Art Tatum Techniques and Concepts

Art Tatum: Musical Techniques and Concepts

In this series I propose to highlight what I consider some of Art Tatum’s most exciting musical discoveries, ideas that I think we can and should use today. The idea of developing a driving left hand, a 'rhythm section' in itself, which at the same time has the subtlety to musically balance the right hand's ideas, is intriguing, a little bit like climbing Mount Everest. The art of making a rhythm section in your left hand hand while still swinging and playing what you want in your right hand - the ideal. It's an exciting challenge, and for me this thought started with Tatum, and has lately been extended by listening to Dave McKenna.

Please don't get me wrong on this: I love Bill Evans' solo style and all the other great players. The music changed with the times, the musicians went with it, even though there's plenty of proof now that those societal changes were orchestrated by external powerful forces that had virtually nothing to do with love of music.

First I propose to examine Tatum's use of descending right-hand runs and patterns, then his left-hand chordal concepts. I will be exploring Tatum's conceptual fingering style on runs, not so much his random 'fill-in' improvisations, found in any of his transcriptions, where his notes choices in the right hand 'disagree' with the chord to the extent that you start to realize he's using the pianistic feel of the keys (what feels good under the fingers/keys) as opposed to a thought-out lick on a chord.

The key point in this concept is to realize is that the right hand is subordinate to the left hand: the left hand is the rhythmic anchor, so everything in the right hand is basically 'whatever you can fit in'. That's why the runs have to sound effortless, because the time feel is all-important. And that's why so many of Tatum's "random fill-ins", as I call them, are not chord-melody based but literally finger or hand-shape based.

To get back to my theme, by examining Tatum's music - much like Charlie Parker's - I believe we are opening the door to further discoveries in the world of classic jazz arranging.

So in describing these runs and their fingerings, I would like to stress that I am concentrating on a relatively minor part of the total musical experience. I get that. What I'm doing is zooming in on a particular facet of Tatum’s performance, for which he is deservedly well-known, and exploring it. But in what we might call good all-round musical practice, harp-like runs such as Tatum used would constitute a relatively small part of the solo jazz pianist’s total bag of tricks. No-one is ever going to outdo Tatum, and more importantly no-one is ever going to sound like the real emotional Tatum, the one I love, not the flashy one. And so we have to realise that in order to incorporate some of his devices in our own playing, we have to do a hell of a lot of work for a relatively brief musical result. To be able to use them at all musically, it takes many hours practicing a run, thousands of times over, until it can be incorporated naturally in a jazz performance. It takes years for it to come out naturally (unless you're working 6 nights a week). Again, that particular run or partial-run maybe lasting for only a second or two in actual performance.

What I am getting at is that we can use and should build upon the techniques of the past, such as these Tatum runs, to add to our arsenal of solo piano techniques.

Future posts will explore Tatum fingerings for descending 9ths and 13th runs.

Tatum Descending Right-Hand Runs

Art Tatum: Descending Right-Hand Runs

Tatum favoured three-finger runs, usually 3-2-1, sometimes just 2-1. The strongest fingers naturally give the smoothest, most even result. As Chopin intimated, it's all about fingering. The runs should be like waterfalls, pearling, cascading. That means cultivating a subconscious response, a Zen-like 'no-thinking'—which in the end just translates to lots and lots of work, at least for me.

Tatum often liked to start these descending runs with an initial flourish, a turn slightly different fingering to the recurring pattern, so I have included a few of them in the following examples. They fit under the fingers and serve as a 'springboard' for the run. I would like to eventually be able to mix and match these runs so what may begin up high as a 9th arpeggio becomes a 13th, then scales (literally) down to the target note. What seems fuzzy at the moment for the artist after due process (practice) can indeed become natural, and express in an organic musical way.

For other viewpoints there are a few Tatum books out there, please check them out if you haven't already. I first discovered the genius of Tatum's conception about twenty years ago, the "I Surrender Dear" transcription, but I began my specific investigation of his runs due to the inspiration of a guy on YT under the moniker rkjp56. Please check them out:

https://www.youtube.com/user/rkjp56/videos

Man, thank you for sharing this beautiful knowledge. 

Besides a couple of transcription albums, there's a great book called "The Right Hand According To Tatum" by Riccardo Scivales. It has some fantastic stuff. But I always thought, what about Tatum's left hand? That subject to me is even more interesting, and after I get through these right hand concepts I'll be delving into those dark and very interesting waters. 

Again, I reiterate these are my Tatum 'pickings', my observations. You may have different views, different conclusions. Everything you find here should be considered my take on Tatum, and music in general. We don’t have any definitive guide as to what fingering he used when, where and in what context. So in this sense, please view the following as my educated guesswork, even though I'm in a constant process of trying out and establishing the best fingerings that work for me.

And by highlighting Tatum in these initial posts, I don't want to convey the impression that many might jump to - that this is the music of the past and somehow irrelevant. My whole take on this is that we should look at these ideas and build on them. To make Tatum's concepts work in a new way, to integrate them with everything else we play from the more contemporary tradition.

I would like to get this kind of information out there because I don’t really find it anywhere else on the net. What I'd like to get across is the Tatum concept of harp-like arpeggios using a simplified fingering which gives an even sound to the descending patterns. Seeing the piano is a just like a harp it seems a pity that modern jazz piano has seemingly forgotten this aspect which so pervades classical and romantic piano tradition. Let's not forget that Tatum was thoroughly schooled in the European classical tradition, just as much as he was in the blues. And have you heard him sing the blues?

https://www.youtube.com/watch?v=eecqu_YSPZY

The plan is to present my current take on all the descending 9ths chords; then the descending 13ths and their variations; then look at other runs and effects, such as whole-tone clusters (foreshadowing Monk). Quite a lot of Tatum is the genesis for later piano developments, which should be acknowledged. In later posts I'll be taking apart Tatum's left-hand concepts, and look at how he made a simple II-V pattern, a two-bar sequence into eight chords as a flowing melodic accompaniment.

Tatum 13ths

The first time I really got into Tatum was through his recording of "I Surrender Dear" (1955).



I had the transcription, and I was fascinated mainly because he was actually playing 'live' all those wonderful arranging concepts I'd been interested in, but never really knew how it could be incorporated into jazz improvisation before. Maybe it sounds dumb to some people. It's hard to explain what exactly motivated me to explore Tatum, because as a piano player I had nowhere near the level of requisite technique usually associated with attempting anything Tatum-esque. Maybe it was because I was such a novice in comparison, I had nothing to lose, in a way: it's like a mountain peak that you're forever striving to reach. And I like an impossible challenge. There aren't too many musicians in history that we can say are close to perfection, but I think Art Tatum is one of them. If we had recordings of J. S. Bach or Chopin, we might all agree on that point.

But jazz is all about making whatever it is you like into your own, so that you own it. Whether I'm playing some Tatum-inspired phrase at 1/10 or even 1/100th the speed and accuracy he played, doesn't matter to me in the slightest. What matters to me is if I can use his concepts in my own playing, to make them mean something, truly incorporating them into my own playing. It can't sound forced. In the literature you will find many jazz commentators who'll tell you Tatum wasn't really 'jazz' because he played the same arrangement all the time. But these people have no idea what a working musician's life was like back then - and I mean a solo pianist's working life. You have to have a plan, a structure, some kind of edifice on which to build your sandcastles.

Tatum was legally blind, yet he mastered not only the 'licks' of the classical repertoire (to put it crudely), but also the whole gamut of jazz piano: so much so that he is a virtual compendium of the complete history of jazz piano, with foreshadowings of the future.

This track was recorded near the end of his life, in the marathon Norman Granz sessions for Verve Records in January 1955. But you gotta hear the mid-1930s Tatum to fully appreciate what he could do and what he meant for musicians of the time.

This was the lick that first got me - the D13 and how it resolved to the Dm in the key of C. The following excerpt is the first seven bars from Jed Distler's Art Tatum Jazz Masters Series, an invaluable book of Tatum's transcriptions. And with acknowledgements by the likes of Dick Hyman and Felicity Howlett (whose thesis I've yet to read), this is an authoritative document.

here is the D13 lick excerpted from bar 6:

the audio excerpt, from a bar before:



This is a great example of the famous 2-1 fingering for which he was famous:

How simple and beautiful!

Once you get the idea, you can expand the concept to the other 13ths.

The point being, the 13th chord is the quintessential jazz chord, the sound that is so captivating to the humans that are really awake out there. As I pointed out in earlier posts, this sound was first introduced through classical music, Ravel and Debussy, but I'm sure it can be traced back much further.

In our next post we'll look at the 13th chord conceptually, and the ridiculous attempts by Wickedpedia to muddy the issue, you can bet on purpose.

The 13th Chord

Today when we want to look up anything 'factual', we go to Wikipedia. It's touted as a "Free Encyclopedia" where anyone is meant to be able to contribute, but the reality is totally different. It's really a castle full of gate-keepers, where the only information that ever gets out is the information that they deem worthy. It's a one-world view.

Sure, if you want to know about the size of an emperor butterfly's wings or the type of soil encountered in the rainforests of Brasil, you can get pretty much the right info. But go to the Wikipedia page on the 13th chord and all you will find is a mish-mash of gobbledygook. Go to the 'Talk' page and read the thwarted attempts by musicians to get some sense into this article.

The particular individual responsible for this particular instance of gate-keeping our musical information is called "Hyacinth":

This is the person responsible for providing information that is meant to be for the benefit of humankind in general. Did you read the 'talk' page? It might take a few readings to understand what's going on. But this page sums up Wickedpedia, because like our whole society, it's a total sham. And they've got the gall to ask for donations every year, as if they're some struggling operation.

It's got a lot to do with 'academics'. These 'contributors' are often doing their wiki-editing in their spare time for free, most probably connected in some way with a tertiary institution. That means they think they're smart. They most probably either studied, or maybe taught (OMG) at some university. But if you've happened to come across any 'degree holders' in your life, or, God forbid, had to deal face-to-face with these people, you'll know that academic qualifications don't mean shit nowadays. Have you noticed the way they rely on 'peer review' so much? Your peers are your competition - in everyday terms. So to say something has been 'peer-reviewed' just means it's been OK'd by the gang, the club. How cosy.

Well, after having bagged the Wiki gatekeepers, I'd better come up with the goods myself.

So, what is a good description of a 13th chord?

In a previous post, I outlined the basics:

https://classicjazzarranging.blogspot.com.au/2016/10/jazz-chords-9ths-and-13ths.html

So many songs, from the most basic to the most sophisticated, end with their final melodic resolution going from the 3rd to the tonic (e.g., E to C in the key of C). That melody note E is on a G7 chord, which makes it a 13th, heading for the C on the tonic C major chord, a V-I progression:

Audio:



As I mentioned in my previous linked post, the 13th can be considered originally a suspension, an appoggiatura for the 5th (of G7, that is the D), which over time gradually became incorporated into the chord itself. It replaces the 5th. This to me is the most natural interpretation of the 13th in general practice.

But the important thing to realize, as I said previously, is that the 13th is the quintessential jazz chord. Not only in its most natural setting, but with different bass notes, the chord quality changes:

Audio:



You will recognize this sound everywhere in jazz. Its characteristics are the minor 2nd (maj 7th) and the tritone (b5th).

Next up: Tatum fingerings for 13th chords

Tatum fingerings: Dominant 9th chords - C9 to Db9

C9: 2-1 fingering

same fingering for F9:

and Bb9:

Eb9 is one of the trickier ones, needing the 4th finger as the pivot.

This one is not very practicable. But neither is the most obvious 5-3-2-1. The whole point of this series is to explore the Tatum 1-2-3 fingerings, fingerings to make for more even runs.

Even though I'm running through all twelve 9th chords fingerings here, some are just not used, even by Tatum. Some phrases work great in one key, terrible in another.

Eb9 is a bit of a conundrum to me at the moment. There are a few variations, such as:

As I mentioned previously, in offering my fingerings for these chords, I'm wanting to share my explorations and discoveries, investigate how they can be integrated into contemporary jazz piano practice, and maybe extended into big band or orchestral concepts. It's all music, but I like investigating the 'nitty-gritty'. And investigating Tatum takes us back to the source, as it were, because a lot of Chopin and classical piano involves harp-like runs, and we know Tatum was thoroughly grounded in the 19th century romantic piano tradition.

In no sense am I suggesting these are definitive fingerings; rather, I encourage you to explore them yourself. I'm simply attempting to illustrate this Tatum concept, backed up by my own research—the idea being to make the fingerings of these harp-lie runs as even as possible without resorting to hand/finger contortions.

Ab9:

Thumb on Gb and C. Thumb on the tritone, the Devil's interval!

You should be getting the general idea that the hand makes a crab-like motion, alternating (pivot) 2-3:

Db9:

Thumb again on the tritone, this time F and B (Cb), alternating 2-3

As I mentioned previously, any dominant 9th chord is like a minor 6th or a half-diminished, so you get three for the price of one.

Continued in the next post.

Tatum fingerings: Dominant 9th chords - Gb9 to G9

Now we've come halfway round the cycle from C to F# - or is it Gb? We don't use the key of Cb major all that often, but I still seem to think of it more as a Gb9 myself. It's whether you like having Bb and Fb instead of A# and E. In this 'borderline' territory between sharps and flats, arrangers always should go with the simplest solution. It's OK to use the more common enharmonic names.

Gb9:

Thumb on the Bb & E, tritone.

If you prefer to see it as an F#9, here it is in the key of B:

B9:

Thumb on the D# & A, tritone again.

E9:

Thumb on the D. This one needs the 4th finger as a pivot on the B to D (4-1).

These excerpts are simply to show my current fingerings, not intended as any kind of specific drill or exercise. The only important point I would stress when practicing them is to play them in time, slow medium and fast, semiquavers or triplets. But they should relate to the basic pulse, they should always be recognizable subdivisions, at least in practice sessions.

To reiterate on the bigger musical picture: all this work on runs amounts musically to fractions of seconds in actual performance, when playing (jazz) time. If playing rubato, of course the runs can be extended or stretched out. But to my mind, the run is all about gesture, not something to be itself 'noteworthy' - pun intended!

A9:

Thumb on B and E, 3-2 pivot.

D9:

Same fingering as A9.

G9:

This one is quite difficult; I think the key (pun unintended) is to feel the pivot between 3 & 2.

That concludes all the Tatum dominant 9th chords fingerings. In future posts we'll be looking at the more exciting 13ths, and some of the variations possible in mixing and matching these ideas.

Tatum 13ths - fingerings C13 to Eb13

When I say "Tatum fingerings' you must realize I'm only speaking from my personal viewpoint. Every pianist is different and every pianist feels things in his or her own unique way. The fingerings given here are just my take on the Tatum concept of using the stronger fingers, 1-2-3, down the keyboard. We will look at ascending runs later on, but for now, here are the descending 13ths from C13:

C13:

C13 audio:



F13:

F13 audio:



Bb13:

Bb13 audio:



Eb13:

Eb13 audio:



These are all 2-1 fingerings.

That's the first part of this particular look at Tatum 13ths fingerings. Now from Ab7 they get a little trickier.

Jazz chords: 9ths and 13ths

Jazz Chords

The history of the musical use of upper chordal extensions (9ths, 11ths, 13ths) in the last century and a half has developed in a way that echoes the evolution of our modern society. In earlier times, supposedly life was 'simpler' and the music probably reflected that. Folk music and church music used simple triads based on one tonic note, key or 'ground bass'. We surmise that music began as simply melody, sung in unison or octaves. 'Homophonic' in its original Greek sense. If you read music history, it seems like we're expected to believe that music was 'unison' for thousands of years, I guess with the odd 5ths thrown in. Then evidently around the turn of the 13th century, at newly-built Notre Dame Cathedral in France, 'Perotin' was one of the first to use voices singing different melodies that came together to form what we would call 'chords'. Separate voices singing over this ground bass (sustained bass or tonic note) to produce chords, harmonies floating over this tonics or series of root notes, you could say modal music. So that's a few millennia of unison, then a thousand years of harmony, if you believe Wikipedia or history generally. And now look at what we've finally ended up - in our mainstream musical world.

The desire in artists to create is very similar to religious devotion, and the goal of 'religious composers' like Perotin and Hildegarde of Bingen was to reflect the wonder of the Universe and God through the magic of music. The Pythagorean divided string, with its pure mathematical ratios of the main intervals of the octave, 5th, 4th and 3rd, defined "Just Intonation". The same ratios also defined the architectural dimensions of their cathedrals. Purpose-built, you might say.

The sounds the made were pure, the major 3rds brighter - but in the Pythagorean system there could be no real modulation possible, as any chromatic notes, even the not-so-foreign ones (e.g., G# on a E7 going to Am) were horribly out of tune.

The history of piano tuning is fascinating. One is gradually made aware of how much 'aural intuition' has played a part in the process, the trained ear, as well as the introduction of technology - the tuning fork in 1711, about the time equal temperament was gaining universal favour. After many years of experimentation the system of dividing the octave into 12 equal parts gradually became accepted, creating our current system of "Equal Temperament". J. S. Bach with his Well-Tempered Keyboard", in all 24 major and minor keys, being the big one. Now modulation between any keys was possible, and it was a very exciting development. 

To continue the analogy with human society: as human beings moved from a rural to an urban existence, things got complicated. Now people were living cheek-by-jowl in a new environment, learning how to get on. It's almost like the notes of the major scale are members of the same tribe, some getting on naturally, some clashing; but over time we have learned how to blend them and make their inherent dissonance more agreeable.

What intervallic characteristics could we say constitute a 'jazz' chord? Simply put, it's the presence of a tritone and a major 7th (or minor second), that give the 'jazz' sound. And this is not unique to jazz, it can be found in many other musical forms, Japanese, Spanish, Brazilian.

The simplest expression of this sound concept is a 13th chord, here on C7. So it's the dominant chord in the key of F:

Note the tritone Bb-E, the Bb-A major seventh.
Really the 13th should be seen as an appoggiatura, a melodic suspension of the 5th:

These melodic suspensions, initially 'foreign' to the chord, eventually get accepted as intrinsic to the chord. But the pay-off is the note that was meant to be resolved to, in this case the 5th, is removed, out of the picture. So you still get a four-note chord, but now the 13th has replaced the 5th.

That's not to say you can't have them both, but you gotta work them differently from your textbook stacking-3rds methodology. For example here's a nice way of getting that 5th on top of the 13th:

If we might venture to say that 13ths are the characteristic sound of bebop and beyond, then dominant 9ths are the sound of the swing era. But let's not forget all these sounds were already part of classical harmony tradition, from at least Chopin in the 1830s, to Wagner, Ravel, Debussy, Satie. These musical ideas were well and truly established before Tatum came to them.

For example, here is a textbook example of this same C13th chord in Ravel's Trio fro Violin, Cello and Piano, to me one of the greatest works of all-time. In the middle of the second movement we get this:

[click to enlarge]

At the key change from A to F, (and after) the voicings couldn't get more jazz:

Straight out of the jazz harmony books. And this is from 1914. We'll have a look at some more of this work later. Ravel's intriguing comment while working on this composition was: "I've written my trio. Now all I need are the themes." I'm still trying to decode that statement!

Dominant 9th chords

The 'forefather' of the 13th chord is the dominant 9th chord, almost like a chakra 'one level down'. And that particular combination of notes in the 9th can sound very different dependent on the bass or root note. Like the roots of a tree, that bass note lays the foundation of the sound of the chord quality. So, this C9 arpeggio is exactly the same as Gm6 and Em7b5: same note family, different bass note, different sound quality or 'mood'.

Tatum's style was often described as 'orchestral', in that he could literally sound like an orchestra. Often three distinct voices can be heard simultaneously—stride bass and chords, a sustained middle-register ('cello') guide-tone melody, and amazing upper figures. How the hell did he do it? We know he had huge hands, and he could easily span any major 10th. (That is freedom: two notes, and you have any major chord. I can get a few but to be able to nail Bb-D, Eb-G, Ab-C, Db-F would be heaven.)

But there's more to Tatum than just tricks. There's some real musical magic going on there, and as musicians we can and should investigate what he's doing. And the more you do, you begin to realize that some of this Tatum magic can be broken down and explained.

One of the most obvious examples that comes to mind is Tatum's rapid-fire ascending diminished arpeggios, which seem to run the whole length of the keyboard. But upon closer inspection what seems like a 6-octave run is more like 4, broken-up, divided between the two hands. Beginning in the left hand with a diminished scale fragment which fits easily under the fingers, which 'kicks off' a two-octave diminished arpeggio, the right hand takes over an octave above, and continues for another 2-3 octaves (top of the piano). If it was an C (or Eb, A, F#) diminished arpeggio, you know he'd be hitting that top C.

Like a great magician, he makes you hear that he's covered the whole keyboard. It sounds like he's 'unzipping' the piano. He's toying with the piano, literally 'playing' with it in the most natural sense. That reminds me of Bird's doctrine of learn everything and then forget it when you play. And isn't that after all what we are all meant to be doing? 'Playing' music, with the child-like spirit that hopefully hasn't been too crushed out of us by all the relentless bullshit.

But of course Art Tatum was only one man and like any other artist he had his high and low points artistically. He was really a solo performer;  how many jazz musicians can you say that about? There are many detractors that will try to tell you he wasn't a 'real' jazz musician, probably because they hear him play the same arrangement the same way, or else the endless runs and florid technique gets in their ear and they can't get past it. But they gotta dig harder. You gotta look between the cracks, between the runs and clichés; there is much gold to be found within his recordings.

Tatum once said in a radio interview that he was far from happy with his technique, that he still had a long way to go...if you can believe that! But as great as he was, this very human sentiment may ring true for some of us out there.

Next post: Tatum dominant 9ths fingerings.

Chord Symbol Nomenclature

For arrangers, the correct naming of chord symbols is one of those topics where everyone has an opinion, and their opinion is right, everyone else’s wrong. It’s a bit like when a singer hands the piano player a lead sheet to a song: the first thing the piano player usually thinks—to himself, hopefully—is: “that chord’s not right there, that sequence isn’t the best there”. But don't you think that this is the way it should be? Because in jazz we are working within the realm of instant composition, arranging and performance, simultaneously, both consciously and unconsciously. So one needs to have strong views on 'how it should go'.

However unlike a set of chord changes, the chord symbols themselves are meant to be a common language that everyone has agreed upon, and whose meaning they can decipher, just like common word spelling in a dictionary. Chord symbols are simply abbreviations, symbols which are meant to imply certain chord sounds.

The arranger’s use of chord symbols is indeed an arcane and hermetic art. What is the common practice, how has it changed over the years? When do we use say C9 instead of C7? When do we write C7#9 and when do we use C7alt? As much as some musicians would cry, 'who cares?', I would like to get into the thinking behind the use of these symbols.

Let’s start with the basics, 'common practice'. Or maybe that should be 'my' common practice, because as you'll see, nobody agrees on anything 100%. Let me give you my thinking, and then we'll 'compare and contrast' with other viewpoints.

In the beginning, as it were, there was the triad. The third interval from the tonic C can be either major or minor, and so we have this basic division:

So we see that to name a major triad, the letter name is all that's required. Simple! No 'ma', no 'maj'. Just the letter name. For the minor triad, the lower-case 'm' is added to the letter name. So for those most commonly played chords - the major and minor triads - we have quite an elegant naming solution. No symbol at all for the major (most commonly played), and one little 'm' for the minor chord. That is what I call a good abbreviation concept. Inherent in this idea that only one 'm' symbol is used ever, the lower-case 'm'. That way when you hand-write it quickly, whether it looks like upper-case or lower-case, it means the same thing.

Next we move higher in the chord, to the seventh. Two options, as before: a major seventh interval from the tonic, or a minor 7th (or dominant 7th) interval. Now here is where it gets confusing to the novice. Our terms for naming the two possible 7ths qualities are "ma7" (or maj7) and just plain old "7" for the minor 7th. Let's look at the possible combinations extending the triads:

In our thinking we draw an imaginary line around the letter name, such that if we see the little 'm' after the letter name, we know it's some kind of minor chord. If we see Cma7, we know that the 'ma' part does not refer to the 3rd (i.e. the triad), but to the 7th, the major 7th interval from the tonic. If we see C7, we know that's a major triad with a dominant 7th.

Clear? Yeah, like mud.

Let me point out at this juncture that there is simply no way to get your head around this unless you have actually spent some significant time playing songs, playing chord sequences with their root notes, and gradually learning how to interpret chord symbols. You can't get it without hands-on experience: literally, hands on the keys. Doesn't matter if you don't play the piano; if you're a jazz musician, or really any kind of musician, the piano is your workbench, your desktop.

Anyway, let's now look at what the respected elders have advised on the subject.

First, Clinton Roemer from "The Art of Music Copying", 1973, p. 137.

First of all we see that he uses 'mi' for minor triad, so right away we have a different idea from your present author. Roemer still uses just the letter name for the major triad ("G"), but wants to use "GMI" for the minor triad. His thinking is to consistently use a smaller, yet upper-case 'M'. Because as he shows in what not to use (a very good instruction technique), and as I mentioned above, there has always the been the potential for confusion between an upper-case 'M' and a lower-case 'm' - especially when hand-writing music, which was the norm until the recent computer age. And still today not everyone uses the computer. But to my way of thinking, Roemer here is going one extra step, a kind of second-guessing defence which just adds an unnecessary layer.

As an aside, and I love asides, did you know Roemer was Stan Kenton's copyist? I went thirty years thinking of him as 'just' the author of the best book on music copying, and a jazz orientated, commercial and inherently practical one at that. Then one day I came across some of Kenton's scores and instantly recognized the handwriting. The funny thing my first impression was that it was a lot sloppier than what's in his book! But I shouldn't say 'sloppier', just written faster. It was all perfectly legible, and that's what matters. Like an extra lesson that is an addendum to his book...reality! So for me it's good to know just whose handwritten manuscript those great musicians on those classic recordings were actually reading.

When Roemer was around, there were more gigs, more live musicians working, more arrangements being required, and arrangers and copyists had to be able to work extremely quickly. In the making of a Hollywood production, the composer is usually the last man down the line, often getting a short window of a few weeks, if not days, before release to complete his work. In the jazz and commercial music world of a half century ago, the copyist was his equivalent - the last man down the line after the arranger got through with it. And we know how these arrangers can take forever! Back then it was no "print-out file to PDF", no instantly-formatted parts like today, with our whiz-bang music software. Then it was all black coffee, maybe a pipe, and keep going till it gets done.

OK, on with the chord symbols. this time from Gary Lindsay, author of one of the best arranging books out there, "Jazz Arranging Techniques" (2005). Let's see what he has to say:

Hmm...looks like I'm being outvoted here, and by the best! Lindsay and Roemer both go for the 'Cmi'. No one seems to like the 'Cm'. What does the ubiquitous Aebersold have to say?

Not even a plain old minor triad in there at all. And now all the triangles and minus signs!
Now I need a coffee and a pipe.

Tatum - "Yesterdays" live TV 1954

Let's have a look at Art Tatum in action. If a picture is worth a thousand words, a video is probably worth a thousand times times that - is that a million? Who cares? It's the lesson of a lifetime in three minutes. It's like a great painting you can keep returning to again and again, and still find new things in it you never appreciated before.

We know this arrangement was one of Tatum's 'party pieces', specialties that any solo performer must have. He recorded it many times. But in Tatum there is always improvisation in the sense that he is making everything sound fresh, and the differences and moods between versions become more apparent, the more you get to know him. Listen to the way he plays along the with the house band, playing their "we're back on" cue after commercials; making his own segue into his classic version of "Yesterdays". There is the whole history of jazz piano here, looking both forward and backward at the same time, but always so hip. The word 'stride' even seems inadequate. Hear the Monk-like seconds clusters? He was a big influence on the beboppers; that story about Bird getting a job as a dishwasher so he could listen to Tatum play every night. He wanted to be able to play his alto like Tatum's right hand.

This was recorded on Spike Jones' TV show in 1954 (he would pass away only months later, in 1956).

 

Just now in writing this post, checking the link from my YT channel to this post and listening to the track again, I thought it sounded 'too' fast, even for Tatum. Mechanically, unnaturally fast. Sure enough, after checking, it's sounding in Eb minor, instead of D minor. The pitch is wrong; we can see that he's definitely playing in Dm. And I'm sure all pianos were A=440 by the mid-50s.

I don't know why I didn't pick it straight away...but there ain't no perfect pitch in my 'shed. And no one said a word in the comments on YT. Where are all the perfect pitch people? 32,500 views so far, and not one soul picked it up.

I've now checked my original video file, then got out the DVD and checked that; both Ebm. Then I looked at other Tatum clips or docos on YT—all are Ebm. What's going on? Can everything on the internet and media be sourced from that one clip? (it comes from "Tatum Art" box set, Storyville Records 1088603, 2008). Maybe that's a sign of just how narrow our 'net' really is. Anyway, I'll be working towards getting a true key of Dm version out soon, somehow, once I figure out how to detune a video down a semitone.

As expected Tatum's harmonies are unusual. Let's look at the opening bars of the melody.
This the original Jerome Kern:

Notice the movement is from Cm in the first bar, to Fm6 in the second - not the more modern interpretation Dm7b5 | G7b9. This is because there is no implied G7, no B natural in that second bar. That is a clue that Kern was after a modal effect, a pivoting or rocking from the tonic Cm (C-Eb-G) to the Fm6/Ddim (D-F-Ab), a I-II swaying. The only difference between the Fm6 and Dm7b5 is what bass note you have: if an F in the bass, it's an Fm6; if a D, then a Dm7b5. Note the slight ambiguity here in Kern's bass figures, almost like the second bar is | Fm6-Dm7b5-Fm6-Dm7b5 |. This is also because the melody note is an F, so he doesn't keep that F in the bass.

Now let's check out Tatum's version compared to the original Kern harmonies for these first two bars. Because I think Tatum's choices had a lot to do with the natural resonances of the actual piano pitches, as well as finger/hand feeling, I'm transposing the Kern up a tone to Dm, so we can contrast with Tatum's take on the harmony:

That's a pretty interesting re-harmonization. Those sixths in the left hand were a common feature of Tatum's left hand chordal concept. I've called them 'diminished', but notice there is no 5th or flat 5th; we could just as easily call it Bbm6 - Am6. In fact if we played an E on the Bb it doesn't sound right, likewise an Eb on the A. So they're neither really 'diminished' or 'minor 6ths' in the strict sense. This Bbm6 voicing is like the top half of an Eb9, the Am6 likewise D9. But theoretical 'missing notes' from chords is what the palette of harmony is meant to be about. Sometimes two notes is all you need.

Tatum could easily span the Ab-C tenth on the Ab6 (sounding more like an Abma7 with the melody note G) although here it looks like he could be playing the C with the right hand. But he could play all 12 major 10ths - bang, down straight. Wouldn't that be a nice ability to have? To be able to nail every major tenth with the left hand, without having to roll/fake them.

As for the G7b5 - well the Db is so low it's a kind of a toss up between G7b5 and Db7b5. Half and half, top 'n' bottom. I've seen this voicing in Johnny Richards charts, in "Monk's Mood", a very ''40s' early bop sound with the low b5 (Db) almost in competition with the root (G). Usually the b5 on a dominant 7th will be more like a #11, that is, an 11th away from the root tone or fundamental. Here it's right up against it. And the Db in this context is kinda sounding like the C# of A7, the leading tone to get back to Dm again.

Sure, Tatum played this arrangement over and over, refining it over the years. But isn't that what 'living' arranging really is - making something that's basically worked-out come alive and fresh, like the performer just thought of it?