What it means to have Absolute Pitch

[wrog]

a healthy respect for the challenges of playing on a transposing instrument.

So I was originally going to answer this in the comments but this is really a huge topic worth its own post. It's always been a bit weird what people think should be difficult vs. what actually is difficult.

See, playing a transposing instrument isn't supposed to be a challenge. In fact, I'm sure the whole reason transposing instruments exist was to make it easier, and for folks who only have relative pitch sense, which is most of the population, I'm pretty sure that's the case.

That is, you see a C on the page, C has a particular fingering ('O'/open, if we're talking about a typical valved brass instrument) and you're done. You blow, and you'll know whether you have the right note if it's the correct interval away from whatever pitch you're using as a reference, whether it's a previous note that you played, or that somebody else played, or there's this huge chain of inference leading back to the start of the session when whoever it was blew on the pitch pipe. If, say, your reference happens to be denoted as B in your part, and you're now being called upon to play a C, then you need to be up a half step, and you'll be able to hear that. And that's really all you need.

The question of what the actual concert pitch might be is quite irrelevant. That you might be playing on a B♭ trumpet so that a denoted C comes out as a (concert) B♭, or a D trumpet where the same written note comes out as a (concert) D, or that the person who's playing your reference tone thinks of it as some note that's completely different from what you're thinking it is, you don't have to care.

This way the individual instrumentalists just get to focus on playing the notes that are on the page in front of them. Making sure that the corresponding sounds will match up with what everybody else is doing is the composer's problem, not yours.

Now, I'm sure that, for the conductor, this is a huge fucking mess, but dealing with that is part of why they get paid the big bucks (hahahaha).

At this point, I'll just note the weirdness of having to reverse-engineer how relative-pitch-only people think of things. Which is the best I can do because I have no personal experience of this.

I can only tell you how it works for me. I can't even be sure how much my experience generalizes to how other absolute pitch people experience things, because this only shows up in 1 out of 30,000 people on average, and I still have yet to encounter anyone else who has it.

(except possibly for my son Philip, who I have reason to believe has it, or, rather, there are certain things that he's done that would otherwise be very difficult to explain if he didn't have it, but because he's autistic and thus doesn't have the language to talk about it, I'll probably never know for sure).

But first we need to straighten out some misconceptions:

What is Absolute Pitch?

• Relative pitch is the ability to recognize intervals between pitches. If relative pitch is all you have, then you can't actually recognize the pitches themselves unless somebody provides a reference pitch and tells you what it is, at which point you can then use the interval between the reference pitch and the pitch you don't know to figure out what it actually is.
  
  
• Absolute pitch is the ability to know what a pitch is in absolute terms without needing anyone to provide a reference pitch.
  
  
• What this really means in practice:
  
  There exists a roughly half-step range of frequencies, say from around 445 to 468, each of which, to me, sounds like a B♭, i.e., everything in that range has this certain "B♭" quality to it that I don't really know how to describe other than that I know it when I hear it.
  
  And there are other ranges corresponding to each of the other 11 notes in the chromatic scale (and likewise for all of the other octaves within the range of hearing). I suppose if I'd grown up in a non-Western culture that prefered 5 or 7 or 19 tone scales, I'd have some different number of pitch categories, but let's not worry about that just yet)
  
  
• We could try the color-vision analogy, i.e., the whole question of how you explain what "red" is to someone who's color-blind (you really can't).
  
  But that doesn't quite capture what's going on here, unless you can imagine this weirdly specific version of colorblindness where people can see relative color, i.e.,
  • they actually will be able to recognize red once you give them, say, a reference sheet that's painted green and you tell them that the color of the reference sheet is indeed green, while at the same time
  • if you fail to give them that reference sheet they will have no idea what red actually is, and, moreover,
  • all shades will still be distinguishable, and
  • they can treat any shade they want as being red, at which point all other shades will be ordered accordingly (e.g., if the color they're treating as red is actually yellow, then they'll see green as yellow, light blue as green, and so on into the blue end of the spectrum)
  Which is not at all how real color-blindness works, or least I've never heard of anybody being this kind of color-blind. The typical thing is for someone to be say red-green blind, meaning their receptor for distinguishing red from green fails, so those two colors just look exactly the same to them, and no reference sheet painted red, green, or any other color will be of any help whatsoever. The auditory analogy would be someone being entirely tone-deaf, or maybe having enough pitch sense to distinguish high from low but that's all.
  
  
• People really just don't get how insanely weird it is that one can have accurate relative-pitch sense by itself and at the same time no sense of absolutes.
  
  
• Another huge misconception about absolute pitch is there's nothing exact about it. This is, by the way, why I severely dislike calling it "perfect pitch", because it's just not.
  
  In fact, if someone were to tune a piano to A-435 instead of A-440, I simply wouldn't even notice. As long as the note is somewhere in the right range, I'll hear it as an "A" and won't think twice about it.
  
  If somebody is singing a capella and going slightly flat or sharp as time progresses, I won't notice that either, at least not until/unless they get sufficiently far off that they're hitting one of my half-step boundaries, at which point I'll be in a something-is-wrong-with-this-but-I-can't-quite-put-my-finger-on-it mode and then maybe go look at the score and realize, "Hey, you're singing the wrong notes," as opposed to "Hey, you're out of tune."
  
  To be sure, if two people are singing together and going out of tune with respect to each other, that'll be obvious, but it'll be the same sense of something-screwed up that a relative-pitch person will also hear right away.
  
  In other words, absolute pitch is really about coarse pitch sense, not fine pitch sense. It's like having a GPS that tells you what city block you're in. For millimeter accuracy, you need other skills/training as well — and in musical terms, there are lots of people out there that are better at the millimeter accuracy than I am.
  
  
• For extra fun, my half-step ranges are not actually centered where you'd expect. 440 is actually at or near the top end of the range of pitches that I hear as A, the bottom being somewhere around 415. My theory on this has to do with the piano in the house where I grew up being tuned to a lower A (400? 415? don't know) up until the time I was 7, when we had it restrung so that it could be tuned to the modern 440 standard.
  
  Which means I can deal with stuff being significantly flat in terms of the modern standard, but not with stuff being even slightly sharp. And the latter is indeed occasionally a problem for me with recordings made by modern orchestras that like to tune themselves somewhat higher than A-440.
  
  And yes, playing Beethoven's 5th Symphony in D♭ minor does actually sound weird to me.
  
  
• The ranges also overlap slightly. Which makes the boundaries into weird places where I can convince myself either way. That is, if there's a note sounding like a B♭ and I am able to successfully convince myself that it's an A, then I'll start hearing it as an A and everything else will flip accordingly (i.e., that D♭ minor 5th Symphony will suddenly switch to being in C minor after all).
  
  This is an effect I can usually rely on to know that a given pitch actually is at or near the boundary.
  
  Which is my one way of doing fine-pitch tests, i.e., if I'm trying to figure out where 440 is, I can take a pitch that I hear as A and keep raising it gradually higher until I get to something that I can't hear as an A anymore no matter how hard I try. That locates the boundary and will typically be a few cycles above 440. But it's not something that can happen on the fly, i.e., I can't just hear that a particular A is not actually 440; I have to consciously work at it to figure out where it actually is.
  
  I also wouldn't be at all surprised if the boundaries moved around a bit (I'm told that the boundaries typically change when you get older but I have yet to experience this) but I haven't really done that experiment.
  
  
• There's also the whole range of questions of where the ability comes from, whether it's innate or something you can learn, whether it involves hereditary/genetic factors or it's purely environmental/developmental.
  
  For a long time I was convinced it was something you could learn; after all, I did. Just take a C major scale and memorize it; what's so hard about this? But the evidence is pretty much against me at this point.
  
  I'm also inclined to think that if there's a hereditary component to pitch sense it's only a small piece of the puzzle. My manager at Microsoft was one of a pair of identical twins. They both liked to whistle, one of them was pretty good at holding a tune, the other was, well,… not. Whenever one of them was within earshot I always knew which one it was.
  
  There's also clearly a huge cultural component. There's a theory that everyone starts out with absolute pitch but you actually have to lose the ability in order to develop language comprehension (i.e., if you're distinguishing too many sounds you can't grok the commonalities), which may be why there's a correlation with autism, unless the language in question is like Chinese in having a significant pitch component. I am also constantly struck by how Americans almost seem to take pride in how out-of-tune they can be (cf. pretty much every performance of "Happy Birthday" in the local restaurants) whereas you can walk along a street in Wales and know pretty much instantly that you're in a different country just from the way people are whistling. I could go on...

So, anyway, this is my mutant superpower. I can also roll my tongue, as it happens.

Not that there aren't situations where absolute pitch can be quite useful -- and I have any number of fantasy plots where there's, say, a nuclear bomb needing to be disarmed, and, oh no, it's emitting an ominous tone, and our oscilloscope is broken, but, yay, Absolute Pitch Guy is here, and he says the tone is a C#, which means we can cut the blue wire, and yay, the city is saved -- but these situations tend to come up way less often than you'd think, probably because our musical culture is designed around relative pitch, because that's the ability most people actually have.

… and, for better or for worse, it has some rather pervasive effects on how I do things musically (surprise)

What Happens With Me and Transposing Instruments

As you can probably imagine, what you get when you bring transposing instruments and absolute pitch together is basically a complete shitshow.

First day of French horn lessons in 5th grade: Let's play middle C (I play a C). No, go higher (I go up an octave). No, lower (WTF?). Just don't go all the way down (I try a note in the middle). There, you got it (Umm, that's an F). No it's a C (No, it's definitely an F). Well okay, it's a concert F, but … oh sh--, you can actually hear that? (and then he goes over to the piano, which is facing away from me) What's this note? (plays a random note, I name it correctly, wash, rinse, repeat, … a parlor trick that I end up repeating many times over the next n years

… which I should point out is not actually a real test of anything. That is, once I've already had a note verified for me, which in this case had already happened once I'd played the note on the horn that he verified was a concert F, the rest of the pitches that followed could be easily be determined via relative pitch sense alone.

in fact, testing this sort of thing is really rather difficult if you believe in pitch memory, i.e., if you can just remember and keep in your head what a C sounds like for days or months on end, maybe that's all this is.

the best I could do was the time I went backpacking for 3 weeks in New Mexico, endeavored not to listen to broadcast or recorded media that whole time (which was fairly easy when we were on the trail and all I had was the group of friends were constantly yelling out lyrics to Bohemian Rhapsody -- yes this was 1976 -- but that was okay because they were essentially tone-deaf), came home, headed straight for the piano, sung a G, and then checked it … which either verified that I have absolute pitch or that my pitch memory lasts at least 3 weeks. Hard to say if there's much of a practical difference there.).

The problem with transposing instruments, in a nutshell, is that if a given pitch is in the right range, then, for me, it's a B♭ and there's no way for me to make it be anything else. And if it so happens that I'm dealing with horn parts that insist on denoting that particular pitch as F, then I just have to find a way to cope with that.

The only real way out for me is to read it as a different clef -- which I didn't consciously figure out until I got to college, but it is effectively what I was doing the whole time with horn parts from 5th grade on.

Update:  If the next few paragraphs are overly opaque, you may prefer the version with the diagrams

To wit: Treble clef means the top line of the staff and the first space from the bottom both denote F. And what I needed to do was find a way to read both of those as being (concert) B♭s. Which means reading the next whole step up, whether this be the space immediately above the staff or the second line from the bottom, as (concert) Cs.

Quick review of how clefs work:

• ledger line below the staff is middle C ⟶ treble clef
• bottom line of the staff is middle C ⟶ soprano clef
• 2nd line from bottom of the staff is middle C ⟶ mezzo-soprano clef
• middle line of the staff is middle C ⟶ alto clef
• 2nd line from top of the staff is middle C ⟶ tenor clef
• top of the staff is middle C ⟶ baritone clef
• ledger line above the staff is middle C ⟶ bass clef

Meaning if you want the second line from the bottom to be a C, then that's mezzo-soprano clef.

So F-transposed treble clef horn music, can also be read as concert pitch music written in mezzo-soprano clef with an implicit extra flat (i.e., what you need to make the bottom space or the top line of the staff be a B♭ rather than a B). Likewise, B♭-transposed (trumpet/etc) music is actually concert pitch music in tenor clef with two implicit flats. And E♭ (alto-sax/etc) music is likewise bass clef with three implicit flats. And likewise for other transpositions but those are the three that matter if you're playing in a band.

What's interesting here is the way this completely divorces

• learning to play the instrument (how to produce particular sounds) from
• learning to read the music (determining which sounds the composer is calling for).
  
  

That is, when I play an instrument, the goal is always to produce particular sounds. And thus what you need to learn is which fingerings produce those sounds, e.g., 1st+3rd valve on a B♭ horn produces a particular sound — and you can call it whatever you want ("D" or "G" or "concert C") and I eventually succeeded in not losing much sleep over this — but the sound itself is what matters, and I'd just skip straight to that without bothering about what to call it.

Correspondingly, reading a part means seeing a note on a particular line of the staff and knowing what sound that's supposed to be. As long as the part is written in a clef that I know how to read, I'm done.

The consequence of this is that I can play any instrument that I know how to play off of any part that I know how to read. Playing a descant recorder off of a trumpet part, playing piano off of an F horn part, playing a B♭ trumpet off of a piano part, it doesn't matter, they're all equally doable.

The only actual difficulties are

1. whatever difficulties are inherent in the instrument itself, e.g., my recorder technique sucks because I only spent a year on it back in 1st grade, I'm much better at doing stuff on a keyboard, and
2. my familiarity with the clef in question, e.g., treble and bass clef I know the best because I started piano lessons when I was 7; mezzo-soprano clef I've been dealing with since I started horn in 5th grade. The other possibilities, not so much.

It's always amusing for me when people get astounded at the wrong things. They see me play a horn off of a flute part and they're all, "Oh wow, you can transpose on the fly." And I'm like, "No, there's no transposition involved at all, here. I'm reading and playing everything in concert pitch." To me, transposing means actually moving something to a different key, which I can occasionally do on the fly if it's something sufficiently simple.

… unless you want to claim that clef-reading is transposing and thereforee I'm always transposing, like when I play an F horn off of an F horn part like I had to do every single day in band practice, but, if so, that kind of renders the word meaningless.

In fact, one of the harder things for me to do was playing B♭ trumpet off of a B♭ trumpet part, mainly because tenor clef is the one clef I never really had occasion to learn, even though I had B♭ trumpet fingering technique up the wazoo (because of all of those years I spent playing a B♭ horn -- modulo the the Small Matter that, as of today, I currently have zero embouchure (lip ability) due to not having touched a brass instrument in decades)

We would in fact occasionally have E♭ horn parts to play. For me these would be a breeze because I knew bass clef from my piano lessons. I'd tell the section, "Hey everybody, bass clef, add three flats, no problem," at which point they'd all look at me like I was from Mars and then proceed to pull out their tuning slides a whole step (effectively turning their F horns into actual E♭ horns) and then playing off the parts as written. And I'd be, "WTF Is wrong with you people?"

Because this was, for me, the hardest thing of all, the downside to doing things the way I do, leading to the one thing I could never actually do: Play an instrument (E♭ horn) that I had never actually learned how to play.

See, when you have fingerings associated with sounds rather than notes, a transposed instrument is a completely different instrument; the technique I learned for B♭ horn was completely useless on an actual F horn or an actual E♭ horn. It was actually worse because I did learn F horn technique at one point but had to forget it in order to learn B♭ horn technique. There's only space in my brain to learn technique for one set of fingerings.

(I suspect the same will be true if I ever try to learn alto recorder, i.e., I'll have to forget the descant recorder fingerings and then I won't be able to play that anymore; for me they're just completely different instruments.)

(Similarly, if I ever learn guitar, it will most likely be impossible for me to use a capo. You get the idea.)

The ultimate nightmare gig was the time I had to play on a B♭ piano. It was New Year's Eve party at a nursing home. They had a piano but the tuner evidently decided it couldn't be tuned up to a modern standard without breaking all of the strings, so he turned it a whole step down: A-392 effectively. If he'd tuned it a half-step down I could have managed, but A-392 was just too low; I could only hear it as a G, which made the C a B♭ and hence this really was a B♭-transposing piano,

… an instrument I did not know how to play, even though by that point, I'd had close to 10 years of experience on the ordinary C-transposing/concert-pitch piano.

What saved us was that the other players in the group were a drummer, a (B♭) trumpet player, a (B♭) tenor sax player, and everything we did was out of the fake book, with everyone else playing parts as written (because they didn't know how to transpose) and me transposing down on the fly (actual transposition, because you can't clef-read a fake book, since it actually says, e.g., "C MAJ 7" and there's nothing to do but translate that to "B♭ MAJ 7").

Which meant they could just keep doing what they were doing and all I had to do was play the chords as written on the transposing instrument and not listen to anything I was playing. E.g., if the page said "C MAJ 7, I'd need to form the finger pattern for a C major 7 chord (C-E-G-B), play it, and then somehow ignore that the actual sound coming out of the piano was a B♭ MAJ 7 chord.

I had to play completely visually.

It was a fucking nightmare. Luckily the alcohol started flowing early and often.

Comments

[annathepiper]

As the commenter whose comment you're responding to here, let me just say: wow.

Yeah, your level of described challenge here is even higher than my personal experience with it. :O

I don't know if this is a matter of me not being a horn player, or my not having absolute pitch or what... the closest I have come to this is that I have flutes in different keys. But my default set of fingerings is either "concert C" or "D" depending on what flute I'm actually playing (keyed flute vs. default keyless flute).

But if I jump over to a G or A flute the fingerings are of course completely different, and I still haven't managed to train myself to account for that properly yet.

[wrog]

But if I jump over to a G or A flute the fingerings are of course completely different

ok, that's interesting. Is this a case of

• the instrument is physically larger/smaller and so the various special keys (the enhancements they add to the base fingerings to make various things easier) have to be different and you actually depend on those for the C/D flute?
• the sounds produced are higher/lower than you're expecting and this is throwing you off? (I'll note I'm still somewhat incredulous at the idea that people don't have any idea re absolute pitch; I keep thinking there ought to be intermediate stages where, say, you know where you are to within an octave or a fifth so that if an instrument is transposed sufficiently far away, it's going to be hard to play — but if that were true it would then seem plausible one could be trained to make the ranges narrower, but apparently nobody ever does this...)

I'll also note woodwinds have pretty much never made any sense to me (the business of having a row of finger holes 1-n, and having 1-k closed get progressively higher as k gets smaller makes sense, which is fine for C,D,E, but then you get to F and it all goes WTF...)

[wrog]

it would then seem plausible one could be trained to make the ranges narrower,

Admittedly, by this logic I should be able to train myself to recognize quarter tones, which I've never actually tried (never really had occasion to do, not being a fan of the sorts of atonal music that tends to use them), and I'm not sure it would work, so most likely that's my answer right there.

[annathepiper]

The flutes I have in different keys are all keyless; they're my carbon fibers. They're all essentially designed the same way, i.e., six primary fingerholes. I still have keyed flutes (the one I bought in college, as well as my keyed piccolo), but I don't play them as often just because I like the tonality of the carbon fibers better when I'm playing in session.

So the fingering works essentially the same way on each keyless flute. But if I put all fingers down on a G flute, I get G. But if I want to get a G out of the D flute, it's the left hand fingers (the upper holes) all down.

The part that keeps throwing me is less a question of pitch and more of a question of muscle memory. Usually I play my D flutes (usually the small ones, the piccolo-sized ones) as that's most generally useful in session and I can usually handle tunes on it in G and A as well...

But there's a caveat. That caveat is the range of the tune. If it goes below D over middle C, then I can't play the whole thing on a D flute. So I have to rearrange what notes I'm playing. Or, I have to try playing it on the G or A as appropriate, as those go lower in pitch.

But my muscle memory is still locked in on the D arrangement of fingers. I've gotten better at learning tunes by ear the last few years, but my brain is going "okay I know that's a B, B means first finger down"... but that's assuming I'm on a C/D flute. B is four fingers down on the G or five fingers down on the A.

I'm pretty sure this is a problem I can solve just by practicing on the G or A flutes more often and teaching my muscle memory to adjust to those flutes as necessary. :)

I can distinguish differences in pitch, no problem. I can go "yeah, that note is higher than that one", and like I said, I'm getting better at learning things by ear after participating in regular workshops. What I have trouble with though is remembering "if the note is this pitch, that means it's A". My brain goes instead "if the note is this pitch, that means I put the fingers in these places".

[wrog]

I see. You're playing by ear and not off of a part (i.e., there's no composer or conductor writing out parts specifically for your D flute or your G flute) so you actually do have to care what the concert pitches are, and "Molly on the Shore" in (concert) G major is going to be a completely different part (different notes, mechanically, hence different technique) depending on whether you're playing it on the D flute or the G flute. And while you can learn both parts, the technique for one isn't going to carry over.

But it wouldn't be an issue if you learned it in (concert) G major on your G flute and the group wanted to switch it to playing it in (concert) D major and your D flute was available. That's the way this stuff is supposed to work.

[wrog]

and I've since learned from the FB thread that alto/bass recorder music is not actually transposed, i.e., it's all written in concert pitch and so recorder players are essentially in the same boat I am (i.e., having to treat fingerings for C instruments vs. F instruments as being different),

and from what you're saying, it sounds like this has carried over into the flute world.

Wow

[pjz]

I *really* enjoyed that explanation, thanks for spending the obvious time and effort that went into it!

First off, I must say that I'm extremely _non_ musical, with no musical training in anything (voice, instruments, etc) beyond the basics in first or third grade or something.

This is, in part, because being analytical-logical-guy that I am, I started asking questions like "how long is a note?" and "if you play the same note on different instruments, why does it sound different if they're all the same note?" The answers I got were part puzzled looks and part "They just are what they are", neither of which I found satisfactory.

Anyway, it sounds to _me_ like you play everything by relating actions (fingerings, whatever) to the result on the 'absolute scale' (concert scale?), which, as you point out, has advantages and disadvantages. You're dereferencing the pointers' to start with, going from action -> sound without the intermediary 'notes'.

From what I can tell, most people play by associating actions with relative changes (action -> note-delta) and then rely on the instrument to give them a reference note to base all their deltas on.. without necessarily caring too much about what that reference note is.

I can only imagine that playing 'completely visually' must, for you, be something like a stroop word/color test.

But again, thank you for the very interesting explanation! And maybe a pointer-dereference analogy might work better than your color blindness analogy. Though of course you may have to run the pointer analogy through a mailbox analogy to communicate to non-programmers :)

Also: I can't decide whether all of this makes me more glad or more sad that I'm so non-musical.

Re: Wow

[wrog]

spending the obvious time and effort

more a tale that grew in the telling and me realizing that I really ought to get it down somewhere; there being people out there that study this stuff, though I have no idea if this relates to any of the questions they currently care about.

makes me more glad or more sad that I'm so non-musical

Yeah. I never learned how to be a pilot. Can't do everything. But also way less likely to die in a plane crash, so there's that.