Most Difficult Piece in Standard Repertoire

[ahinton]

That Brahms work is far from undemanding, but the implication that it is "harder" to learn or to perform that, say, Alkan's solo concerto surely stretches credibility well beyond breaking point. That said, the very issue about some difficulties being in the hands of the beholder seems very much at issue here, given that I've never quite managed to identify with the barbs that I have heard from time to time about some of Brahms's piano writing being "uncomfortable" and "against" the instrument or the pianist or both or some such...

Best,

Alistair

[djealnla]

You did not shorten it as one might typically use the word.  You omitted the majority of it, including its explanation.

I know. 

In the meantime, check this out:

https://www.pierre-arnaud-dablemont.com/blog/5-most-difficult-piano-pieces-541

The guy doesn't have a Wikipedia page, though.

[sucom]

Perhaps an easier question to ask might have been "Which piece have YOU found to be the most difficult to learn and why?" 

I play the piano because it puts me at peace with myself and takes me away from the strains and stresses of the world, which is more than reading through this particular thread has done!

I can't think of one particular piece that has been the most difficult although the glissandi in Ravel's Alborada del Graciso made my fingers bleed and practising the repeated G sharp loosened the G sharp key on my piano. Also any piece where the composer assumes the player to have a large hand span are the most challenging for me personally.

[djealnla]

Hm, some posts have apparently been deleted.

[retrouvailles]

Hm, some posts have apparently been deleted.

Yeah, the admins here have a reputation of deleting posts without any sort of notification, which can seem rude, in my opinion.

[lmpianist]

Edit: modified original post.

I think that part of the trouble we're having is that you're approaching this problem through a philosophical lens, which is all well and good in determining whether there exists an answer to your question, but it doesn't do much to help us actually determine the answer, assuming it exists.  To do that, i.e. to determine the piece X[k] which is by some definition the most "difficult" of pieces X[0], X[1], ..., X[N] (or equivalently that two or more pieces are equally the "most" difficult of the finite set), I believe we are required to employ mathematical methods.  In doing so we may well find that there is no solution, or that the solution cannot be determined.  Anyway, it's been a while since my days of studying real analysis so my mathematical nomenclature might be off a bit; apologies in advance.  

Assuming your definition of difficulty is exactly as you say in your first post (which btw is not 100% clear, suggest refining it to mathematical precision), you would have to (1) define a finite set of (ideally) independent variables that somehow encapsulate your definition and (2) quantify the difficulty of each piece in terms of a function f over those variables.  Then you would need to (3) define a metric '<' for the range space of that function with which to compare the results for each piece.  

Since I have no idea what the variables or range would even be, not to mention the function, I can't begin to answer your question, unfortunately.  It seems conceivable that an answer might exist, but the effort required to compute the value of the function for each piece would probably be insurmountable.  Think about evaluating it for the Brahms Second Concerto, for example... you'd almost need to set up a distributed computing project along the lines of GIMPS or SETI@home to get even close.  This is not an unattainable goal, however.

I think we can agree that f(Czerny study) < f(Chopin etudes).  This would be a good test vector.

In the meantime, I suggest you design some simulations to get approximate values for the pieces in your poll, and then post the results here for us to analyze in greater detail.  Of course you should also provide mathematical proof that your choices (1) and (2) together satisfy your original definition, otherwise expect bitter opposition.  Good luck.

[rob47]

ligeti etudes by a considerable margin in my humble opinion

[ahinton]

ligeti etudes by a considerable margin in my humble opinion

That's provided that they can be regarded as "standard repertoire"; I'm not suggesting for a moment that they shouldn't be, but others might disagree, with or without good reason...

Best,

Alistair

[john11inc]

I think that part of the trouble we're having is that you're approaching this problem through a philosophical lens, which is all well and good in determining whether there exists an answer to your question, but it doesn't do much to help us actually determine the answer, assuming it exists.

Who says someone has to find "the" answer?  Nobody is going to.  That's been stated and accepted multiple times.

To do that, i.e. to determine the piece X[k] which is by some definition the most "difficult" of pieces X[0], X[1], ..., X[N] (or equivalently that two or more pieces are equally the "most" difficult of the finite set), I believe we are required to employ mathematical methods.  In doing so we may well find that there is no solution, or that the solution cannot be determined.

The solution exists; it can be determined therefor.  This has been proven.  The computability has not been determined, or even discussed.  This in no way operates within the realm of whether or not there is an answer.  Nobody will find that there is no solution, or that the solution cannot be determined, or that there is a solution.  Nobody is going to attempt to do so.

Assuming your definition of difficulty is exactly as you say in your first post (which btw is not 100% clear, suggest refining it to mathematical precision)

The definition of difficulty I have proposed has both been shown to be objective, and has been defined well beyond what is necessary to operate within its parameters.  Regardless of whether or not that definition was completely sufficient in the original post is not important, as it's been further explained several times in later posts.  Please refer to them.  And, I do not understand your comment regarding breaking it down into mathematical terms.  Such a thing seems unnecessary and obfuscating, if you mean to literally parse the definition into analytic logic expressions.  Please explain what you mean.

you would have to (1) define a finite set of (ideally) independent variables that somehow encapsulate your definition

I don't have to define them.  Anyone can define them however they see fit, at the cost of the verisimilitude of their answer.  The "variables", which I will refer to as criteria (as that is what they have been referred to as previously), will necessarily be able to be expressed independently of each other, but the system would be complex, and their relationships would have to be defined regarding the value we place on unions/intersections of them.

and (2) quantify the difficulty of each piece in terms of a function f over those variables.

Of course, as I've stated.

Then you would need to (3) define a metric '<' for the range space of that function with which to compare the results for each piece.

If we've already quantified difficulty, this is superfluous to (2).

Since I have no idea what the variables or range would even be, not to mention the function, I can't begin to answer your question, unfortunately.  It seems conceivable that an answer might exist, but the effort required to compute the value of the function for each piece would probably be insurmountable.

Of course it's insurmountable.  I have stated this as well, several times.  As the degree of specificity of an answer increases (towards a truth value of 1), so will felicitous complexity.  Nobody here will answer this question in adherence with epistemological laws of "knowledge".  To expect that would be ludicrous.

In the meantime, I suggest you design some simulations to get approximate values for the pieces in your poll, and then post the results here for us to analyze in greater detail.  Of course you should also provide mathematical proof that your choices (1) and (2) together satisfy your original definition, otherwise expect bitter opposition.  Good luck.

Perhaps you should do it.

[mephisto]

Could the one who started this thread please explain what "playing a piece" means?

[john11inc]

Could the one who started this thread please explain what "playing a piece" means?

It's been done a few times, now.

[ryguillian]

From one-troublemaker-from-time-immemorial to another...

While the tact and measure with which the poll was elaborated on surprised the hell out of me, the whole thing is ill-conceived, but not because of any would-be objective vs. subjective problem.

The rigor, integrity, and general cadence and flow betrayed by most posts here suggests two qualities about many of this forum's members:

A.  Illiteracy.
B.  Excess chromosomes.

A late—and most happy—2011,

Ryan

[mephisto]

From one-troublemaker-from-time-immemorial to another...

While the tact and measure with which the poll was elaborated on surprised the hell out of me, the whole thing is ill-conceived, but not because of any would-be objective vs. subjective problem.

The rigor, integrity, and general cadence and flow betrayed by most posts here suggests two qualities about many of this forum's members:

A.  Illiteracy.
B.  Excess chromosomes.

A late—and most happy—2011,

Ryan

Welcome back;)

[11ngerman]

I am an 18 year old pianist, and I have a few very difficult piano pieces that should be added to your list.
1) Liszt "la campanella"
2)Liszt "totentanz" for piano and orchestra
3)R. Strauss "burleske"  in D minor
4) Dreyschock piano concerto (VERY difficult)

Although not all of these are considered "standard", La Campanella is, somewhat.

[lontano]

An interesting, if a tad odd, selection to choose from. Surprising to me is that a few works on the list have just become familiar (or long ago forgotten). Anyone know where I can get the score to Corigliano's "Etude-Fantasy"?

[scottmcc]

1) Liszt "la campanella"

Liszt etudes was one of the choices--he didn't separate out individual etudes.

[ch101]

I wouldn't say Mozart or Haydn is the most difficult. Nor would I say Liszt or Barber is.
If you can play all transcendental etudes, doesn't mean you can play any Mozart sonata, in a good way.

mozart is a delicate composer, while liszt is broad, carefree and virtuoso. you need intellectual understanding more than technical skills to play his music.     
the definition of "difficult" is quite broad in this case