There is an entire genre of jazz writing built on how long dead pianists sat at the keyboard. Eight hours. Twelve hours. Through the night, every night, for years. We went looking for the primary sources behind those numbers for the ten pianists on this site and could not find them.
We went looking for what is actually documented about the ten pianists on this site, using the same source discipline their pages are held to: a claim gets printed only if it traces to sources that agree. What came back was more interesting than the legend, and in two cases what came back was nothing at all.
Why the hour counts survive
What follows is our own argument rather than anything the sources say, offered because the pattern will keep producing these stories after you finish reading.
An hour count is close to the perfect piece of misinformation. It is short, it is memorable, it flatters both the subject and the teller, and it is almost impossible to disprove — nobody was standing outside the room with a stopwatch, so the absence of evidence never looks like evidence of absence. Then it gets quoted in an interview, and the interview becomes a source, and the source gets cited by the next writer, and within two decades a number nobody ever measured reads as established fact.
The information that would actually help a student has the opposite properties. Who the teacher was, what repertoire was in front of them, at what age, in whose front room — all specific, all checkable, none of it quotable. So it gets left out, and what survives is the number.
What the record supports
Art Tatum was taught by someone who would not teach him jazz. His formal teacher, Overton G. Rainey, was visually impaired like Tatum, did not improvise, and discouraged his students from playing jazz at all. Tatum attended the Jefferson School in Toledo, then the School for the Blind in Columbus from late 1924, then the Toledo School of Music. So the technique everyone still marvels at was assembled by a man whose formal lessons were about something else, which is a far more useful fact than an hour count. The training was real; the jazz was self-directed on top of it.
Oscar Peterson started at home, then went classical. His first teachers were his father and his sister Daisy, and he began classical lessons at six. He also left the clearest statement any of them made about what practice is even for:
Technique is something you use to make your ideas listenable. You learn to play the instrument so you have a musical vocabulary, and you practice to get your technique to the point you need to express yourself, depending on how heavy your ideas are.
Read that again, because it inverts the usual order. Technique is not the goal and it is not a fixed target either. The amount you need depends on how demanding your ideas are, which means the question “how much should I practise” has no answer until you know what you are trying to say.
Herbie Hancock taught himself the jazz and studied the rest. There were no jazz teachers on the South Side of Chicago when he wanted one, so he worked the idiom out himself. He had already played a Mozart concerto with the Chicago Symphony at eleven, and he went on to a music degree at Grinnell. The instrument and the theory were schooled; the language was not.
McCoy Tyner learned from a neighbour who had no piano. Bud Powell lived around the corner in Philadelphia and, with no piano in his own apartment, came to Tyner’s mother’s house to play. Tyner’s own account is that it was hard to understand everything Powell was doing, but that he liked it and learned a great deal from Bud and from his brother Richard. That is a description of learning by proximity rather than by curriculum, and it is how a startling amount of this music was actually transmitted.
Chick Corea was given arrangements to work out. His father, a Dixieland trumpeter who led bands around Boston, started him on piano at four and wrote simple arrangements of popular tunes for him to figure out. Formal lessons with the Boston concert pianist Salvatore Sullo began at eight. Both things are documented; what the record does not say is how the two fed each other, so we will not tell you which one made him.
Ahmad Jamal studied everything and called the result discipline. He named Art Tatum, Bach, Beethoven, Count Basie, John Kirby and Nat Cole as what he studied, and said he was working on Liszt. Formal study began at seven with Mary Cardwell Dawson, who founded the first African American opera company in the country. He rejected the word “space” for what he did and used “discipline” instead — a discipline he said came from having worked in every configuration a musician can work in.
Thelonious Monk was learned FROM, in a band. We have no account of Monk’s own intentions as a teacher. What we have is John Coltrane, writing in Down Beat in 1960 after their Five Spot residency:
Working with Monk brought me close to a musical architect of the highest order. I felt I learned from him in every way; through the senses, theoretically, technically.
Keith Jarrett records classical repertoire too. He has recorded Book I of Bach’s The Well-Tempered Clavier for ECM, along with Shostakovich and an album of Gurdjieff’s compositions. Whatever that represents in hours at a keyboard, it sits alongside the improvising rather than before it.
Bill Evans took a degree in it. He went to Southeastern Louisiana College on a scholarship and graduated in 1950 with degrees in piano performance and music education. Before his recording debut he could read at concert standard and had absorbed Debussy, Ravel, Bach, Milhaud and Stravinsky into his working vocabulary. That is a conservatoire education, and it is the most formal training of anyone here.
What the record does not support
For Bud Powell we found nothing about how he practised that we were willing to print.
That is not a gap in the research so much as a fact about the sources. Powell has an enormous literature and it is far more interested in his illness and in his tragedy than in what he did at a keyboard on an ordinary Tuesday. The stories circulate; they are simply not traceable, so this site does not repeat them.
The same caveat applies to everyone above, and it is worth being precise about it: what the record gives us is TRAINING — who taught whom, what repertoire, at what age. For not one of the nine does a documented daily practice routine appear in the verified manifest behind these pages.
We are stating that plainly because the alternative is what everyone else does: fill the silence with something plausible and let it become the consensus by repetition. A page that says “we do not know” is more useful than a page that guesses well.
What the documented cases actually have in common
Not hours. No hour count survived the check for any of the ten.
What did survive are two patterns, and because this piece is about counting rather than asserting, here are the counts.
Documented formal musical education outside jazz: six of the nine. Counting anyone for whom the manifest names a teacher, a school or a degree: Tatum (Rainey, three institutions), Peterson (father and sister, classical lessons from six), Evans (a degree from Southeastern Louisiana), Hancock (a music degree from Grinnell), Corea (Sullo from eight) and Jamal (Dawson from seven). Hancock is in on the degree, not on the Mozart concerto, which proves he could play rather than that anyone taught him. Jarrett’s classical recordings are evidence of repertoire rather than of schooling, so he is out. For Tyner and Monk the manifest says nothing either way, which is not the same as a no.
A formative person close by who was not a schoolteacher: four of the nine. A father writing arrangements for Corea. A father and a sister for Peterson. An uncle setting Jamal a puzzle at three. A neighbour with nowhere else to play, for Tyner. The Monk entry above does not belong in this count and it is worth saying why: it documents Coltrane learning from Monk, which tells us nothing about how Monk learned.
Neither pattern is universal and we are not going to round them up to one. What is fair to say is narrow: in the cases where anything at all is documented, it is those two things that got documented. The practice-room isolation the hour-count stories imply appears in none of these records, which is an absence of evidence rather than evidence that it never happened.
Questions people actually ask
- Did Art Tatum really practise twelve hours a day?
- The verified source manifest behind his page on this site contains no primary support for it, which is why the number does not appear there. That is a bounded statement about our own evidence rather than a claim that no such document exists anywhere. What IS documented is who taught him and what that teacher refused to teach, which turns out to say more about how he learned than any figure would.
- Why do so many practice stories about famous musicians turn out to be unsourced?
- Because a number is easier to repeat than a method, and nobody checks. A claim like "eight hours a day" is memorable, unfalsifiable and flattering to the teller, so it survives being copied from article to article until it reads as common knowledge. The specific, checkable facts — who the teacher was, what repertoire was studied, at what age — are less quotable and are the ones that actually help.
- Does this mean the practice advice on this site is different?
- It means it is traceable. Every biographical claim on the pianist pages here is tied to a manifest of sources that a build check enforces, so a story that cannot be sourced does not get printed. Where the record is silent this site says so rather than filling the gap with something plausible.