Sunday, 31 March 2013

Review: An Anatomy of Musical Criticism (1968) by Alan Walker





Alan Walker

An Anatomy of Musical Criticism

Chilton, Hardback, 1968.
8vo. xii+114 pp. First Edition. Preface by the author, February 1966 [xi-xii].

Contents

Preface

Part One
A Solution in Search of a Problem

Part Two
Creative Principles

Part Three
A Theory of Unconscious Assimilation

Conclusions

Bibliography
Index

=================================================

The most important landmark common to all of my favourite authors is that they have changed me, my views and opinions, even my character. The British-born, but Canadian-based, musicologist Alan Walker fits the criteria to perfection. His monumental three-volume biography of Franz Liszt, first published between 1983 and 1997, has changed my perception of the Hungarian genius out of recognition. It is by far the most comprehensive, scholarly and perceptive study of Liszt's amazing life and unique personality available in English, and it is most likely that it will remain so for a good many years ahead. If there is any other work of musical biography which combines compulsive readability and stupendous research is so fine a way, I have yet to read it. Alan Walker (born in 1930) has repeated the feat with one of Liszt's famous pupils, Hans Von Bülow: A Life and Times (2009, I wish he would tackle Carl Tausig too, to say nothing of Richard Wagner) and he has also published a wonderful collection of essays, Reflections on Liszt (2005), that make an excellent companion volume to his magisterial biography.

Since I owe Alan Walker so much, the least I can do is to read some of his earliest and most obscure books. And this is where An Anatomy of Musical Criticism comes into play, for it was first published long before Mr Walker had even started his exhausting research on Franz Liszt, let alone published the first volume of his biography.

In book form or in the online version of the famous magazine Gramophone, I have read quite a bit of musical criticism. Except for the writings of two great men – Harold Schonberg and Bernard Shaw – I have found it of little value, and even this is often obscured by severely technical language that's all Greek to the layman. For the most part musical criticism is indeed perfectly ridiculous. If anything, I am vastly amused by those fellows in the Gramophone. As a general rule, they put on gigantic airs and attach an astonishing importance to themselves; they are often stupendously presumptuous, pretentious and condescending. I am not sure what's worse: if they sincerely believe the inane stuff they write or if they deliberately write nonsense for their own perverse amusement. Certainly I am deeply sorry for everybody who takes these guys seriously and let them interfere with his musical experiences. He misses the best music can offer.

So it is hardly surprising that, my great admiration for Alan Walker notwithstanding, I have been sceptical about An Anatomy of Musical Criticism. I should have known better. Despite its forbidding title, it makes an excellent read packed with stimulating observations. Mr Walker's view of musical criticism strives for complete objectivity and thus it is refreshingly different than the highly personal approach of Shaw or Schonberg. Significantly, however, all three writers share exemplary prose – perfectly lucid and beautifully written, with regular flashes of wit and wisdom – and formidable argumentation which completely excuses the obvious fact that all of them are strenuously opinionated. Mr Walker, of course, is not nearly as witty as Mr Schonberg, to say nothing of Shaw, but this is to be expected considering his scholarly approach. Nevertheless, his writing is often enlivened by charming humour and – this is very important – no matter how much I disagree with some of his notions, I cannot possibly be angry with him.

Before I go into appalling detail about the contents of this lovely book, one serious caveat to be kept in mind: the whole second part makes a truly tough read for the layman. The reason is simple and, I have to say, to be fully expected: the chapter is full of numerous musical examples, no fewer than 99 of them. They constitute at least half of its volume and range from piano pieces to full orchestral scores: from Chopin's famous (and stupidly nicknamed) ''Heroic'' polonaise to Wagner's prelude to Parsifal, his last music drama. I daresay this is inevitable and I really am the wrong person to peruse this book; a trained musician, or anybody who can read music fluently, will surely gain a lot more from it than I possibly could. Still, even for the layman these pages are not entirely unreadable for they do contain a number of interesting points that can be understood by everybody who is able to read. Besides, the other two parts are perfectly free from musical examples.

What attaches a peculiar fascination to this book is what shines on every page of it, and this is Mr Walker's genuine passion for music; not something one would expect in such rigidly intellectual approach. As for the purpose of his book, I can't do better than quote extensively from his fascinating preface:

The practice of criticism boils down to one thing: making value judgements. The theory of criticism, therefore, boils down to one thing also: explaining them. If you formulate a theory of criticism, it is not enough to know that one work is a masterpiece and another a mediocrity. You must also explain why they are different. It is on this issue that my book takes a stand.
[...]
Faced with such extreme opposites as Mozart and Schonberg, Beethoven and Stravinsky, one is hardly encouraged to postulate the existence of constant, creative principles which all these great masters have followed. Each seems to pursue his own arbitrary path. Yet I believe this to be an illusion. Masters do not write masterpieces by accident. Nor do we recognise them by accident. Far from being arbitrary, I believe that masterpieces unfold according to timeless, creative principles, that they would not be masterpieces unless they did, and that it is the chief function of any theory of musical criticism to tell us what these principles are. This is the task I have set myself, although the present book amounts to no more than a survey of the basic issues.

Rather tantalising! Further in these two remarkable pages, Alan Walker demolishes some notions of his colleagues (Hadow, Calvocoressi, Newman) who have tried before him to provide musical criticism with such sound theoretical basis. He is convinced that they were terribly wrong chiefly on two points, and he tries to show that the very opposites are true, namely that: 1) there is no such thing as a critic with standards; and 2) the critic must possess value-judgement before he even starts criticising. Never a man to mince words, Mr Walker bluntly states that, as regards the first point, there is music with standards but certainly no such critics, and that the opposite notion of the second point has been ''a major disaster for criticism.'' Then he makes the startling statement that one of the main paradoxes that he proposes to explore in his book is that criticism actually only explains what one already knows on intuitive level; in other words, you needn't prove that Beethoven's Eroica is a masterpiece to know that it is, for ''its mastery is self-evident'', but the much more important question is why it is self-evident (his italics).

Yes, I agree, the subject is thoroughly abstract and monstrously abstruse. It is definitely a difficult stuff to grasp since, quite apart from the fact that it does require solid knowledge of musical theory, it goes deep into human psychology and there is nothing more arbitrary than that. Yet there is something really inspiring in Mr Walker's ideas about music as – no other word will do – mystery, as something completely self-sufficient, rather esoteric and playing weird games with the subconscious. But he does have a fine point, as every music lover knows only too well. Music is a mystery. It either grips one or it doesn't; and if these conditions are fairly easy to distinguish, it is devilishly difficult to say why a certain piece of music – often in a certain performance indeed! – gives you a unique thrill of excitement, whereas another work – or even the same work in another performance! – gives you a strong nausea. At any rate, Mr Walker offers a fascinating journey that may well prove worthwhile in the end:

We sleepwalk our way through music observing, often with somnambulistic certainty, what is good and what is not. But we still know next to nothing about the way in which we do it. Perhaps, it may be objected, we do not need to; a theory about our musical reactions will not necessarily make us better critics. True; but the crucial function of theory is not to improve practice: it is to explain it. Its great task is to bring into sharp focus the principles behind musical communication.

This quote comes from the first paragraph of the first chapter, which is indeed by far the most important one in the book. Here Mr Walker explains the principles of his theory concisely, with rare subtlety and beautiful lucidity. He separates the chapter into five main sections:

(i) ''A Priori'' Principles
(ii) Creative Determinism
(iii) Methods of Comparison
(iv) Means versus Ends
(v) Vox populi, vox Dei

Each of these five parts does contain tons of compelling reflections. I make no apology for the extensive quotation.

''A Priori'' Principles. In one of his remarkable footnotes Mr Walker tells us that he uses the term ''a priori'' in its strict logical sense, namely meaning ''from cause to effect'', not as ''arbitrary'' as it is sometimes misused by musical theorists. I can't for the life of me put it better than the author did:

It is often asserted that a theory of criticism based upon an acceptance of a priori principles is doomed to failure. My book was written in the belief that such an assertion is almost certainly wrong. A masterpiece is not a masterpiece by chance. Neither, for that matter, is a mediocrity. Both are symptomatic of deep, far-reaching principles; music's fate is sealed internally, as it were. When these principles function, you have a masterpiece; when they malfunction, you do not.

Another landmark of my favourite writers is that it's always a pleasure to disagree with them. So let me immediately take issue with Mr Walker's statement here. For one thing, I think he makes a little too sharp a distinction between masterpiece and mediocrity, putting the former on a kind of pedestal and dismissing the latter contemptuously. Anyway, he is convinced, not only that these ''a priori principles'' do exist, but they actually are ''immutable laws to which all genius gives unconscious utterance.'' There are three consequences of extreme importance that emerge from such an assumption. Mr Walker sets them in his usually straightforward manner; as we shall see presently he is not in the least as dogmatic as he looks like:

First: cutting across all musical barriers – light, serious, chamber, orchestral, vocal, etc.; cutting across all historical styles; cutting across all shades of opinion, all varieties of taste, there runs a simple line of division. On one side fall those works which express music's timeless, creative principles; on the other fall those which do not. The division is crucial. It is one of my deepest convictions that it splits the world's best music from the world's worst.

Second: ''taste'' is not a tool of criticism but a symptom of people. Tastes change. A switch in taste must, logically, involve untruth. How can today's masterpieces become tomorrow's mediocrities? A masterpiece either is, or it is not. You cannot have it both ways.

Finally, there are no critics with principles. There are only works with principles. Criticism does not settle music's fate; it observes music's fate. An important concept now emerges.

Before tackling this ''important concept'', I should like to quote a footnote which Mr Walker uses as a very amusing illustration which parallels the job of criticism, namely to observe music's fate:

The story goes that a man once went to an exhibition of paintings. He walked round all the galleries paying particular attention to the old masters. As he left, an attendant said to him: ''I hope you enjoyed our collection of paintings, sir''. ''Not bad'', came the reply, ''but I didn't like your old masters.'' The attendant paused. ''With respect, sir'', he said, ''it isn't the old masters who are on trial here''. There is no need to labour the parallel.

Creative Determinism. This, of course, is the important concept Mr Walker mentioned in the end of the previous paragraph. He starts with the curious notion of music as an end in itself, its intrinsic value being completely independent from any external factors. That seems a little high-handed to me. Can one separate a musical work from the historical period in which it was created or from the personality of its creator? Is it worthwhile doing that? Would Eroica have been hailed as a masterpiece had it not been created in the very beginning of the XIX century as the first flame of the newly born Romanticism? I doubt it. Yet Mr Walker does have a point, the same one I remarked on earlier. Every person susceptible to music, by which I mean listening to music not for relaxation or fun but from internal compulsion, knows that knowledge about the life and times of a composer whose work he likes can, and does, increase his appreciation. But one must first like the music, to say the very least, and this early stage, I think, is the one which Mr Walker set out to explore. His own philosophy as far as musical criticism goes – creative determinism – is certainly provocative and controversial. But it is compelling and stimulating, too. Accepting that music is ''autonomous'', Mr Walker tells us, the consequences are threefold (all italics are his):

(1) A musical communication is complete; it refers to nothing outside itself. Mendelssohn was once asked the meaning of some of his Songs without Words. His reply summarises an entire philosophy.

There is so much talk about music, and yet so little is said... Words seem to me ambiguous, so vague, so easily misunderstood in comparison with genuine music... The thoughts expressed to me by music that I love are not too indefinite to be put into words, but on the contrary, too definite.

The totally musical nature of musical communication is something I want to stress. Music is non-conceptual. It neither requires nor demands an ''explanation''. It is purely musical truth which can be comprehended on a purely musical level.

(2) If music is autonomous all our knowledge about it must flow out of our experience of it. I do not think it is sufficiently realised that there is no valid theoretical concept in the entire history of music which did not first emerge as an intuitive part of creative practice. Musical theory is always wise after the creative event:

Robert Craft: ''What is theory in musical composition?''
Stravinsky: ''Hindsight. It doesn't exist. There are compositions from which it is deduced. Or, if this isn't quite true, it has a by-product existence that is powerless to create or even to justify.''

(3) Unless you already understand music, a theory about how it works is beside the point. Knowledge without understanding is a curse of our age. I know this is heresy in some circles. The entire field of musical education, for example, is slanted towards the belief that the ''way in'' is through knowledge – theory, analysis, history, etc. Modern psychology, however, shows that this is really the ''way out''.

[...]

The philosophy I am expounding, creative determinism, leads inevitably to a definition of criticism. It is a definition that I shall follow rigorously throughout my book. Criticism is the rationalisation of intuitive, musical experience. I maintain that you cannot criticise anything of which you have no previous intuitive understanding, for your value-judgment is already contained in such understanding. Nobody postpones responding to music until he has set it against a critical yardstick. We respond first, criticise last. This is my basic objection to the so-called ''scientific'' approach to criticism. It encourages the notion that we do not know whether music is good or bad until we have subjected it to an ''objective'' test. But the whole point is that we do know. As soon as we have taken that first intuitive step towards a work, we know instinctively whether we are dealing with something trivial or something great. As for the ''criticism'' itself, this is merely a rationalisation of something we already know to be true on an intuitive level. The real dilemma facing most critics is not that they cannot recognise musical quality without yardstick; it is that they can, but do not know how they can.

[...]

The notion that stylistic developments can be explained in terms of historical ''pressures'' – political, social, economic, etc. – the bread and butter of the history books, is misleading. The case against it is really very simple. There are no successful stylistic changes in music which are not, at the same time, artistically necessary. I find this fact impressive. If you wish to explain a musical event, you do not have to go outside music to do it. The history of music lies in music. The historical background against which music unfolds, I submit, plays a passive, not an active, role in music's development. That is to say, history merely offers an ever-changing series of alternatives – political, social, economic – through which music might, or might not, develop. The final choice is, I maintain, always musically, always creatively determined.

[...]

In the light of the foregoing discussion, I should like to consider two aspects of critical practice which seem, to me, fallacious.

But before we consider these too, let's look at the above quote again. Lots of food for reflection. Mendelssohn and Stravinsky hit the nail on the head with mighty force, to begin with. Mr Walker's notion about music's total independence of history is rather interesting, but I wish he'd given us several examples. His philosophy is highly appealing for it fits perfectly with mine: everybody is his own best critic. In other words, if you do know/feel that you love certain music, you may well send all those who don't to mind their own business. I do not know if Mr Walker would agree with me that the value of this ''rationalisation'' is purely personal and, a friendly pun intended, autonomous, but this is beside the point.

The notion that comparisons in criticism are evil is more than sensible: it is brilliant. This appalling practice is what, to my mind, makes so much of art criticism – music, literature, painting, you name it – simply detestable; to say nothing of critics who do inflict their opinions on you as if they were the only right ones in the world. ''Knowledge without understanding is a curse of our age.'' This is frightfully true, much more so today than it must have been in the 1960s. Today, with Internet and Wikipedia, knowledge is devastatingly easy to be acquired. But understanding – now, that is quite a different matter. I might even go further and suggest that it’s only information that is easy to obtain today. Knowledge is one level above that – and understanding is what but few of the knowledgeable fellows manage to achieve.

The two fallacious notions – disastrously so, if I may add – of musical criticism are points three and four above: ''Methods of Comparison'' and ''Means versus Ends''. In short, the former refers to the vastly misguided passion for yardsticks and the latter is even sillier for it implies that technique may have anything to do with the value of the result, and of course it can't and it doesn't. I hardly need to add that both of these anathemas have pervaded all art criticism. These are problems of paramount importance and Mr Walker has explained, and demolished, them stupendously well, to say nothing of his elegant witticism:

I have never understood the high premium some critics attach to comparison. It has always seemed to me unreasonable to use one composer as a rod with which to beat another – for this is what ''comparison'' invariably amounts to. Work A is neither good nor bad because it is like work B. For the question then arises: why is work B considered worthy as a critical yardstick? Because it is like C which is also considered worthy? This is a charmed circle from which escape is difficult. Comparison brings us back to the all-important question: Where is music's value to be found? We are faced with a simple choice. Music's value is either intrinsic – in which case we shall know about it without our ''yardsticks''; or it is extrinsic – in which case we shall arrive at the embarrassing conclusion that a great work of art is ''great'' through no fault of its own.

[This is a very powerful argument very powerfully expressed; a real gem. And so is the next one, and a great fun too.]

A particularly invidious form of comparison arises when critics appoint themselves to the rank of H. M. Customs and Excise officers whose function it is to spot composers smuggling contraband ideas from one work to another. To ask a composer if he has anything to declare while he is busily unrolling his music to public view is not a very intelligent question. Each act of composition is a declaration. If it did not owe something to somebody it would be intelligible to nobody. Elgar may be said to have ''smuggled'' the closing pages of Tristan into the final bars of his own Second Symphony. But the comparison is so obvious only a bad critic would make it; and only a fool would ''devalue'' the Elgar as a consequence. The likeness sheds no light whatsoever on the respective ''value'' of either work. The way pieces resemble each other is the least interesting thing about them. It is one of musical criticism's blind alleys.

[...]

Another is the way in which means are sometimes confused with ends. Means are precompositional; they are the concern of the composer. The critic's concern on the other hand, is the creative result. We put the cart before the horse when we censure a composer for employing a technique which, for one reason or another, we happen to dislike. By itself, a technique is neither good nor bad. It is incapable of receiving censure until it has fulfilled itself in a creative result.

Means are continually gaining the upper hand over ends in criticism. We are always being asked to admire a way of doing, not what is done. Witness the so called ''experimental masterpiece'' which commentators are apt to refer to with monotonous regularity. The fact is, there can be no such thing as an experimental masterpiece and it is a critical blunder to suppose that it can. An experiment is a process – a means towards an end; but a masterpiece is an end. For the rest, if the experiment comes off it is no longer ''experimental''; if it does not come off the result can hardly be a ''masterpiece.''

Transfixing passages! They do give me a considerable pause. I may mention in passing that in a footnote Mr Walker mentions as an example of an ''experimental masterpiece'' which has had that ''meaningless label'' for many years Sibelius' Fourth Symphony. And have you never seen the charming fellow who insists on showering you with comparisons between composers, painters and writers, discussing with subtlety and insight the technique they used to create their masterpieces. I have. This is called ''culture'' and showing it off is the favourite pastime of these ''charming fellows'' who can be described with the generic name ''intellectual snob''.

Vox populi, vox Dei. The last of Mr Walker's five maxims is the most dangerous one – and he knows it. This, of course, is the elusive and so often misunderstood correlation between recognition and mastery, or to put it in a rather more vulgar way: between commercial and artistic value. It's a conundrum as old as art itself. Does popularity prove that a work is a masterpiece or not? That is the question. It's a very complicated one. Popularity is not so easy to define, to begin with; what do you mean, really: national, continental, worldwide popularity, one that transcends different cultures and religions, or something else? Then there is the question of time. Popularity means nothing if it doesn't last sufficiently long. But how long? Most importantly, perhaps, a work of art must first be given the opportunity to survive the fickle human nature and the severe test of time. In the case of music, fine performances of a work must be heard often enough for a considerable time: if it still fails, then surely it is no masterpiece; if it survives, then it is. But there are complications here, too. The quality of performance and the duration of the survival time are vastly subjective. Yet one thing is sure: no masterpiece can gain any recognition if it is dismally performed or not at all. This, I maintain, is the reason for Liszt's symphonic poems to languish into obscurity century and a half after they were written. Instead of rambling like that, I may as well give the word to Mr Walker:

I should now like to put forward a hypothesis which is of some importance to my theory. The potential aim of a great composer is to communicate a universal, artistic truth. [...] Immediately, certain objections arise. There are many works about which there has never been universal agreement. [...] Yet there are two reasons why I want to retain the hypothesis. First: its operative word is ''potential''. I am not maintaining that a great musical communication is universal, only that it may become so. Second: overwhelming as such objections to it may be, they only represent one side of the case. It is equally certain that there are works about which no one disputes.

[Here, again, I would have loved to see some examples on both sides of the case. The first point is rather evasive. It works both ways: if a communication has not become universal for decades or even centuries, maybe the work in question is no masterpiece and the theory which claims the opposite is false. The second point is controversial: even the most universally renowned works have had their detractors. They may be in the minority, those hard-headed iconoclasts, but their eminence is usually impressive and gives their voice a certain prominence.]

As for the question ''What does music communicate?'' no theory of criticism can avoid it. Music is a vehicle for transferring a psychological situation. A listener who responds to music does so because he has unconsciously identified himself with it; [...] No musician needs to be told that the intuitive musical experience is a vehicle of truth far superior to that of rational thought.

[I venture to claim the above is true of every serious listener, too. In a fine footnote Mr Walker mentions what he discusses more fully in Part Three, namely that the relationship composer-listener is based on powerful unconscious factors that are most probably beyond human understanding.]

Why do masterpieces survive across the ages? Surely because musicians are in general agreement that they are masterpieces. Survival is a symptom of profound and widespread recognition. Critics who cannot accept the notion that a universal response is symptomatic of musical value often counter with the argument that, if it were true, the most popular works would be the best – a conclusion nobody accepts. Actually, popularity is not significant because it increases music's value but because it confirms it.

[Rather muddled that, or evasive is a better word perhaps, either is a rarity in Alan Walker's writing. So, is popularity proportional to value or not? If not, as seems to be the case, doesn’t this invalidate the whole argument? How do we measure popularity? Can we? Should we? Is there some kind of threshold above which the value of a work is ''confirmed''? Whatever. The next paragraph redeems the deficiencies of the previous one. Here Mr Walker is positively profound:]

Vox populi, vox Dei is a dangerous doctrine to espouse in aesthetics. Yet it is equally dangerous to ignore it. The fact is, the vox populi is often a symptom of the vox Dei. To dismiss it arbitrarily is the essence of snobbery. To regard it as evidence is the beginning of understanding.

[...]

Assuming that he gets a hearing, history shows that a master will never remain unrecognised for ever. How great a master he is, and how widespread the recognition accorded to him, these are subsidiary matters which should not deflect us from our main objective. A theory of criticism must explain what it is that constitutes the difference between the two extremes of mastery and mediocrity.

Having set the stage brilliantly, Alan Walker then goes on to illustrate how he applies his principles to some musical works which he, apparently, considers masterpieces, unless he explicitly states the opposite as he does in a few cases. Unfortunately, as I have already said, Part Two of this little book is largely incomprehensible for me. But since Mr Walker mentions a number of my favourite works, and waxes lyrical about some that are not but I have some familiarity with, I think I may allow myself several cautious remarks.

Sometimes I really am sorry that I can't read music and in such cases I have to satisfy myself with a very lame analysis. But I absolutely refuse to believe that the ability to read music, or any other special knowledge of music theory, is in any way essential for a true understanding of music. The musical perception of a trained musician is unimaginably different than that of a layman, that's for sure, but who is to say which is the truer or more important one? I dare not even think about it.

One of the most wonderful things about Part Two of Mr Walker's book is that he spends considerable time on Tchaikovsky's Fifth symphony and, to a lesser extent, his Violin Concerto, overture-fantasy Romeo and Juliet and the Fourth symphony, all of them outstanding works that are very close to my heart. One of Mr Walker's most curious footnotes is also dedicated to Tchaikovsky, and I of course cannot resist quoting it:

One day, when the history of composition, the history of musical ideas comes to be written (as opposed to the conventional, socially inspired histories of music), Tchaikovsky will emerge as a leading figure in the development of the contrast-potential of large-scale musical structure.

I wish Alan Walker had elaborated more on this; he did, indeed, though indirectly: in the same footnote he directs the reader to his analysis of Tchaikovsky's Fourth symphony in his previous book, A Study in Musical Analysis. Yet even for the perfect layman, he can be illuminating. Why does Tchaikovsky's Fifth Symphony have a waltz as a third movement? This is not without historical precedent (Berlioz's Fantastique, for instance), but it is surely very unusual. So is the inclusion of a march in a symphony, and that's exactly what Tchaikovsky did in his Pathetique. Alan Walker builds a case that these at first glance bizarre choices actually make a good deal of sense when the works in question are carefully examined. I am sorry – yet again – that his explanation is in the form of musical examples and thus well beyond me. (Still, why does Beethoven's Eight have a minuet as third movement? Beethoven himself had made this dance obsolete in a symphony, replacing it with scherzo, long before he came to compose that work.)

Speaking broadly, in the second part of the book Alan Walkers asks point blank many questions of paramount importance. I am indeed surprised that I had given them almost no thought before, especially considering the vast number of examples I am familiar with. Why is a particular composition written for a particular medium? Does change of the medium matter? Why is every masterpiece constructed of such and such movements and why the themes in each of them are arranged in that way? Does change in either matter? The short answer is ''Yes, they do matter'', the medium, the themes and their order. Alas, I can do more here than point out at some examples.

As usual, the exceptions are the most fascinating part of the story. Though Mr Walker is convinced that the right medium is essential, he is perfectly aware that every medium has its limitations and the great composer is the first one, when his inspiration dictates it, to write against the medium. Since both instruments and performers evolve, this leads directly to the amusing historical paradox that some works were condemned ''unplayable'' at the time of their composition, yet today they are extremely popular and played by a good many people. The most famous example here is Tchaikovsky's Violin concerto which became ''playable'' just two years after Leopold Auer, the distinguished Russian violinist to whom the work was dedicated, had declared it to be ''unplayable''.

Sometimes the recognition was much slower, in a few cases it has yet to come. Here Mr Walker gives as an example Beethoven's famous, or notorious, Hammerklavier, one of his late piano sonatas which, nearly a hundred and fifty years after its composition (at the time of Mr Walker's writing), still posed severe technical problems even to accomplished virtuosos. Some parts of the final fugue are simply physically impossible to play, something Mr Walker memorably describes as Beethoven composing beyond the medium. The author is convinced that one day the time will come when pianists will reach the level of mastery required to play faultlessly the Hammerklavier. I don't know if that day has come already, but I wonder whether the truth lies more in Mr Walker's speculation or more in the possibility of Beethoven, towering genius as he was, being carried away in the realm of the nonsensical.

Mr Walker's notion that all masterpieces grow from diversification of one main theme may sound exotic but I daresay he knows what he's talking about; it may well be that the aspect is comprehensible only to trained musicians. Certainly, I too find it rather interesting that Brahms' Second Symphony is his longest yet the one composed in the shortest time. Despite Mr Walker's touching descriptions of this work as ''conceived whole, rather than made whole'' and that there is ''impressive totality about it'', I still consider Brahms' Second by far the dullest of his four symphonies. Just another example how different a view – just different, nothing more – a musician and a layman may have of a single work. All the same, Mr Walker remains convinced that every musical masterpiece, no matter how diverse on the surface, exhibits a profound unity in the background. He backs this up with an amazing quote from a letter by Tchaikovsky himself:

We begin to understand what Tchaikovsky meant when, discussing symphonic form on one occasion, he remarked: ''The details can be manipulated as freely as one chooses according to the natural development of the musical idea'' (my italics). I should like to draw a corollary from Tchaikovsky's observation. All great music is variations; to compose is to diversify! In a masterpiece, no event will take place for which there is no precedent. Every direction the music follows is predetermined by the ''natural development'' of the basic idea.

Did you spot the most important part in Tchaikovsky's quote? It is the singular form of ''idea''. It is from one idea that a masterpiece grows by constant diversification. This sounds positively fanciful to me, but I am pretty sure people who can read music may well find some not unconvincing evidence in Mr Walker's examples. Amusingly, he drew this ''corollary'' while discussing Tchaikovsky's Fifth Symphony which, as every admirer of the Russian composer knows perfectly well, does have a main theme, a motto, a leitmotiv, call it what you like, that appears in all four parts of the work, sometimes heavily disguised. Of course this is so obvious, even to the layman, that it is not at all what Mr Walker means when he talks about unity. He goes far beyond that, to the very foundations of a musical work, and there, alas, I cannot follow.

The most obvious consequence of this principle is that ''there are no ''free'' passages in music – except bad ones'', but there also are some rather far-reaching implications not without interest. Mr Walker, for instance, argues that, not only does it matter which themes a composer chooses to build a dramatic contrast in a movement, but it does also matter in what order he arranges them. His example is absolutely charming: he re-writes Mozart, no less. He takes the entire exposition of the first part of the Sonata in F major (K. 332), which contains no fewer than seven ''sharply contrasted themes'', and re-distributes them without omitting a single note. Yet the difference between the two versions ''might almost be called spectacular'' and here comes the most enchanting moment:

If you do not already know Mozart's original, I do not think you will reject my version, although I hear no chance of your accepting it as the work of a genius. It could have been written by a minor eighteenth-century composer.

Somewhat more accessible example about the importance of the order in which musical themes are organised is Mr Walker's re-retelling how Beethoven's Diabelli Variations came to be written. Now, I by no means subscribe to his opinion that this is ''the finest set of variations ever written''; even from Beethoven's own sets of variations I would much sooner listen to the more concise and powerful 32 Variations in C minor, though they are regarded contemptuously by more or less everybody, Beethoven himself included as he didn't give them even an opus number. But this is neither here nor there. The story is highly entertaining and has been told so many times that it will surely bear another repetition. And I certainly cannot better Alan Walker's version, to say nothing of his using it as an illustration:

In 1822 the composer and publisher Diabelli invited fifty-one of the most prominent musicians then living in Austria and Germany to contribute a variation each on a Waltz-theme he himself had written. Everyone accepted – Schubert, Moscheles, Czerny and the eleven-years-old Liszt – except Beethoven, who refused. Yet sometime afterwards Beethoven did, in fact, turn his attention to Diabelli's theme and his ''contribution'' flowered into the finest set of variations ever written. A comparison between these two sets is illuminating. ''Diabelli's'' set, by fifty different composers, is a curiosity of musical history. The chronology of the variations is governed not by any musical logic but by the alphabetic order of the composers' names! The musical results are ludicrous. Beethoven's set, on the other hand, unfolds a chronology which sounds absolutely right. I would stress that both sets are unified; they are, after all, based on a single theme. But the one remains a mediocrity while the other is a sublime masterpiece. One contributory factor to this distinction, I submit, is that the principle of contrast distribution operates in one, but not in the other. ''Diabelli's'' variations are thrown together casually. Beethoven's emerge causally.

The last major question that Alan Walker discusses in Part Two is also the most elusive one. This is the applying of what is known in logic as the ''law of parsimony'', which is in fact a ''law of economy'' and translated in musical terms means to compose music that contains neither fewer nor more notes than it is absolutely necessary. This may seem like a simple affair but, in fact, it is devilishly difficult; anybody who has ever tried what I am trying at this very moment, namely to write something not altogether unintelligible, knows only too well that simplicity is a really hard thing to achieve.

Starting from this rather unassuming point, Mr Walker goes beyond even the wildest dreams of the musical layman, delving deeper and deeper into complex matters like difference between an idea and its utterance, revisions of own works and transcription of others', ''co-extension'' between form and content, the principle of audibility and, perhaps most important of all, the creative process as function of the unconscious. This, indeed, may be regarded as the main topic of the whole book: it pervades the pages like a Wagnerian leitmotif of epic grandeur and disturbing power. The author is absolutely convinced that musical composition is largely – if not entirely! – an intuitive process, that great music cannot communicate universally, and thus can't be great, if it does not affect profoundly the subconscious of both performers and listeners. It is a kind of spooky notion, but extremely appealing to speculate about. Here Mr Walker resorts as often to psychology as he does to musical examples, and his prose is so sublimely succinct, that I can – yet again – do worse that resort to shameless quoting with but occasional annotations.

Good musical argument is likewise subject to a law of economy. The reason may be summarised in two words: maximum comprehension. To introduce into musical communication either more material or less than is actually required to convey the meaning behind that communication leads to ambiguity and distortion

[Here, in one of his wonderful footnotes, which really rival Gibbon's in terms of additional dimensions to the main text, Alan Walker mentions that he uses ''economy'' in the strictest sense of the word: ''Economy not only involves the elimination of surplus: it also involves the preservation of essence. Too little is as uneconomical as too much.'']

Composers through the ages have been acutely aware of the need to express themselves with unremitting clarity. Some succeed; many fail. Josef Suk once showed Brahms one of his youthful quartets. That Brahms was well aware of the function of economy is shown by his reply, characteristically pointed. ''The essential thing is that every note should be in its place. I can't do that – nor can Dvorak – and you, of course, least of all.''

[Brahms was quite a character, wasn't he?]

Beethoven rarely succeeded in correctly articulating his ideas the first time. His Sketch Books put his creative process under a magnifying glass. They reveal an obsessive search for a true formulation. To ''fix'' the idea, to define it, to pin it down – that Beethoven knew this was the essence came out strongly, on one occasion when, after hearing the Funeral March from Paoer's opera Achilles, he observed ''I must compose that!''

Now the general law of economy which all these composers were hinting at, and musicians of all times have had an unconscious awareness, divides into a group of subsidiary principles which I should like to define separately.

First: the principle of identity between the idea and the utterance. In a masterpiece, the ''idea'' and its ''utterance'' are one. To put this proposition another way: it is a function of creative mastery to cloth an idea with a certain number of notes; no more, yet no less. I derive this hypothesis from the simple observation that you distort a masterpiece the moment you add to it or subtract from it.
[...]
Few composers formulate their ideas precisely the first time. They are constantly re-adjusting and re-defining until their ''outer'' notation matches their ''inner'' vision. This habit of sketching offers us an ideal starting-point. Indeed, it brings into sharp focus the distinction I have already made between an ''idea'' and its ''utterance''. Beethoven's sketches are more fascinating than most – partly because they reflect every stage of his creative process (often, even the sketches have sketches), and partly because the contrast between the initial, unpromising formulation and the final, inspired result is so spectacular. Every detail is exposed – like watching a film in slow motion.
[...]
It is clear that sketches are of limited use to us in demonstrating the existence of the principle of identity between idea and utterance. They can furnish an illuminating starting point; yet their great drawback is that the evidence they offer is precompositional. Sketches are not part of the work at all; they are means, not ends, and I only introduced them by way of furnishing an analogy. Let us push forward the argument a stage by returning to the finished composition itself.

There is a field of musical creation well suited to a demonstration of this principle. I refer to revisions. Most composers revise. They may return to ''completed'' works, often after an interval of many years, and re-compose them. Nothing could serve better to illustrate the presence of economic tendencies in creative genius. Revision is an acknowledgement by the composer himself that what he actually wrote is not what he actually meant. Revision is an act of self-criticism. It aims to re-define, and therefore enhance, a musical communication.

[Here Mr Walker inserts a footnote which is well worth quoting in full:]

The very word ''revision'' implies that a composer has a revision, a new vision of the work. True, what prompts him to revise may not always seem like a new vision. External conditions arise which sometimes force a composer to adapt his work. One thinks of Mozart re-composing opera arias owing to last-minute cast changes. But this is not, strictly speaking, ''revision'' at all. It is the creation of genuine, musical alternatives which arise from a sense of expediency; and frequent as it is, it is not nearly so common as that caused by creative dissatisfaction. It is this last activity, revisions which are artistically determined, which discloses an unconscious drive towards economy.

[Examples here include Liszt's simplifying in 1851 the early version of his Transcendental Studies (1839), ostensibly to make them more accessible but really, Mr Walker believes, to discard lots of surplus material and thus express much better his original ideas.]

It could be held that these examples of revisions are open to precisely the same objection as my earlier examples of sketches. That is to say, the composer himself has accomplished the revision. What we have witnessed is an act of self-criticism – a very different thing from criticism! All this is true. Let me push the argument to its final stage.

Composers are not only self-critics. They are critics, too. When Beethoven said of Paoer's Funeral March ''I must compose that!'', and when Brahms told Suk that ''the essential thing is that every note should be in its place'', they were defining a tool of criticism; they were comparing the real music, the music behind the notes, with the way in which it had been wrongly expressed. Such observations can sometimes lead to highly productive results. Composers will sometimes re-compose composers. This can be criticism on a grand, creative scale. So devastatingly effective can it be, the original composition can be totally replaced by the new version which succeeds where the other failed. History is full of examples.

[As a striking example, among others, Mr Walker discusses Liszt's Paganini Studies (1838) which, significantly, he also revised and simplified later.]

A comparison between the originals and Liszt's adaptations is revealing. Like Bach, Liszt does not hesitate to re-compose the originals. I do not mean he merely adapts to the new medium. He does this and much more. He adds new figurations, transforms the harmony and even adds extra bars. One of the best transcriptions in the set is the second, in E flat major (number 17 in the Paganini collection).

[Two musical examples unfortunately impossible to include here follow.]

One is tempted to say of Liszt's version that it transformed the original not out of, but into, recognition. Posterity would scarcely remember this Paganini caprice, if Liszt had not shown a creative interest in it. Hearing the two versions one after the other is like seeing a blurred image suddenly leap into sharp focus.

[It is very pleasant to see Mr Walker referring to Liszt's works, and some of his most virtuosic transcriptions at that, with so much sympathy in the relentlessly anti-Lisztian climate of the 1960s, long before he started his work on what is now Liszt's definitive biography.]

There are other directions, too, in which the creative tendency toward economy operates. One of the more important manifests itself as a principle of Co-extension between form and content. Debussy once told Satie that he should pay more attention to form. Satie responded by writing his Trois Morceaux en forme de Poire (Three pieces in the form of a Pear)! There was a right idea behind Debussy's advice, just as there was a right idea behind Satie's response. Debussy was regarding form as an aim of musical content; whereas Satie was trying to show that form is a result of musical content. Now many musicians have observed that the distinction between ''form'' and ''content'' is a false one. They rightly point out that you cannot have ''form'' without, at the same time, having ''content'', that the one is the organic result of the other, that it is misleading to talk about form as if it were an empty husk, a mould into which ''content'' is poured. ''Form'' and ''content'' are different aspects of the same thing.
[...]
I contend that when we understand music really well, when it has become a part of us, we also intuitively come to know the potential of its material. It is the non-fulfilment of this potential that enables us to diagnose a ''split'' between content and form, between the distance the music has already covered and the distance it might yet cover, between what the music actually is and what could still become. Why is a masterpiece as long as it is? Text-books on form remain silent. Yet it is of paramount interest to know why music ends when it does. The one sure thing about a masterpiece is that it completes its allotted span with the punctuality of a planet completing its orbit. Music has a certain propensity to unfold a certain distance. If it stops before it has done so, it is incomplete. If it goes on after it has done so, it is over-complete; there is no room in great music for the pleonasm.

[I find this unconvincing; perhaps that is to be expected for the matter is really very abstract. The question is: why do I think most of Liszt's symphonic poems masterpieces and none of Mahler's symphonies anything but monstrous pleonastic orgies? Many would disagree with me, of course, but this is not the point; of course I would think differently if I could like Mahler. The question is: can I ever appreciate any music I don't identify subconsciously with? I don't think I can – nor do I think that it makes any sense trying to. That said, men of Mr Walker's erudition and intellect probably can and so far as theory of musical criticism is concerned he probably does.]

There functions in all great music a principle of audibility. To get across, to make a total aural impact; that is an objective towards which all notes travel. Not all of them arrive.

On a most primitive level, anything in a musical structure which is strictly inaudible is strictly unnecessary: if you cut it out, you cannot hear the difference, anyhow. But this kind of inaudibility is not all that fruitful to study. I know of no great master who is guilty of such gross miscalculations (although I can think of several lesser masters who are).

One of the most striking cases of inaudibility occurs towards the end of Grieg's A minor Piano Concerto, where the theme of finale's central episode returns in full orchestral splendour, and where it obliterates the solo pianist totally and absolutely.

In the concert hall it is an extraordinary effect to see the soloist racing up and down the keyboard, fortissimo apparently without producing any sound. The observation is beyond all question, and I invite anyone to check it for himself the next time he sees the Concerto played on the stage. The conclusion you draw, of course, will depend largely on whether you think that Grieg meant it to be heard. I personally think that he did, and I consider it a serious miscalculation in a work which is otherwise supremely well composed.

[Now Mr Walker does have a point here. In yet another meaningful footnote he reminds us that he refers to live concerts, not recordings where the sound may be so manipulated ''that, paradoxically, the notes come across despite Grieg, rather than because of him.'' Of course he is right about the ''silent'' pianist on the stage, this no one who has ever heard, and seen, the concert performed live will dispute. But such moments occur in many a piano concerto; certainly in Tchaikovsky's and Liszt's Firsts, occasionally even in Brahms and Rachmaninoff the orchestra comes dangerously close to drowning the piano. And if Grieg may be regarded as a ''lesser master'', shall we say the same about titans like Liszt, Tchaikovsky, Brahms and Rachmaninoff?]

[Towards the end of Part Two Mr Walker summarises his philosophy in a beautifully concise and vastly amusing way.]

Buridan, a logical philosopher of the 14th century, was reputed to own a remarkable ass. Placed between two bundles of hay, the animal is said to have correctly deduced that both were equally succulent. It then starved to death because it could discover no rational reason for proceeding in one direction rather than in the other. This astonishing beast was well on the way to becoming a bad music critic. Indeed, asked to decide between two works, many a critic might follow his example; that is to say, having exhausted his intellect weighing one possibility against another, he might be faced with a logical obligation to starve himself into silence. So much the worse for the intellect. So much the worse, too, for musical criticism. I have never understood the somewhat exalted position to which Buridan's ass has been raised by some philosophers. It has always seemed to me fully deserving of its fate. But perhaps this is because my own philosophical position is different. Essentially, that position is a Pragmatic one. Had Buridan's ass been a Pragmatist, it would have consumed both bundles of hay and ''rationalised'' the event later. Likewise criticism. We ''consume'' music, then we ''rationalise'' the experience. It is simply not true that critics wait until they have set a masterpiece against a critical yardstick before they know it is a masterpiece. Criticism does not work like that. Music's value is there a priori; it comes across intuitively as part of the musical communication. You know, you criticise (= rationalise); the converse is unthinkable.

[In other words – in this case, mine – criticism is absolutely useless unless you are of singularly inquisitive turn of mind and want to know why Beethoven's Eroica or Mozart's Jupiter are masterpieces. And what good will this ''rationalisation'' do for you when you have known all time perfectly well, though intuitively, that this is as sure a fact as the endless universe? The answer is obvious.]

[The last chapter from Part Two is called ''The Musical Underworld'' and deals exclusively with musical composition as a function of the unconscious. I start the quotes with two excerpts from an earlier part of the book since they are very relevant to what follows below. After discussing in detail the structure of Tchaikovsky's Fifth Symphony, Mr Walker writes these memorable words which cut straight through the heart of the matter.]

Tchaikovsky, we may be sure, was blissfully unaware of his complex integration. He would no more deliberately compute his thematic organisation in this fashion than a pedestrian, guided only by a desire to get to other side, would work out a geometrical progression before crossing the road. Tchaikovsky was guided only by his creative conscience. Yet for that reason, this sleepwalking genius remains a fascinating study for all musical analysts.

Some musicians might disagree. They might maintain that what is not consciously intended can hardly become part of a meaningful, artistic communication. This is a naive point of view. It is misleading to suppose that what is conscious is meaningful, while what is unconscious is not. Indeed, what is unconscious is, by its nature, dynamic and far-reaching in its creative consequences; all the more likely to be meaningful, then. Composers themselves have made penetrating self-observations which provide a great deal of evidence in favour of unconscious musical organisation. Moreover, from the listening end, we observe far more than we know. We hear unconsciously, too. The notion that the complex demonstrations of unity sometimes revealed by analysis cannot possibly be heard unconsciously receives no support from depth psychology. What cannot be grasped unconsciously, cannot be grasped.

[...]

From time to time, I have observed that the creative process, like the critical, is an intuitive activity. Music cannot communicate, let alone survive, unless it expresses a high degree of unconscious content. The unconscious is the womb of musical creation; all masterpieces are born there.

Such categorical assertions are likely to sound provocative. They impute to the musical unconscious almost uncanny powers of creative organisation. Yet there is a great deal of evidence, from composers themselves, suggesting unconscious sources for their ideas. Many of them have self-confessedly picked their way through the complexities of a composition as if in a sleepwalking trance. Indeed, there are a great many so called ''intellectual'' achievements in music which were arrived at with no more effort than it requires to produce a dream. My comparison is nearer the mark than may be supposed. Composers who dream music are not at all uncommon.

[There are several absolutely amazing examples about the musical unconscious and how it worked brilliantly for composers as diverse as Tartini, Handel, Mahler, Walton, Copland, Brahms, Schubert, Schönberg and Stravinsky. I will limit myself to the most famous example – Wagner – and would like to stress that Mr Walker's attitude is infinitely more sensible than that of many a Wagnerian scholar who have been only too eager to show themselves learned by trying to disprove the episode as fictional. I also quote Mr Walker's extremely perceptive footnote in full.]

The trans-like condition which can accompany the act of musical creation was described on one occasion by Wagner while discussing the initial inspiration of the orchestral prelude of Rhinegold. After entering a ''cataleptic state'' he suddenly felt, he says, as if he were sinking into a mighty flood:

The rush and roar soon took musical shape within my brain as the chord of E flat major surged incessantly in broken chords: these declared themselves as melodic figurations of increasing motion, yet the pure triad of E flat major never changed, but seemed by its steady persistence to impart infinite significance to the element in which I was sinking. I awoke from my half-sleep in terror, feeling as though the waves were now rushing high above my head. I at once recognised that the orchestral prelude to the Rhinegold which for a long time I must have carried about within me, yet had never been able to fix definitely, had at last come to being within me.

I think it would be a mistake to dismiss this kind of testimony – highly coloured though it is. It is of the utmost interest to the student of modern psychology. And Wagner only confirms what many other composers have experienced.

[Footnote.]
To judge from his frequent remarks on the subject, Wagner had an unusually keen insight into the creative process. In Meistersinger, it will be recalled, even the inspiration for the ''Prize Song'' comes to Walter in a dream!

Part Three of Mr Walker's book does not contain a single musical example (though it does have several simple schemes), but it is the most difficult one to grasp. It deals almost exclusively with psychology and it contains a one-page quote, the longest in the book, from Freud's Introductory Lectures on Psycho-Analysis (London, 1952); it is not an accident, indeed, that Freud is the writer with most books in the small, but thoroughly and wisely used, bibliography in the end of the book. But I daresay this is inevitable. Being the least visible and the most intangible of all arts, music plays on your unconscious as nothing else does. Besides, as Mr Walker wisely points out in the beginning of his final chapter:

Let us begin by acknowledging that when music brings home a crop of contradictory reactions it is a symptom of people, not music. It is to psychology that we must turn for some answers.

I confess I am fan neither of Freud and his libido stuff nor of psychology in general, but to dismiss Mr Walker's reflections only because of this prejudice of mine would be foolish beyond measure. After all, he asks a number of questions of paramount importance for everybody seriously interested in music, be he performer or listener. Why a masterpiece is recognised by some people and not by others? Why a seemingly irrelevant matter like the identity of the composer is so important to appreciate a piece of pure music? In a nutshell, Mr Walker presents a case that music is an expression of the repressed unconscious by a composer, therefore the first and by far the most important factor for a listener to appreciate it is to identify unconsciously with it. Music lovers – true music lovers – know perfectly well what this means: music either grips one or it doesn't, and that's that.

Only it's not that simple. Here Mr Walker introduces his concept of ''Moving historical backgrounds'' which leads directly to investigation of the nature of aesthetic experience. This, again, is something that everybody susceptible to art knows, yet many try, stupidly, to discard. Music's value may well be entirely intrinsic but its appreciation by the layman, or by the critic for that matter, certainly is not; there are powerful ''secondary'' factors which are extra-musical. If I understand correctly, this is where the ''moving historical backgrounds'' become really important. Just like one can't quite ignore one's past, not for any other reason but because it has had a marked effect on his personality, nor can one ignore his ''personal musical history'', so to say, or the musical spirit of the times he lives in. As for the complexity of the aesthetic experience, I may safely leave this to Mr Walker:

The fact is, as we have seen, our aesthetic experience of music does not consist only of what we hear. It consists of what we do not hear, too. On the one hand, there is the work itself – the ''primary'' source. On the other, there are all the works by the same composer (and many others historically related to him) – the ''secondary'' source. What we already know of a composer's music conditions the way in which we react to something unfamiliar by him. The aesthetic experience is always compounded of this interplay between the ''primary'' appreciation field and the ''secondary'' appreciation field. As soon as we are told that a work is by (say) Beethoven, this one piece of information is enough to trigger into activity a ''secondary'' field, and which colours it – albeit on an unconscious level. But suppose that we later learn that the work is not by Beethoven at all, that it is by a nonentity. Then its links with the ''secondary'' field are severed. We listen to it in a different ''frame of mind''. And this, too, is the whole case against anonymity.

[Earlier in this chapter Alan Walker gives some vastly amusing examples of this well-known phenomenon: when a work is known to be by a great master, it automatically gains some of his greatness; and when it is known to be by a lesser master, it automatically gains some of his mediocrity; and when there is a mistake, deliberate or not, with the identity of these two fellows, then there is a good deal of fun. Mr Walker calls the guys who play such tricks ''hoaxers'' and, a little further in the chapter, he is rather harsh – rightly – with their attitude:]

All hoaxers go wrong in ignoring this dual aspect of the aesthetic experience. They imagine that aesthetic reactions are based exclusively on the observed work of art, and that when a listener rhapsodises over a work later shown to be a fake, he exposes himself as a fool. It is the hoaxer who is the real fool, however. He overlooks that vast, pre-existent reservoir of positive responses which all of us gradually acquire and then unconsciously project on to new works by great composers. What you know influences what you do not know. Without this unconscious tendency, no hoax would be possible. For the hoaxer creeps in under its cover and exploits it. He robs a work of its true background and fabricates a false one. The listener is easy game; he takes music on trust. The hoaxer does not; his attitude is essentially that of the perverted lawyer who regards everybody as guilty until proved innocent. But musical criticism never has, and never can operate on such principle.

Far from representing a detour, a careful consideration of what is that constitutes a hoax and how exactly it differs from what constitutes ''honest'' musical communication seems, to me, to be an ideal method of disclosing the presence of historical backgrounds, and their function. A historical background can push a listener towards a work, or it can pull him away from one. It is an indispensable factor in fixing the view from which the listener observes the composer. No theory of criticism can afford to neglect it. But what has it to do with the first part of our theory, that of ''unconscious identification''?

Well, it doesn't make much sense to stop quoting just right here. For in the next few paragraphs, subtitled ''A Synthesis'', Mr Walkers brilliantly brings together the two major components of his theory mentioned in the last sentence of the above quote. I daresay there are people who would denounce Part Three, if not the whole book, as trite and superficial; then again, the author himself has never claimed that it is anything more than the very foundations of something he hoped to develop more fully in the future (alas, he never did). Personally, I find his reflections compelling and his arguments stirring, if not always entirely convincing. But let's go back to the point – ''A Synthesis'' – which Mr Walker explains with rare clarity:

At first sight, I appear to have changed horses in mid-stream. The concept of ''moving historical backgrounds'' seems to have little to do with that of ''unconscious identification'' – an aspect of our theory which we apparently abandoned some pages ago. Indeed, these two concepts give the impression of being somewhat contradictory. For if it is historical backgrounds which determine whether or not music shall communicate, why postulate unconscious identification at all? To put the contradiction another way: Was it not a mistake to have proposed unconscious identification as the key to our theory of musical communication in view of the crucial role played by historical backgrounds?

I do not think it was a mistake. Nor do I think there is any contradiction in our theory. What I have called the listener's ''historical background'' is nothing less than the sum total of all the music he has previously identified with. He has no ''historical background'' unless and until he ''identifies'', for this is his only means of acquiring one. The two conflicting ''concepts'' are, in fact, different sides of the same coin. It is by a continual process of unconscious identification, particularly in his formative years, that the musician gradually builds up the historical backgrounds against which he listens – or composes.

[…]

All influence, cultural or otherwise in T. S. Eliot's phrase, ''introduce one to oneself''. Psychologically, ''influence'' is largely what you need to be influenced by in order to develop. It is a deeply mysterious process, but one to which everybody can testify. It is equally true of composers and listeners alike. When a composer ''influences'' a listener he is doing nothing less than revealing a part of that listener's own musical personality to him. And where do the musical principles, the principles I formulated in Part Two, fit into the psychological theory I have just outlined? The answer is that they emerge as rationalisations from within the framework of our positive responses towards composers and their works. Theory, as I said before, follows practice. The proper philosophical position for all principles in art criticism, subordinate though it is, I believe to exist well and truly after the event.

Being the supreme egoist, I naturally ask myself the following and very simple question: what has all that to do with me? Well, quite a lot! Every layman can hardly fail to notice how often Mr Walker refers to the listener as well as to the musician or the musical critic. One of his most awe-inspiring notions is that the unconscious identification with certain music is entirely pre-determined. Of course one may improve one's appreciation with the passing years, personal experience and repetitive listening, but if the most vital part is not there from the very beginning, there is no use in trying at all. For want of a better word, I am bound to call it ''unconscious identification''.

For me, personally, this means what I have already realised, if not unconsciously at all events subconsciously, namely that it makes no sense whatsoever to try to like any music that simply does not correspond with my inner self. In my youth I used to be mortified that I always found Bach and Handel monstrously boring or the New Viennese School something a little above pure noise. I no longer have any such qualms. Nor do I wonder why my preferences in the nineteenth-century music lie very much in the realm of ''revolutionists'' such as Beethoven, Liszt and Wagner, rather than in the realm of ''traditionalists'' like Mendelssohn, Schumann and Brahms. Least of all am I worried that some apparently great composers, such as Bruckner and Mahler, that I really should like, this assumption being based on some feeble intellectual criteria, I actually find perfectly dull.

Among my favourites, there are peculiar cases of course. What about Mozart, Schubert or Tchaikovsky? Wolfi is rather a mystery, for whatever I have heard of the eighteenth-century music I have found it tedious stuff; I am forced to conclude that here is a strong case of powerful unconscious identification; it is absolutely beyond explanation. As for Schubert and Tchaikovsky, the unconscious factor reigns supreme here too, but there also are some fascinating (extra)musical factors. Aren't these two great composers the greatest ones to precede and follow Liszt, respectively? Indeed, they are. The case is particularly strong in the case of Tchaikovsky, who composed a number of outstanding symphonic poems apparently inspired by great literature such as Shakespeare, Dante and Byron. Had Liszt ever had Tchaikovsky's unique melodic gift, he might well have composed Romeo and Juliet or Manfred. Had Tchaikovsky ever had Liszt's extraordinary inventiveness, to say nothing of his infernal insight if I may put it so, he might well have composed the Dante and Faust Symphonies. As a matter of fact, both composers being as they are, Tchaikovsky's Hamlet and Francesca da Rimini do sound extremely Lisztian, do they not?

I may safely say that with this little book Mr Walker has forced me to reconsider all of my musical tastes, or to rationalise them perhaps. He might, even, have changed my whole musical perception, hopefully for the better. Pretty good achievement for a hundred pages full of incomprehensible musical examples.

I certainly can't finish this rambling in a better way than with a long quote from Mr Walker's conclusions, which are as meaningful as anything written by him. Ironically, my claim in the beginning was proven wrong: Alan Walker's approach to musical criticism is every bit as personal as that of Harold Schonberg or Bernard Shaw. The significant difference is that Mr Walker tries to explain his reactions on a purely intellectual basis; but he remains convinced that criticism itself in an intuitive process. In other words, we've gone a long a way round to come at the same place as in the beginning. But it's been a most exhilarating ride.

Even more ironically, though he doesn't quite say that, I believe Mr Walker realises that musical criticism, theory or practice, is essentially useless except insofar as the individual who does criticise and does rationalise is concerned. But this is surely an oversimplification of mine. The fact that Messrs Schonberg, Shaw and Walker do exist proves that musical criticism, or some tiny part of it at any rate, is very useful indeed.

Lastly, Mr Walker makes a fine case against negative criticism which, in fact, is a logical impossibility. Of course this does not always apply to criticising performers, but it does fully apply to criticising works. Keep this in mind next time you read a scathing review of a favourite work.

My book is not an attack against criticism; it is a defence of criticism. Even so, the theory it unfolds will almost certainly not remain unchallenged. The twofold conclusion towards which it drives is too stark for that.

First: the practice of criticism is destined to remain at the litmus-paper stage. Critics take a dip into music and we see what colour they turn. That, basically, is all critical practice is about. Second: the practice of criticism is really a solution in search of a problem. How a critic reacts is largely pre-determined by deep-rooted musical and psychological force. That, basically, is all critical theory is about.

How does this constitute a ''defence'' of criticism? – in view of the nature of my theory the term seems almost ironical. I think that we do criticism a great disservice by emphasising its intellectual aspects. An act of criticism is not an act of intelligence: it is an act of intuition. The intellect is there only to move in after the intuitive event, to explain it. It would make no difference to a critical reaction if it did not move in at all. Litmus-paper still changes colour even when we are not interested in knowing why.

[…]

For the rest, there is one, indispensable symptom without which both the theory and practice of criticism remain sterile. […] Like performance, criticism should proceed on a basis of positive, intuitive involvement; paradoxically, it leads to the truth about music far more quickly than a negative search for defects. This, incidentally, is the unanswerable case against ''objective'' criticism which, being psychologically impossible, is correspondingly truthless. How do you criticise music you insist on holding at arms' length? Until you have embraced it, assimilated it, there is nothing to criticise; but once you have, your ''objectivity'' has vanished. The dying echoes of the great ''objective-subjective'' debate can, as a matter of fact, still be heard today in the drawing-rooms of the musical intelligentsia. They need not detain us, however.

[…]

A critic functions most truly, I believe, when he plays the role of counsel for the defence. If he finds himself out of sympathy with the evidence he ought not to accept the case. I am by no means the first musician to think thus about criticism, although I may well be the first to want to base a theory of aesthetics on that belief.

If he is also the last, so much the worse for aesthetics.

I have a notion that the best review of a really great book is a careful selection of quotes – and nothing else. As it turned out, this is more or less what I have done with Alan Walker's An Anatomy of Musical Criticism. This must be one of the most extraordinary books I have ever read. About one third of it is completely incomprehensible to me, I disagree with a good many observations of the author and I by no means always find him entirely convincing in the other two thirds. Yet I have no hesitation to rate this book as highly as I am able to. It is beautifully written, extremely readable and fabulously thought-provoking. In a hundred pages or so, Mr Walker gives you a simply unparalleled insight, quite out of proportion to the length of the volume, into the minds of many a great composer. Most of all, he offers lots of food for reflection about the elusive nature of music, how it is born in the realms of the unconscious and why, though the least tangible of all arts, it is perhaps the most genuinely universal and powerfully affecting one.