Something that's interested me a great deal in the time I've been in the business is indeed the development of electronic instruments which allow for some kind of feeling of 'performance' - for it's a bit ridiculous in concert just to watch people clicking on their mice and fortunately one doesn't see too much of that kind of thing any more. There used to be a lot of writing on the electronic/acoustic instrument relationship which rather spectacularly missed the point in failing to take into account that aspects of performance are indeed what most of us go to concerts for.

There was an article on Ferneyhough's bass clarinet solo Time and Motion Study I once upon a time which concluded with something along the lines of 'this piece is much too demanding for a human performer and would be better realised electronically'. The fact that the whole idea of the piece was an investigation of what happens when a performer is put consistently at the limits of what is realisable in concert (hence the title of course) had clearly escaped the writer. There were also plenty of articles in which composers rejoiced in the fact that composers could now in tape pieces realise speeds and tempo relationships no human could manage. The fact that that wouldn't in itself make going to watch someone press 'play' any more interesting, especially when the resulting sounds were no different from one performance to another, seemed not to be a factor.

Fortunately there are now many composer/performers on electronic instruments (Richard Barrett is of course one) who employ instruments in which the physical actions of the performer do bear some relationship to the sound that comes out. So there's something to watch. What those relationships are is something that can be determined in ways other than those dictated by traditional instruments. When a pianist attacks a chord from a couple of feet above the keyboard it's quite clear what's going to happen; but Richard's setup (not only his of course but that's the one I've seen in action the most) lets him determine in advance what the speed of depressing the keys and the pressure with which they're held down will do and they can be just changes in dynamic or radical changes in the sound.

But then on top of that there's the question of how much longer the whole concert business is going to last and I'm not always very optimistic about that. If you're not dealing every time with the response of the audience then the music will clearly be quite different...

"There was an article on Ferneyhough's bass clarinet solo Time and Motion Study I once upon a time which concluded with something along the lines of 'this piece is much too demanding for a human performer and would be better realised electronically'. The fact that the whole idea of the piece was an investigation of what happens when a performer is put consistently at the limits of what is realisable in concert (hence the title of course) had clearly escaped the writer."

How much of a Ferneyhough (or any 'complex' score) is realisable by the performer? Is this almost a kind of improvisation where a performer is, to a certain extent, asked to create his/her own music? (Obv the performers are always creative but this is another type of creation?)

Of course, there are quite a few performers (who aren't involved in composition) doing work on laptops that are quite dull to watch, although MIMEO (where all those borders between performers/composers/instrument builders/programmer/audience seem to be tested), an excellent ensemble (10-12) made up of improvisers, many of them with their army of laptops/electronic instruments did put on a show at the Serpentine Gallery a couple of years ago where you could move around while they played. The members of the ensemble took breaks themselves, leaving others to improvise in mini-ensembles of 4-5. The first 45 mins I'll always have fond memories of, the evening (musically) did run out of steam, but I suppose that ws their approach to some of the problems regarding electronic music performance.

I can't say it much better than can he who is asked this question more than anyone else:

"But my point is: what is interpretation? If you 'interpret' a Beethoven sonata you don't play exactly what is notated on the page in front of your nose. In a certain sense you are interpreting an entire tradition of interpretation already several generations removed from the original. ...

It is certainly possible to have a bad interpretation of a piece which accurately realizes more of the written notes but signally fails to reflect the mental tensions involved in the enterprise [as in Beethoven, I'd editorially add]. ... In any case, as I say, all performance is imprecise in either a positive or a negative sense. ...

Even the most exhaustive notation only offers a semblance of precision.  It is useful for telling us where to put our fingers, say, but even more important is some capability of offering perspectives on the position of the entire piece in a larger context, its connection with other works of the same genre, or of the same composer."

-- Brian Ferneyhough (interview with Philippe Albera), Collected Writings pp. 318-9

This is just a longer-winded version of Aaron's answer above, which I will rephrase thusly: To suggest that "complex" (!) music is impossible to perform accurately is to imply that "non-complex" music of any stripe is performable "accurately" in an unproblematic sense, an untenable assumption; the nature of the performance of instrumental music is such that a literally "accurate" performance of the "Moonlight" sonata is as superhumanly impossible as one of "Time and Motion Study I".

There are oft-cited reproaches to some composers: 'Do they know what they want?', or 'Do they know what the music sounds like?' or 'They should indicate whatever they want in the score'. These all assume a singularity of interpretation which is ridiculous in the context of music of any period - if there's simply meant to be one type of result that the performer diligently achieves at best, why not write for electronics alone? There are to the best of my knowledge five recordings of Ferneyhough's Lemma-Icon-Epigram available at present (soon to be a sixth when my new recording of it comes out); one of these is thoroughly cavalier with rhythms, dynamics, articulations in the text and makes it all into a big wash, but the others I think all try as best as possible to engage with the detail of the text in their own individual way. And they all sound very different. It's not a matter of which one best 'plays what's written'; this music delineates a quite wide range of interpretative possibility as does much else past and present. But critics and many performers alike seem to assume that playing modern music is simply about 'playing the notes', perhaps with the additional proviso of 'making a nice sound', and as such there can simply be a 'definitive performance'. That is no more true of any decent new music than it is of that of any other period.

Just a quick point of clarification there, Ian, since you mentioned Coigitum (though there's no horn in that piece, you must be referring to the slightly later piece Anatomy). At that time (mid-1980s) I was indeed interested in the somewhat problematic idea of creating a particular kind of sound by deliberately placing the music beyond the possibility of accurate or complete realisability, but subsequently I tried instead to find ways of having the music reach that sonic/expressive world much more precisely and flexibly, by composing so to speak "out of" the instrument rather than "against" it. (I know you know all this from first-hand experience!)

I'm not sure I agree that the impossibility in, say, Evryali is "banal."  It's obviously a matter of literal impossibility rather than simply extraordinary difficulty, and is less subtle in that sense, but insofar as the extreme technical situations in that work are of a piece with the context, and grow out of it in a sense necessarily and convincingly, I think the situations are not in fact all that different.

I know you agree, for example, that there is a "wrong way" to confront the impossibilities of Evryali, such as that for which poor Peter Hill was so roundly and publicly pilloried; and the fact that there are options - the fact that anyone plays the piece at all! - means that there is more to it than the phrase "impossible in a banal sense" implies.

While this is true in a sense, I think it's worth pointing out differences of extent and scale. While musicians are not machines, able to play an unflinching beat accurately for an unlimited duration, and there are many parameters to consider in terms of articulation, phrasing, dynamics etc. that aren't always notated in older scores, I would think you'd agree that executing a series of straight 1/8 notes (the opening of the Waldstein Sonata, say) is less mentally or motorically demanding than a series of nested tuplets of constantly varying durations and speeds, or that quarter- and eigth-tones are harder for a violinist or wind player (the latter due to frequently awkward fingerings) to execute precisely than chromatic intervals. One might argue that playing a C natural on a violin also demands 1/8 tone accuracy, but I don't think that's ultimately true in traditional practice, which always compromises such absolutisms. The point is not simply that BF's notation spells out many of the things a traditional score tacitly assumes, but also that the MUSIC makes cognitive demands that Beethoven does not. Beethoven, on the other hand, demands more in the way of interpretation to bring it to life, as the performer of BF's music will probably often be too tied-up mentally to consider the different expressive possibilities of a passage - and much of the rhythmic complexity is concerned with nuances that are carried out more intuitively, or more in accordance with performance conventions, in older music. All the talk of notation re Ferneyhough is a bit of a red herring, in my opinion, as it easily overlooks the fact that the MUSIC is actually hard to play, that it's not simply a matter of complex notation.

I'm sure Evan will agree that it's a form of meta-negational circumscription. 

Personally I'd sooner suggest that performing a C natural on a violin in many contexts, for example in unison with a piano, requires very considerably closer than 1/8 tone accuracy... also that in a passage such as the opening of the Waldstein, maintaining a constant balance over the sequence of chords while shaping the overall dynamic is damn hard work, more so than in a 'Ferneyhovian' context in that in such transparent music the slightest deviation from regularity is immediately perceived.

I would certainly also argue strongly (and from a certain amount of experience) against the notion that "the performer of BF's music will probably often be too tied-up mentally to consider the different expressive possibilities of a passage". I don't think the subject of interpretation is so simple as putting the notes and rhythms (blah blah, in short, the black-and-white demands of the score) in place and then once you've finished you put expression on top of that. Engagement with (not only) Ferneyhough's music involves you (well, it involves me) with expressive choices from the outset - for example, the answer to 'how loud is ff' depends on what the ff is there for and expression can't be factored out of that consideration; exactly how you (well, I) shape the proverbial "nested tuplets" in time depends on many considerations beyond the merely mathematical from the beginning, for example how any one rhythm relates to the rhythms around it.

The famous case of the four Cs in Superscriptio springs to mind: there are two dotted demisemiquavers in a 3/48 bar, then two straight demisemiquavers in a 'normal' bar. (I forget what it is - I don't have it in front of me.) In other words, two dotted demis under a triplet followed by two straight demis; in other other words four equally spaced notes, and they're all staccato. Felix Renggli does indeed manage to phrase them so that the second note sounds as an offbeat and in my opinion that's absolutely right - but it's an expressive and a formal decision as well as a purely mathematical one.