The conventions of a dictionary are as formal as those of a sestina, a minuet, or the architectural orders. In a preface, Dr. Willi Apel ominously compares the use of his volume to a visit to the dentist, but that, too, has its formalities and its ritual. In this work, revised and enlarged after twenty-five years, the useful is evidently intended to outweigh the sweet. Nevertheless, no literary genre can so easily combine instruction with delight as the dictionary. The majesty of the OED, the intimate charm of the Petit Larousse, depend on the classical sense of propriety that they exemplify, works of art never read as wholes, but whose unity radiates and reveals itself in all their parts. In a dictionary, lack of grace entails a loss of utility.

The Harvard Dictionary of Music offends against decorum on the first column of its first page. It explains con abbandono only as “unrestrained, free” and not as “passionately”: under Abbreviations it gives con 8va (i.e., with octave doubling), but not 8va itself (transpose an octave upward or downward), which is more often used; and it implies that there is only one way to interpret on the piano, when there are two clearly distinct meanings to that notation (the Harvard Dictionary’s interpretation would simplify a lot of wrist-breaking music, including Schubert’s Erlkönig). Even the initial entry, A, betrays its innocence of the properties of lexicography by listing a 2 as an example without explanation, apparently because it needs none. Yet if there were a cross-reference to A due, which comes up thirteen pages later, the less wary would learn that the meaning of a 2 is very ambiguous indeed.

In many cases, of course, if not in all, the uninstructed will be able to figure out what is intended (just as most anglophones will assume correctly that con abbandono means “with abandon”), but if this is to be the plea, why bother printing anything at all about A due beyond an identification of “due” as the Italian for “two”?—and this could safely be left to Italian-English glossaries. The Harvard Dictionary is cavalier about cross-reference, and even that first article, A, is miserly with the asterisk that indicates a separate entry under the word lucky enough to be starred: partbook and a piacere are among the elect, but antiphon must hope for the enterprise of the individual reader.

Now in none of these opening articles is the information given absolutely false: they contain errors of method rather than of fact. True howlers and misinformation of course abound in all dictionaries, and the Harvard Dictionary of Music is no exception, but they are to be expected and even welcomed; they complement the more serious parts of a work of reference as the satyr-play sets off the tragedy. I am as delighted as the next reader to find Ravel’s Jeux d’eau defined as Water-games, as if it were not the play of fountains but a form of water-polo. To read again that Beethoven introduced the trombone into symphonic music (to say nothing of the triangle and the big drum) should excite more sympathy than censure, and the idea that Schoenberg actually intended his Music for a film sequence as part of the repertoire for silent films, like the pieces labeled “Help, Help,” is too ludicrous to mislead, and too engaging to wish corrected.

The biggest howler of the first edition is left unaltered: on page 4, perhaps the most famous phrase that Mozart ever wrote, the opening of the G Minor Symphony, is quoted to exemplify a combination of agogic accent with an irregular dynamic accent: only that is not what Mozart really wrote. His phrase, of course, has neither a dynamic accent (even the editors sense this as they put it in parentheses) nor an agogic accent: the B flat is short and followed by a rest. The reviser must have left this error because of its grandeur.

Methodical errors, however, are more damaging than comic mistakes, 8va can only be omitted by a slapdash approach to listing musical abbreviations, and con abbandono can be translated “unconstrained” in a musical dictionary by someone who has before him only an Italian dictionary and not a single instance of the use of the term in a musical work—one, say, by Liszt. That is why the opening page of this dictionary is disquieting; it seems to be the work of a committee that never asked itself why anybody would look up a given entry. In the rest of the dictionary, the separate entries are largely composed as if specific musical considerations did not exist. They are brilliantly crammed with information, and often dangerously misleading.

How a correct and informative article can deceive the ingenuous may be seen from Swell:

In organs, mechanism for obtaining a gradation of sound, crescendo and diminuendo. It consists of a large room (swell box) built around one or more divisions of the pipes and provided with shutters similar to Venetian blinds, whence comes the name Venetian swell (G. Jalousieschweller). The chief enclosed division is called swell organ, a name that also applies to the manual from which it is played. The swell box is opened and closed by a swell pedal, operated by the feet. The first practical swell mechanism, invented in 1769 by Shudi, was used in harpsichords before it was adopted for the organ.

Leaving aside the minor irritant of a “pedal operated by the feet,” let us affirm at once that the information is absolutely accurate. The article, however, implies that the swell was little used before 1769; yet Burney, when he visited Europe in the early 1770s, was astonished to find only one swell mechanism in all the organs in Germany because, as he said, they had had them in England already for half a century. No doubt the English mechanism was not a “practical” one, but it existed and this may have its importance for Handel’s organ concertos. Most students and music-lovers will look up swell to find out when and where they were used, and the Harvard Dictionary will only fool them; the entry is useful chiefly for people who want to settle a bet about Shudi and the date of his invention.


This kind of lexicography is corrupting in an odd way; it gives the reader who happens to know better and can avoid the traps innocently set for him the pleasant sense of knowing more than the team of musicologists, many of them distinguished, who compiled the dictionary. This agreeable glow of self-satisfaction is unwarranted. The Harvard Dictionary knows perfectly well that English organs had swell mechanisms early in the eighteenth century. It does not relay Burney’s comment, nor is there any reason why it should, but it does inform the reader in the entry under Organ about the English swell. It does not, however, add a cross-reference to the entry under Swell because it is content to get things abstractly right without thinking of the reader or indeed of the musical reasons for consulting its pages.

For this reason, the praise that must be accorded the Harvard Dictionary with one hand must often be withdrawn with the other. The article on Swell is not an isolated instance, but the reflection of a fundamental irresponsibility. A Positive organ, for example, is indeed “a medium-sized medieval organ,” but the “very small positive” mentioned two sentences later existed during the eighteenth century. Under Rückpositif we are referred to Positif but not to Positive organ, and the complete entry for Positif reads “choir organ” with an asterisk referring the reader there. Why Rückpositif sends one only indirectly to Choir organ, and not at all to Positive organ is not easily explained, at least not charitably, and one must add that Choir organ sends one to Organ part III, and Rückpositif to Organ parts II, VII, and XII, II being here evidently an unchecked error.

Digging information out of this dictionary can therefore be a major excavation problem, and at the end one might still be forgiven for assuming that a small positive was exclusively medieval. Mit Andacht is thoughtfully and ridiculously indexed under both M and A, but I cannot see that the definition “with devotion” will enlighten anyone who does not already know that a religious context is meant. Luftpause is defined only as “breathing rest,” but not as a slight pause that is not indicated except by a comma, and which forms no part of the regular rhythmic structure.

If one wants to know why the finale of Mozart’s Entführung is called a vaudeville, he will not find here a definition of the form: a song for which the whole cast lines up, while each one separately sings a verse (this is sometimes called a vaudeville finale, but more often simply a vaudeville). Anyone puzzled by the eighteenth-century use of the term modulation will not be told that it meant voice-leading. The examples of typical rhythms of ars nova and ars antiqua are oddly chosen: the purpose of ars nova notation was to accommodate easily a wider and more complex range of rhythm, and the ars nova example is both simpler and more limited than the one of ars antiqua. This is another case where a correct presentation can misrepresent the essential musical point.

The failure to observe the fundamental courtesies of a dictionary makes a hash of the simplest musical concepts. Are the words and phrases, following Samuel Johnson, to be interpreted “with brevity, fulness and perspicuity”? Let us take as elementary an article as Answer, with its definition at once verbose, incomplete, and obscure (I quote the entire entry):


In fugal writing, the answer is the second (or fourth) statement of the subject, so called because of its relationship to the first (or third) statement. Hence, the succession of statements is subject-answer-subject-answer. See M. Zulauf, in ZMW vi. See Fugue; Tonal and real; Antecedent and consequent.

What is this mysterious relationship from which the veil is not to be torn? Simply that the answer is on the dominant or more rarely after 1700, on the subdominant, or—in the case of counterfugues inverted on the tonic. The further references will not help much: in Fugue, one must plow through a page of turgid, badly organized prose before finding that an answer is on the dominant—and the matter is left at that. Tonal and real makes no further advance on this, and Antecedent and consequent is even more reticent, and contains, besides, a misprint in the musical example doubly unfortunate because it is just plausible. In any case, why is there no reference to Dux, comes, which would at least give students a chance to display some unnecessary erudition? The editors of the Harvard Dictionary know all about answers at the subdominant (see Imitation) and at the inversion (see Counterfugue) but they do not know how to tell anyone that they know.

These omissions are not isolated, but systematic. No work, even of almost a thousand double-column pages, can embrace the whole of music, and it is an unfair game to play with dictionaries and encyclopedias—particularly revised editions—to see what is in and what is out. But it is an instructive game to play with the Harvard Dictionary, because it reveals the principles with which the revision was undertaken. Composers do not get separate entries, so the amusement of seeing that one’s friends have been left out is barred. But individual works are listed. La Cosa Rara is in, but Le Devin du Village, one of the most influential works of the eighteenth century, is out. Oberon and Euryanthe are also out although L’Africaine is in. Theodora, which Handel said contained his greatest music, is out, as is Samson, Semele, Esther, and Susanna, but Susannah by C. Floyd and even Il Segreto di Susanna are in.

The twentieth century is heavily represented by Hindemith and Honegger, with comically disproportionate results. I am unmoved by Dr. Apel’s loyalty to his youthful idols. Jeux by Debussy is out, but Rugby by Honegger is in; Le Martyre de St. Sebastien is omitted, but room is made for Jeanne d’Arc au Bûcher. Le Marteau sans Maître, Gruppen, the Gesang der Jünglinge, and the Improvisations sur Mallarmé are unlisted, while Pacific 231 and the Schwanendreher keep their place although they have long since disappeared from the musical scene. La Fanciulla del West is there, but not Fancy Free, to say nothing of West Side Story. The principle on which the dictionary was revised and brought up to date seems to have been that the research assistants jotted down a few items which occurred to them. And they live in a small, tight, closed world. There are newly written articles on jazz and even on bop, but there is nothing at all on rock, not even a brief definition of the term.

If this haphazard revision merely applied to contemporary musical life, while the musicological side had been done with care, there would be no excuse for niggling. There have certainly been considerable additions to bibliographical lists, and many articles have been entirely rewritten. New terms have been defined, and there is a considerable expansion of material and discussion. Some, but not enough, of the silliness in the first edition has been removed: we are no longer told that the “solemn dignity [of Parsifal]…does not always escape the danger of monotony.” Dr. Apel’s stern admonition that even the greatest piano virtuosos need to practice finger exercises every day has unfortunately been blue-pencilled; in compensation, we may still read the splendid alliteration that some wiseacre slipped into the first edition: “Among later composers Tchaikovsky showed a particular predilection for pedal points.”

But why has Leo Schrade’s interesting and brief discussion of the sixteenth-century use of the term maniera been excised? The one sentence that replaces it now does not even tell us what maniera is Schrade died since the first edition but that is surely not sufficient reason for the change. His article was a serious attempt to relate music theory to the other arts which makes the articles that have been retained (e.g., Expressionism and Impressionism) childish by comparison.

The improvements are naturally many: the new article on Development is considerably better than the old one. The entry on Krakowiak is more informative. There are excellent entries on Descant (written by the late Sylvia Kenney), Parody Mass, Glass Harmonica, Fauxbourdon, and numerous others. The second part of the article on Sonata form, the historical section (revised by Leonard Ratner), is now intelligent and sophisticated, but the first part is little changed, and the quality of thought and writing is still so flat here that the student will receive more illumination from the new Britannica article by Donington, or even the old one by Tovey. Melody is as nonsensical as ever, and cannot even be summarized: Part IV must be read to be disbelieved.

Articles on individual countries are much expanded. I read the one on Java with great interest. Czechoslovakia has now, it seems, the longest article of any country, longer than France or Germany—or even Harmony. There are many more works represented, although Schumann’s Carnaval has disappeared by mistake, I presume, as many much less famous works of Schumann are there. Someone must have confused it with the Carnaval des animaux, which has been added in the new edition, ousting the Schumann to avoid duplication. In addition, there seems to be no editorial policy at all on the admission of expressive terminology: mit Wärme has been added, aufgeregt has been left in, and mächtig has been taken out. There is only whimsy in all of this.

The updating is seriously defective in many places; it is also contradictory. Pitch lists the recent and well-known articles by Mendel in its bibliography (he is called R. Mendel—R. for Arthur), but the text itself takes almost no account of his researches, and is largely a slightly reworded repetition of its twenty-five-year-old predecessor. Chiavette, however, does take account of these same articles, but renders the discussion totally unintelligible by misprinting the first F clef on the fourth instead of the third line—the point at issue being exactly the unusual position of the clef.

Until now I have dealt largely with errors that could easily have been avoided by a respect for the nature of a dictionary—its necessary clarity, elegance, and order. But there are graver charges to be preferred against the Harvard Dictionary; it is often tendentious, its approach is too frequently unhistorical, and it is consistently stuffy in outlook.

It is unsportsmanlike to have an entry on a controversial subject written by a fanatical partisan of one side. An argument has been raging for some years on notes inégales, and while the article on the subject takes a position that seems to me judicious and even unexceptionable, it does not fairly represent the present conflict of opinion. Serial music is an example of a different kind: written by Henri Pousseur, its opening is interesting and lucid. Its last section, however, is a long, apocalyptic vision of the future, of what serialism will become (with the obligatory structural linguistics thrown in). At the end, the heavens open: “A vast system is now beginning to emerge, able to assimilate all the earlier systems that have contributed to its development (including, e.g., the tonal system).” I congratulate myself on not being able to understand this.

The tendentiousness gets its come-uppance. One of the most egregious lapses in the first edition is rendered even more ludicrous in the second. In the article on Musica ficta (the problem of adding, or not adding, sharps and flats to medieval and Renaissance music), the first edition remarked with justice that no answer could be found that “would apply equally to music of the 13th and of the 16th century.” Then, with a straight face, examples of “unusual but entirely legitimate formations” were appended—all undated and unidentified.

The examples of ficta in the second edition have been dated—1955, 1939, 1959, 1947, etc. They are the dates when various editors have inserted, or have not inserted, accidentals. The article makes it clear that it is not the musical text which is at stake, but musicological fashion. The brief reference to Lowinsky’s theories in the first edition has been excised, and the article in the second edition, without naming him except in the bibliography, is conceived largely as a covert attack upon his position. Lowinsky may be right or wrong (and my own opinion on the matter carries no weight even with myself), but the article represents the general state of musicological opinion only in the most biased fashion. It is, therefore, a pleasure to note once again the foolishness of giving examples that span almost two centuries of revolutionary developments without dating any of them, and to remark that no account is taken of the evidence of lute arrangements (although organ tablatures are mentioned), and that no hint is dropped that the use of ficta probably varied from country to country as well as from age to age.

The lack of dating is endemic in this book. Modulation is treated as if the process remained relatively unchanged from 1450 to 1900 and only its application to different formal schemes was altered. How change of key can mean the same thing in fifteenth- and eighteenth-century harmony is difficult to comprehend. In any case, a discussion of modulation that does not once mention the preparation of a modulation declares itself out of court to start with, and the bibliography omits the most famous of all books on modulation, Reger’s little treatise.

Examples in Counterpoint are not dated, either, and they span four centuries: identifying the composers does not help in a book where composers have no separate entries. The kind of reader who knows the dates of Perotinus and Dunstable will probably not read the article Counterpoint unless he is reviewing the book. If it is noted that it is by these examples alone that contrapuntal practice between 1200 and 1620 is covered, that they are presented with the barest minimum of commentary, in captions, and that contrapuntal theory is treated cursorily and is largely a list of theorists, it will be seen that the approach is not historical but taxonomic.

This is at the root of many of the deficiencies in the book: the role of Thoroughbass after 1750, for example, has occasioned a good deal of research, not a word of which seeps into the article on the subject. Only the lack of historical sense could characterize Bach’s music as without “a definite emotional character” (p. 2) and say that “baroque instrumental music tends to appeal to present-day listeners because of its detached, non-expressive character” (p. 301), and at the same time, print an excellent article on the Affections, doctrine of in which a Baroque theorist is quoted as describing “the affections (characteristic emotions) of numerous dances, saying that the gigue expresses ‘heat and eagerness, the courante ‘sweet hope and courage.”‘ The relation of theory to practice is never simple, but the conception of a single Affekt as the basis of every piece of music is fundamental to the period 1700-1740, and an allemande by Bach is as “expressive” as a nocturne by Chopin.

The same unhistorical turn of mind makes a terrible confusion of the article on consonance and dissonance, combined with an inability to define a musical term in a musical sense. Chamber’s Twentieth Century Dictionary defines a dissonance as “a combination of musical sounds that calls for resolution or produces beats,” but you can read all the many hundreds of words in the Harvard Dictionary article without finding resolution mentioned once. Dissonance is defined only as “a disagreeable effect,” as if music had neither significance nor structure.

This produces the historically incredible statement that the intervals of the fourth and the fifth, “from the point of view of musical composition of all eras…must be regarded as consonances,” although (p. 329) the most important of late fifteenth-century theorists, Tinctoris, is quoted as saying that the fourth “was considered by the ancients the foremost of all consonances, but actually, taken by itself, it is not a consonance but an intolerable dissonance.” Back in Consonance, dissonance, the dictionary becomes uneasy and adds that “the fourth has a decidedly unpleasant effect for an unbiased listener.” Contemplating this, we peer into an abyss of naïveté.

In harmonic theory, the Harvard Dictionary has not progressed beyond Riemann. Fingering at the piano stops with Clementi (what happened to the Liszt fingering of the scale, and Chopin’s use of the 3rd, 4th, and 5th fingers for chromatic passages, etc?). The list of theoretical works stops at 1900, with a few more added pell-mell in a final short paragraph. Performance practice is called “the study of how early music, from the Middle Ages to Bach, was performed,” and the article concludes, “in the period after Bach the problems of performance practice largely disappear, owing to the more specific directions of composers for clearly indicating their intentions.” This breathtaking provinciality makes it possible to claim (p. 731) that “after 1750, metric developments are relatively uninteresting until Beethoven” because the editors have not heard of Haydn’s and Mozart’s revolutionary treatment of rhythm, and to write (p. 120) that “the cadences of the classical and romantic period are of little historical interest since they usually conform to the standard types outlined”—and Beethoven’s minute-long hammering at the tonic and Schumann’s poetically unresolved cadences were as if they had never been.

It is an analogous provinciality of style that makes the often pontifical and patronizing tone of the Harvard Dictionary offensive, whether one agrees or disagrees with the judgments gratuitously offered. We read (p. 7) that Tchaikovsky “rarely went beyond a chordal accompaniment in lush harmonies of rather ephemeral interest”; that Tovey’s point of view in analysis was one-sided (p. 36); that Hugo Wolf’s songs are better than Schumann’s—and this in a long article on Lied which omits even a mention of Loewe. The Harvard Dictionary thinks that late fourteenth-century French music is decadent (p. 58), but, in compensation (p. 350), Italian music of the fourteenth century “is perhaps too ‘earthy’ and ‘lively’ (too ‘proto-Renaissance’) to be termed Gothic”—it must be many years since a major university press allowed itself to print so grotesque a statement on the visual arts.

The Harvard Dictionary does not, of course, adequately represent the view-point or the achievements of American musicology. The real glory of the volume is its two immense lists of Libraries and Editions, historical. The bibliographical notes, if unsystematic, will be very useful to graduate students. It is the only musical dictionary in one volume of such pretensions and range, taking in terms like Echegiatta, Echiquier, and Ecphonetic notation. The learning and labor behind such an enterprise is formidable, and it is a tragedy that it has been marred by poverty of thought, inconsistent editing, and a rebarbative style of writing.

I feel myself competent in only a small part of the range of this book, but it is difficult to take on trust a dictionary that defines the scheme of the medieval ballade incoherently (the last two lines appear to rhyme with nothing at all), and is both embarrassing and inaccurate about Chopin’s ballades. It is, perhaps, merely bad luck that (p. 94) of the two examples of ternary form given, one (the slow movement of Beethoven’s op. 10, no. 3) should be binary according to the dictionary’s own discussion of an often-confused term (first paragraph of p. 95), and in any case is not ternary. Even simple terms like Waldhorn and Arabesque are improperly defined. “Waldhorn” is only used today to mean a French horn without valves (and that is, in fact, the way the dictionary itself uses it on p. 392), and the unmentioned analogy with a curved, unbroken ornamental line is essential to any musical use of the word “arabesque.”

The dangers of this pedestrian, unreliable, and often useful book may be illustrated from within its own covers. The Harvard Dictionary (p.517) thinks that Schoenberg’s Erwartung consists of spoken dialogue with a musical background. Not even Sprechstimme—real, genuine, honest-to-goodness spoken dialogue. How can this be? Even if we admit Apel’s ill-mannered remarks elsewhere (p. 482) about Schoenberg’s “unvocal line,” the soprano in Erwartung sounds at least as if she is trying to sing. How did anyone get the idea there was any spoken dialogue?

It is possible to reconstruct what happened. The correct term for spoken dialogue against a musical background is melodrama, and the article so named remarks that “if only one or two actors are involved, the terms ‘monodrama’ or ‘duodrama’ may be used.” Monodrama reads “See under Melodrama.” Schoenberg called Erwartung a monodrama because there is only one singer. Somebody working on the dictionary had a sudden twinge of conscience and did some rare but this time unfortunate cross-referencing.

It remains to be added that the book is handsomely presented, the paper is splendidly white, the print agreeably black and legible. The text has been admirably proofread (a difficult task for such a long work), but the musical examples swarm unprecedentedly with errors. In Arioso, to quote only a single instance out of a hundred, one mistake from the first edition has been corrected and three new ones have been added. Many of the notes are printed so that they are neither quite on nor off a line. The volume is sturdily bound, and sports its crimson colors bravely.

This Issue

February 26, 1970