considers the nature and perception of visual and musical entities. The de- pendence of accurate musicological formulations on the historical orienta- tion of our musical hearing is discussed by Ruth Halle Rowen, and Edward T. Cone explores the historical meanings, criteria, and conceptual implica- tions of our basic terms for the musical work. Bernard Stambler, responding to Professor Crocker's plea for the development of historically appropriate formal senses, deals with 18th-century thought, its character and background, and the manner in which it has conditioned our present-day approaches to the entire musical repertoire. Problems of successive and simultaneous apprehension with regard to Miss Carpenter's notion of the musical object are investigated by David Burrows. In the forthcoming issue, Saul Novack explores the various criteria for the cohesion of multimovement works, and Maria Rika Maniates compares composition as "objectified entity" with composition as "continuous pro- cess" in an ontological and historical framework. Philosopher Arnold. Berleant notes, with respect to the problem of musical explanation, a distinc- tion between auditory and analytical forms of perception. The relationship of music to the emotional experiencing of reality, in a historical context, is discussed by the philosopher Robert Hall; Kenneth Kaiser, from the viewpoint of architectural theory, considers the notion of the "piece of architecture". Finally, Miss Carpenter and Professor Crocker will have the opportunity of responding to the commentators. Since it is the purpose of this project to promote an exchange of ideas, we welcome further comments from our readers. Richard L. Crocker, Some reflections on the shape of music In a recent article ("The troping hypothesis", The musical quarterly 52: 191, 1966) I dropped what I thought was an innocent remark, to the effect that one might reflect on what constitutes a piece of (Western) music. Now I find myself asked to do the reflecting. Please look for no more than that; my intent is merely to begin a line of discussion, not to conclude one. The question came up in connection with an introit trope, which seems at first glance not to qualify as "a piece of music", because it is interpolated as three separate sections in a pre-existing piece, the introit. Why do we hesitate to call such a trope a "piece"? Obviously because it seems to be three pieces, not one; for we expect a piece to be continuous, or at least not inter- rupted by some other claim on our attention. But then, each of the three pieces into which our trope has fallen is too short to be a piece all by itself; for we expect a piece to have a certain minimum substance. Or do our three pieces of a trope, along with their introit, rather form one large composite piece? But such a piece seems too composite, its elements too diverse; for do 50 we not ask that a piece be by one, and only one, composer? And if the trope cannot be considered together with the introit, but rather must stand on its own merits, still we are bothered by the fact that its sense seems (in principle, at least) to depend upon something outside itself, upon the introit. Without reflection, I think we agree that such a trope is not a piece. As we consider the reasons upon which our reaction is based, however, I think we may well begin to have doubts. For the reasons we produce so confidently at first, do not seem so universally applicable when we try them out on different repertories, especially on very recent repertories, and. even the repertory that we-or perhaps more precisely our immediate forebears-call tra- ditional. I do not offer either a comprehensive sllrvey of Western forms or a new theory of what constitutes a piece; instead, only a cursory discussion of a few of those bothersome cases that refuse to abide by the most obvious canons of form, canons we share by common inheritance and perhaps uncritical con- viction. It may well be (and I hope it is) that if the canons here so crudely formulated are merely rephrased, they will gain a sophistication sufficient to extend their validity as far as necessary. In trying to arrive at a clear idea of what constitutes "a piece", one of the most persistent obstacles is the popularity of "multimovement forms", as they are sometimes called in the traditional repertory. Those symphonies and sonatas that are regarded as masterworks (and hence generate canons of taste) are so persuasive in the succession of their movements that we in- stinctively take the "piece" to be the "whole piece", consisting of several organically related parts, the divisions between these parts or movements being merely articulations in the larger form. This sense oflarger form mayor may not be present in any given symphony or sonata; the point is, we assume it should be present, and when we look for it in other repertories-as we do-we are inclined to note its absence as a fault. But this is to ignore a prior reality, a reality both historical and phenom- enological in nature. The whole genesis of multimovement symphonic form betrays an intense effort to join together entities once separate, whether we look to the opera sinfonia, whose movements could be played separately as overtures to the several acts, or to the suite, which for a long time retained elements of its original ad libitum constitution. The sonata da chiesa, on the other hand, is perhaps to be understood in terms of an over-all governing idea, but only because it involved an artful selection and ordering of distinct entities that elsewhere appeared in ad libitum formats. In the 17th century the structure ofa keyboard suite apparently depended upon the selection made by the performer from the dances and other pieces available to him. The only guiding principles were general ones; the constituent pieces were usually by one composer (that is, usually the performer), usually in one "tone" or key, and tended to follow a conventional order, for example, allemande- courante-sarabande. These principles, however, are distinct in nature from the principles that govern the over-all shape of a symphony, and we would be 51 mistaken to apply the canons of one to the other. The concept of "a piece" involves a basically different level of form in the two cases. Another clue to the position of multimovement form in the traditional repertory is its elusiveness and the zeal with which composers of the 19th century pursued it. They tried hard to fuse together the four movements of the symphony into a continuous musical experience, one that would be a true image of the organic ideal of form they heard within. As an alternate solution they sometimes expanded a one-movement form to the point where . it could rival the over-all dimensions of a multimovement form, while avoiding its over-strong articulations. Schoenberg's Verkliirte Nacht is a par- ticularly convincing example; yet is it an example of a normal concept of a "piece", or only an exceptional concept, an ideal held only by some people at some times? In between the 19th-century composer and his ideal of a con- tinuous, organic form stands the obstinate reality that multimovement form is multipiece form, that the constituent movement or piece is transcended only by exceptional effort not always crowned with success. If we grant, for sake of continued argument, that symphonies and sonatas are multi pieces, and that the idea or ideal of a "piece" can be represented by something other than the scope of a symphony, then the problem of assessing a multi piece becomes one of assessing the kind and degree of linkage realized against that intended. This problem seems slightly different from the prob- lem of the inner shape of a constituent piece-even though, as is evident, similar principles may govern both. And if we are willing to accept in com- parison a variety of kinds and degrees of linkage, then we can come to some- what novel conclusions. We usually grant without thinking that pieces get bigger as Western history unfolds. That may still be so, at least, up to 1890: Les Troyens, and even more, Der Ring, are "pieces" offormidable length. But if we accept four successive evenings of Der Ring as a "piece" (I reckon about 20 hours over an elapsed time of 77 hours), then we must compare it, not to a one-minute antiphon, or even a ten-minute tract but rather to a medieval rhymed office, a Gesamtdienstwerk involving seven or eight hours of one man's music and text spread out and informing an elapsed time of 24 hours. Wagner is longer; but the comparison does not demonstrate, to me at least, that earlier pieces were significantly shorter. Obviously, we should not compare extremes but rather more normal phenomena. If we did so, I think we could come to the very rough hypo- thesis that a piece of (Western) music is of the order of magnitude of five minutes (plus-or-minus four minutes) in length. By that seemingly whimsical observation I mean that a "piece" is not something an hour, or 20 hours, long. If a piece ends before a minute is up, the audience (if it is indeed con- vinced that the piece has ended) is apt to betray mild, and often delighted, surprise. If a piece lasts more than 9 or 10 minutes, they start to fidget, inter- nally if not visibly. Even symphonic movements do not seem to run over 10 minutes so very often. 52 This five-minute hypothesis would have to be subjected to extensive verification-more, certainly, than it is worth. But even as a crude estimate it suggests that pieces did not get significantly longer between, say, 890 and 1890. An interesting corollary is that pieces may not have diminished in length since 1890 to the radical degree we usually imagine. Perhaps, in Webern, say, we are witnessing merely a reversion from a special solution to a more general one. Or, we should say, to a broader spectrum of possibilities; for it would be foolish to substitute one norm for another. We do, of course, recognize as legitimate pieces all sorts of things besides symphonies. Still, I suspect we unconsciously measure them against sym- phonic dimensions. (Post-Wagnerian composers from Schoenberg-if not Hugo Wolf-on have been trying to tell us differently, but we tend to take their sayings as idiosyncrasies of contemporary music rather than literally.) Songs are pieces, after all, and we have songs (of comparable dimensions) from one end of Western music to the other. It seems to me we should de- velop a sense, or senses, of form that react specifically and positively to the dimensions of an antiphon, a 13th-century motet, a 17th-century aria or canzona, a Stilck-or a Piece. Surely we have now learned that a sonata by Domenico Scarlatti has form even if it does not have Sonata Form. I think we need to extend this education of our ears to a great many other areas, even including tropes. The problem of linking pieces into multipieces is cap-able of the most varied solutions, each of which suggests a different set of canons of judgment. Leonin's pieces of polyphony (organa or "works") are placed on top of a chant continuum, as are some motets in the early stage of development. 'Some pieces of music in the late 16th and early 17th centuries are placed on top of a drama continuum, as are some pieces in the 20th century (such as L' Histoire du soldat). The grouping of 17th-century keyboard pieces- preludes, allemandes, courantes, say, by Louis Couperin-into ad libitum arrangements has been touched upon; what about the grouping of 19th- century keyboard pieces-preludes, nocturnes, mazurkas, say, by Chopin- into other ad libitum arrangements? How did Chopin use these pieces, or how did he imagine them to be used? Twenty-four preludes at one sitting? I think not; rather a few, as chosen and ordered by the performer. And should we call this order by the Latin term ad libitum, or the English aleatory? It is a curious thing that while Chopin's Preludes, Op. 28, are often played ip. an order to some degree aleatory, Webern's Bagatelles, Op. 9, usually are not, even though Webern would be commonly understood as the composer with the less perceptible "form". But if we take seriously the order of Webern's Fun! Siit;;;e, Op. 5, or even his Fun! Stucke, Op. 10, we might almost be persuaded we were hearing an (admittedly diminutive) five-movement form alIa Beethoven or Bartok. Yet we should honor Webern's-and Schoenberg's-sometime insistence that they were writing Stucke [Pieces], not multipiece symphonies or string quartets. It is evident, in any case, that music is not presented to us as a never- 53 ending continuum, but rather in discrete units, or pieces, and that an ex- tended exposure to a composer's consecutive thought is intermittent. Operas come in acts, Der Ring (gratefully) on four separate evenings; in between we walk, talk, eat, and perhaps sleep. Acts come in scenes and arias; and some- times we applaud in between. Orchestral music comes in movements, and only the more devout maintain fixed attention during the interim. What about our trope? How much of an obstacle is it that the lines of the trope (which constitute a believable form when considered by themselves) are interrupted by fragments of the introit? If it is too much of an obstacle, we could in some cases imagine the trope sung without the introit, just as concert suites are often formed (and variously so) by extracting music from ballet. Is the loss of logic, of consecution, really greater in the one case than in the other? Ifwe do have to deal with trope and introit together, in alternation, as a single artistic happening, then the formal problem is most severe, for here we confront our conviction that a piece of music should be by one composer at a time. The pasticcio, the arrangement, indeed all such symbiotic forms, seem to be repugnant. The question is actually a complex one; it involves pur- posiveness and artistic identity, and their relation to idea and formal control. There is one approach that might be useful in the present case. Once upon a time it was thought that all aspects of a sounding piece might be brought under a composer's control, and it was further thought that this would be a good thing. On a broader front, however, we have since learned that inde- terminacy in some aspects is a better thing and in any case an inevitable one. Almost everyone accepts the fact that tempo, tone color, balance, inflection, can vary among performers and performances. Yet we feel that the format of an artistic conception should be inviolable, as if the format or rhetoric of a piece were in some way connected more closely than the inflection to the basic idea or identity of the composer's utterance. Why? It is not too hard to imagine music whose utterance is so intimately associated with tone color or inflection that these should be strictly controlled, while the large-scale format might just as well be left unordered. That could include renouncing unilateral control over the audience's attention for an uninterrupted stretch of time. We accept a conversation, a panel discussion or debate, as a form of verbal ex- pression; why not a musical counterpart as well? All this has been concerned mainly with variant solutions to the multi- piece problem. There are many other problems associated with the idea of a "piece"; I would like to consider only one, and that briefly. We share a strong conviction, I think, that a piece should have a beginning and an end, and a middle that bears a reasonable and perceptible relationship to the beginning and end. (Actually these are several problems.) But already we are not sufficiently precise: all pieces have beginnings and endings, simply be- cause everything does-everything in the world. Here is no canon of taste, but only of necessity. Is this the meaning of our persistent partitive expression, 54 "a piece (not the whole) of music" ? No matter. What we require of a piece, I think, is that the beginning sound like a beginning, and especially the end like an ending; and that the middle move us from the one experience to the other. But now the matter seems quite different than at first, and nowhere near as obvious, or as obviously universal. Clear beginnings and endings, then, are a matter of taste, and the taste may vary. Some like a very clear beginning, a "premier coup d'archet"; others, including Mozart, feel that the second "coup" has an equally strong claim on our attention. Many instances of beginnings, middles, al,ld ends that do not live up to our categorical proposition will leap to mind; let me add one or two that might not otherwise l>e mentioned. Psalmody, a widespread musical experience in the Middle Ages, has a singularly undirected middle, the same melodic for- mula being repeated a (musically) indeterminate number oftimes, each time for a (musically) indeterminate length. I find certain similarities with in- stances in the rock-and-roll idiom and its derivatives, such as folk-rock. Here, too, there are interesting attempts to give the illusion of no ending. At some seemingly indeterminate point the music starts to reiterate a phrase, and after a seemingly indeterminate (but in practice controlled) number of reiterations, the music fades out to be replaced immediately by the recitative-like con- tinuity of the disc jockey. The intent, clearly, is an eternity symbol; it is not that the music ends, rather that the listener stops listening, while the music goes on and on. Often announcer and song overlap at both ends. If this ex- ample is trivial, it cannot be called uncommon. I doubt whether the discussion has made our trope seem any more like a piece. Actually, such tropes are relatively extreme examples of the kind of formal disassociation touched upon throughout these reflections. The trope may remain a nonpiece; but it may help us to understand other less extreme, and more important, instances. In these, it becomes increasingly apparent that a strong quantitative factor is involved-how much linkage, how much continuity or conclusiveness, how long the piece. Are we to conclude that art knows no eternal rule? Nb canons ofform? No criterion for distinguishing a piece from a nonpiece? Even ifwe were to con- clude the substance of such a proposition, we should rephrase it to say that we know no rule, rather than that none exists. But we hardly need to retreat to such extremes. One of the things we mean by "creative", I think, is that creation results in the existence of something that was not there before, and in music this seems to have as corollary that a radically new musical creation precedes the canoll;s of judgment that concern it. (The art of the Meister- singer, while another extreme case, is still representative of what happens when musical activity remains within existing canons:) A historian, if so in- clined, might conceivably formulate canons so general as to hold for all music ever written. But he could not, as historian, ever find canons that would hold for all music yet to be conceived and created. It seems doubtful that even an aesthetician, from his lofty lookout, could see far enough for that. There will, hopefully, always be some pieces disorderly enough to call forth new 55