a more legato style, while sharp, defined conducting indicates staccato, for example. Considerations of volume, balance, tuning, and style fall under a conductor's purview, and patterns indicating beat are not so much an indication of where the beat actually is, but rather what part of a given measure the conductor is communicating to the orchestra.

To illustrate my point, I'll close by telling you that something funny happens when you watch professional orchestras with very good conductors. You won't notice it at first, but before long you realize that the orchestra is often playing several seconds behind the conductor. It's not that they're a bunch of musicians with strangely coordinated horrible timing, but rather that they don't rely on the conductor for beat. Chances are, they're familiar enough with the material (or they damned well ought to be) to fend for themselves, more or less. If there are timing problems, then the conductor will deal with it, but generally not until then.

Oh, and I should mention this applies primarily to performance conducting. Certainly as an orchestra is just beginning to tackle a piece of music, the conductor adopts a much different role than during performance, when they've heard it a million times and know a piece cold.

I spent many years playing in school and amateur orchestras. A large chunk of the conductor's job comes during the hours and multiple times that a piece is rehearsed. Although the musical score will contain hundreds of suggestions regarding the dynamics and shape of the notes, there can be huge variations in what is forte, or pizzicato. On top of that, acoustic dynamics are different from harp, to horns, to strings. And if you are a violist surrounded by cellos and violins, you have no idea how you sound compared to the other sections.

So a conductor might say something during rehearsal like, "the English Horn is important here, can you give me a bit more volume? and the strings need to be more in the background." or "we need a sharper contrast here." The hand gestures not only keep time, but often they are visual reminders of stylistic decisions that were discussed verbally during rehearsal.

To use an olympic analogy, coaches sit or lurk on the sidelines in most team sports. But the actual competition is preceded by hours of technical practice and discussion.

On good and bad conductors: if a group of performers needs someone to hold them together by being a "human metronome" and giving them the beat, a good conductor gives it to them. A bad conductor defies the group and sacrifices the beat to convey emotion. This always makes for a disaster.

However, the better a group of performers is, the less they need this kind of visual metronome. This requires both experienced and talented musicians and group that has played together for a while. At this point, a good conductor will begin to anticipate the orchestra more and more. At this point the conductor is communicating tempo (not quite the same as beat), style, volume, emotion, etc.

This is an important step for a great orchestra to achieve some of the spontaneous feeling and emotion of a performance that takes it to the next level. Anticipation helps the orchestra know what to do in advance of playing the notes, and frees the conductor to do something a little more spontaneously and freely in the performance, rather than reproducing rehearsal by rote.

A bad conductor will stay a human metronome.

In short, a good conductor helps the orchestra sound better and, if the ensemble is sufficiently skilled, brings the music to life for the audience and performers. A bad conductor fails to help the orchestra realize its potential.

Perhaps the fact that people notice the deceleration the most is because people imagine the conductor striking and hitting an imaginary object (which could be any kind of striking instrument, like a drum).

Then again, this study needs to replicated on real musicians, with real conductors, so that it is more applicable. Maybe amateurs just don't know how to keep beat with the conductor properly.

Keep in mind that it's not just the hours of rehearsal that tightens an orchestra to a conductor, it's also the *years* of experience of watching conductors (and having your own performance judged by them for your grades) that makes us aware of every nuance, every subtle motion than the "amateur" would not see.

It's akin to how a MLB baseball outfielder that knows where to move *on the pitch* or at least on the sound of the hit, and knowing exactly where it's going to go within inches to make the catch.

Such conditioning is part of the musician's training for years.

While we're at it, accelerando is actually extremely rare in classical music. ritardandos are more common, but coming out of it is usually back to speed on the next beat "a tempo". Now, "Romantic" conductors specializing in 19th century music can play with the tempo a lot more than the amateur listener might notice (Bernstein, Stokowski, and Karajan are most famous for this) - again, very subtle changes to the amateur mean subtle changes to the performers, but leading to great changes in the listener's experience.

To me, there's little in the way of mental magic worth studying when it comes to conducting - it's a language (a body language more than a verbal one) that develops as specialized to musicians as the specialized language of psychology and neurology (and statistics) is to you lot.

By Joe Shelby (not verified) on 21 Aug 2008 #permalink

BTW, Bugs in Baton Bunny was specifically making fun of Leopold Stokowski, in Warner's 3rd parody of the famed Disney Fantasia conductor. It wouldn't be the last, as Elmer would do it again in the opening of What's Opera Doc. :)

BTW, there's a good podcast, "Do We Need Conductors", from NACOcast featuring an interview with Britain's Oliver Knussen. http://radio.nac-cna.ca/podcast/NACOcast/NACOcast_20070313.mp3

By Joe Shelby (not verified) on 21 Aug 2008 #permalink

Being an amateur musician (whose father is a professional musician), one of the things we both talk about often is the clarity of a conductors beat, and one of the ways a conductor makes his beat clear is the "striking of a drum" analogy Frieddie points out. A good conductor is able to quote-on-quote "manipulate" the orchestra into playing in certain styles by changing how he metaphorically "strikes the drum". A conductor once explained to me that if he wants the musicians to play smoothly, he'll move the baton smoothly. Although one might think that the orchestra, by the time of the performance, would have gone through the pieces so many times that they would know exactly how the conductor wants them played. In actuality, most orchestras (at least in the USA) only have one or two rehearsals--they need to rely on the conductor's physical movements (to a point) to know what style to play in. The only parts a (good) conductor will work on in rehearsal (besides running through the whole piece) are the sections in which he feels the orchestra isn't responding to his "conduction".

Also, conductors don't just rely on how they "set the beat". When you see a conductor moving his entire body, he isn't (necessarily) just swinging with the emotions. For example, one way to get the orchestra to play louder is to lean back as if beckoning more sound.

By Avi Steiner (not verified) on 21 Aug 2008 #permalink

I used to play in a semiprofessional community band, and one exercise that our director had us do was to conduct the band using only his facial expressions - basically, directing with his hands behind his back. It was comical AND informative. We didn't really need him to give us the tempo, but it was pretty universally understood what kind of playing "happy face" and "sad face" were meant to convey.

Another exercise we did often was playing warmups with no conducting at all beyond an initial count off. Again this was not difficult because we didn't need a "human metronome" (a phrase I like btw), and warmups typically do not have tempo changes or extremely expressive parts.

There are two main schools of conducting: American and European. The American style is as you describe, with the beat defined at the point of change, called the ictus. The European style defines the beat before the ictus, during the downswing. This is to make the sound coordinate with the visual for the audience. Since light travels so much faster than sound, if the musicians play at the same time as the ictus, then to the audience they sound late (the effect mentioned by Brian). If the musicians define the beat before the ictus, then to the audience it seems lined up. I hate playing for a European-style conductor.

One thing I realised in my years as a DJ is that the speed of sound is actually quite low, and seemingly minor distances in human terms will have a major effect in musical timing terms. for every meter a monitor speaker was away from my head, a 3 ms delay is introduced that must be compensated for when the signal in your headphones that you are trying to align the beat with is not delayed at all.

assuming a few approximate distances here...If the conductor were replaced with an audio click, rather than a visual cue of the beat, then the players at the front would hear that click around 16 ms after the players at the front, and the error would be compounded going back out to the audience, with the audience hearing the players at the back 16 ms after the players at the front.

This is why large audiences cannot keep a ryhthmic clap going easily.

16 ms is right at the limit of musical timing perception - under 10 ms and you only notice on very tight percussive sounds, under 25ms for softer instruments, so 16ms (about 5 meters) is somewhere in middle ground.

When a conductor gives the orchestra a visual time cue, all the orchestras notes hit the conductor with just the single time delay, not the compounded double, and they never then drift because of that delay. The sound from the front of nearest instruments will hit before the ones at the back, but the audience will never hear any timing discrepancy larger than this, and the error does never gets compounded and is always consistent, not changing by following an audio cue from instruments with differing places.

"Since light travels so much faster than sound, if the musicians play at the same time as the ictus, then to the audience they sound late (the effect mentioned by Brian).

I don't think that's what Brian was referring to. In my experience (as a player), some orchestras simply play a particular amount of time after the beat on purpose, and some conductors conduct in such a way that it is impossible for the orchestra to be precisely on the beat (but not impossible to be together at some point after the beat). Some conductors change how much in front of the orchestra they are based on what sort of music they're conducting. For example, I once played the overture to Cosi fan Tutte (a very light, fast piece) with a conductor who was close to half a bar ahead of the orchestra at all times - it gave the music a peculiar and very effective feeling of urgency as the musicians kept trying to catch up.

By konrad_arflane (not verified) on 23 Aug 2008 #permalink

What Brian said. It's not the conductor's job to express the beat or control every aspect of the orchestra. The musicians have the sheet music, or the song memorized; they know the notes they're supposed to play and have a general sense of when to play them. What the conductor does is keep the orchestra coordinated and on task. He slows down the people who play too fast, speeds up the people who play too slow, and gives a visible marker to the rate of progression through the music, which helps to cut down on instances of people reading ahead.

It was a trademark of a lot of European orchestras vs American that the Europeans aimed to play half a beat behind the conductor. Then the conductor can cue things on the beat but the orchestra can still respond.

As several people have mentioned, with a decent group, once you've started together, the orchestra will generally hold together. This probably isn't true for, say, Mahler's 8th symphony for purely physical reasons, but Orpheus does just fine, and the vast majority of early music ensembles don't have a conductor at all, just a web of rules about who leads timings (harpsichord or first bass string instrument unless the first violin is doing something, in which case first violin, unless someone has a neighboring or passing tone off the beat leading into a cadence, in which case that person, unless the cadence is elided, in which case...).

Hi. Professional conductor here.

(Just to clarify, yes, readings are often conducted differently than concerts.)

The music determines the conducting techniques employed.

Perception of the ictus (point of "beat"; infinitesimally short duration) is up to the performers. The conductor employs techniques to convey this.

Of course a change in direction of stroke implies a "dead stop." Clear conducting motion is, paradoxically, nonlinear. Acceleration is essential to avoid losing time during the "dead stop."

Perception is based on changes to otherwise static phenomena.

Common interpretive techniques include compression and expansion of beating time. The ensemble recognizes the Gestalt and reacts appropriately.

Ictus is either pronounced or effaced. Again, this is purely a result musical determination.

I find this an inappropriate forum for droning about the role of the conductor, because, again, it depends on the music...

Remember that performers are conditioned by the affections of a piece's interpretive history. Contemporary music lacks, by nature, a long performance history, and thus often calls for different conducting techniques, to aid the performance of the music.

Finally, ensemble size significantly affects the size, altitude, ictus, and range of the beat pattern. (N.B. The choice to use a baton is very physically significant and is determined musically.)

Hope this clarifies.


"Viewers tended to indicate a beat when the conductor's velocity was high, but acceleration was negative (i.e. slowing down)."

Well, that's what happens when you strike a drum. The sound happens when the stick stops (decelerates sharply). Drumming is rather interesting in that a person can instinctively factor-in the time it takes to get the stick from A to B and hit the drum right on the beat. Some people lack the ability - can't even clap in time.

By Paul Murray (not verified) on 23 Aug 2008 #permalink

This article is answering the wrong question. Conductors don't just show up at the concert and start hand-waving. The vast majority of their influence is during rehearsals, where they adjust all kinds of balance and textures and colors, and sometimes even inspire. There are some good conductors who really aren't doing anything useful on the stage, but have still greatly and uniquely shaped the music.

By Epistaxis (not verified) on 24 Aug 2008 #permalink

Hello. I play in a community orchestra, and our orchestra is not very advanced. The conductor gives every beat because most of the musicians don't have much experience in playing and doesn't pay much attention to listening to each other. Therefore that little delay doesn't really happen because we basically follow the conductor all the time, or should I say that the conductor emphasis on staying exactly together. I always wondered about the delay too because I conducted bands before and if that happened while I was conducting, I'm sure I felt pretty uncomfortable. May be it was because wind ensemble, not an orchestra. I have never seen a wind ensemble having that little delay. My teacher who played in many professional orchestras explained to me that it happens mainly because of the string sections. Their instruments don't speak as fast as wind and brass instruments; string instruments increase in sound as they move the bow, so the very first moment when they are starting to sound the note is weaker, where the other instruments would have the clear and articulated sound from the first attack of a note. And this is why all the other brass, wood wind and even percussion instrument just listen like crazy. She said that the first day she sat in a real professional rehearsal was a shock. Being a flutist, it was very hard to understand at first, but she said all she needed to do was to listen to others and feels the music together. She thinks the conductor is there to help them sound more musical and to unify the emotion of the piece. So this is what I thought it was... or am I wrong...?

By University mus… (not verified) on 25 Aug 2008 #permalink