The development of aural analytical skills is a must for anyone who desires a thorough understanding of musical form and structure. It is of course an integral part of that study we call music theory—it is the wedding of the aural experience to abstract musical constructs. As both a theorist and a theory teacher, I believe that to hear music is to analyze it and to analyze music is to understand it more fully. This I would above all like to impart to my students. Unfortunately, this is easier said than done. So what is the best way to approach an ear training, or aural skills class? I have pondered that for a long time, and as a result, have developed and implemented some new strategies which I believe to be more successful than others I have used. It is these new strategies that I would like to discuss here.
Important note: Permission to add this article was given by the Gail Boyd de Stwolinski Center for Music Theory Pedagogy at the University of Oklahoma. It is a violation of copyright to download a copy of this article. This article was published in the Journal of Music Theory Pedagogy, VOLUME FIVE, No. 2, FALL 1991.
Random intervals?
One vexing problem for many aural skills teachers is the study of intervals. Do students need to be proficient at identifying random intervals before they can move on to something else? No, I don’t believe so. [1] Do they need to be proficient at hearing scale degrees and relationships within the context of a key? Most certainly. Many students do not do particularly well with random interval identification, but can do well with other aspects of aural analysis. Nevertheless, the identification of intervals seems to be a major component of many ear training and sight singing texts, CAI music software, and presumably, most ear training programs. But at the same time, many aural skills teachers question their importance, or the value of the method by which they are most often taught. A similar situation exists with regard to triads. Is there more to the study of triads than recognizing the four qualities as isolated events, and hearing which voice has the root? These are the questions and concerns that I will address here, and I will argue for a contextually-based approach to teaching intervals and triads in an ear training program.
Sight singing and ear training goals
Over the past several years I have experimented with different strategies for teaching aural skills, all of which emanate from my basic beliefs about exactly what skills I am trying to teach, how to integrate them with written theory and how to do it in the most efficient way.
What skills am I trying to teach? There are different opinions about the purposes and goals of sight singing and ear training [2], so an aural skills teacher must decide which goals he or she is pursuing before developing a program of study. With that vision in mind, all activities must be coordinated in such a way that they all work towards that end. My goal is to produce musicians who can hear and think about music equally well — musicians who can understand and see what they hear, and hear and understand what they see. I am not trying to merely provide a service by producing better sight-readers for the choirs and ensembles. If improved sight reading is a consequence, so much the better—a felicitous by-product, but not the point of the exercise.
Theory and ear training: flip sides of the same coin
The underpinning of my approach to teaching aural skills is that nothing a student learns in written theory or aural skills exists in a vacuum, independent of anything else. Theory and ear training are flip sides of the same coin — every component works to reinforce every other one. Consequently, everything in theory and ear training should be as coordinated and interrelated as possible. It is unfortunate that as a practical matter, the written portion of a standard theory class progresses much more rapidly than the aural skills portion, so there is a lag between what one has learned in theory and what one is learning in aural skills at the same time. This is not to say that examples of whatever is currently being taught in theory should not be played in class: quite the contrary. This is one of the best ways to interrelate the two and reinforce for the students the idea that hearing, seeing and understanding go hand in hand.
Tonal music
One caveat: the strategies I am proposing are without question biased towards tonal music. Since most of a music major’s two year theory sequence (at least 1.5 years in most schools) concerns itself with music of the common practice period (i.e., Baroque through Romantic), this bias seems justified. Also, in light of the fact that this music is still very much with us in the concert hall, as well as in the music school, and because I believe that later styles can profitably be studied as outgrowths of, or reactions to, the tonal system, this emphasis seems warranted.
Functional relationships
Since I am proposing a “holistic” approach, the discussion of intervals and triads needs to be prefaced by a discussion of what I believe should be the goals of sight singing, and how best to work towards that end. My goal, when teaching sight singing, is for students to learn to hear the functional relationships between notes in a key: their relationships to each other, and their relationship to tonic. Ideally, scale degree numbers should be sung, but since that would be too cumbersome, especially when alterations are added (which I think are absolutely imperative—flat-6, sharp-4, etc.), the monosyllabic solfege system is a very happy compromise.
Sight singing
I believe practicing sight singing is one of the best ways to practice ear training. To memorize what certain patterns sound like from singing them is to recognize them when you hear them in music. Learning position within a key by singing with that in mind becomes a tool one can use in either the classroom setting (taking dictation), or, more importantly, in any analysis one embarks upon, wherein one begins by listening to the music in question. Thus, sight singing is an extremely important facet of ear-training — it is how one begins to learn to hear music analytically.
Movable-do / tonic-do
Since I am advocating the use of sight singing to teach students to hear and recognize scale degrees (a skill I believe to be far more important than learning to hear isolated intervals), then only one method of sight singing will be appropriate: the moveable-do/tonic-do system (do is tonic for both major and minor keys). Moveable-do has as its goal the learning of the tonal system, unlike the fixed-do system, or the movable-do/minor-la system. Fixed-do can work towards improving reading skills, since the same line or space on any given clef always has the same syllable. But it imparts no functional meaning or significance to any note. La-minor is an interval approach. It measures the distance between syllables: ti to do is always a half step, whether in a major key functioning as the all-important, key-defining leading-tone to tonic relationship; or in its relative minor key, functioning as the much less critical supertonic to mediant relationship. At best, this system requires learning two sets of syllables to signify the same functions: one set for the major key, and one for the minor key. At worst, this system encourages hearing all minor keys in relation to their relative majors, rather than as independent keys of equal stature and validity. A tonic la is in actuality heard as being two steps below do, as opposed to being heard as a tonic in its own right.
Tonic centeredness
Beginning exercises in my sight singing classes stress learning to hear and retain tonic. At first, everything is done from do in order to emphasize tonic as the reigning entity and to develop a sense of tonic retention and tonic centeredness. For example, after learning to sing a major scale with syllables, students begin singing each note of the scale from do: do-re-do, do-mi-do, etc. One sight singing text that correlates well with this approach is the Berkowitz/Frontrier/Kraft A New Approach to Sight Singing [3], the text we use at Butler University. It begins with stepwise major-key melodies, all of which start and end on do—again, developing a sense of the primacy of tonic. Next, the text introduces melodies with skips in the tonic triad. Thus, major and minor thirds, perfect fourths, perfect fifths and even major and minor sixths are introduced early on, as the residual intervals, or resultant distances, between notes of the tonic triad in its various guises, as opposed to being introduced as intervals for their own sake. In fact, one need not even refer to intervals at this point. Students are simply learning to sing from one note of a tonic triad to any other note of that same tonic triad. This skill is fundamental: to be able to recognize and sing a tonic triad in any of its presentations as soon as possible.
Intervals vs. tonic context
A different approach employed by some sight singing and ear training texts is one in which melodies are categorized and presented according to the intervals they contain [4]. These books progress by adding melodies with successively larger intervals: seconds, thirds, fourths, etc. However, I contend that singing a third in a ii, IV or vi chord, for example, is much more difficult for a beginning sight singing student than is singing a third in a tonic chord. Why? Because a beginning aural skills student has not yet progressed to the point of being able to hear internally all those different harmonies within a tonal framework.
There is a critical distinction here which lies at the heart of my approach to teaching intervals and triads. A sight singing program that progresses by interval reinforces the “interval” way of hearing tonal music, which I don’t believe is the primary way in which most of us really process music [5]. In tonal music, intervals have no meaning outside of the context of a harmony and/or key, and in this tonal context, like intervals can sound very different. The major third from do to mi (as part of a tonic harmony) is the epitome of stability, and should be experienced as such, whereas the major third from sol to ti (as part of a dominant harmony) is anything but stable. It is the antithesis of the tonic third and needs to resolve to tonic. The urgency of the upper note of this third, the leading-tone, to resolve up by half-step to tonic is palpable. This is not at all paralleled by the major third from do to mi. Or, to cite a different kind of example, think of the case of enharmonic intervals. Out of context there is no difference. But in a tonal context there is a world of difference between, say, an augmented sixth and a minor seventh. Similar examples abound in tonal music [6].
So what is the purpose of teaching intervals per se? I would argue that we should instead be teaching students to hear the larger relationships: scale degrees, harmonies, and the affinities of notes for each other. Intervals should be taught and understood only as parts of harmonies, not as discrete units to be recognized in the absence of a tonal context. Clearly then, when practicing sight singing with these harmonic goals in mind, one should progress not by interval, but by harmony. Consequently, after practicing melodies with skips in the tonic triad, one would proceed to melodies with skips in the dominant or subdominant triad, and so on.
Since I have for a number of years taught sight singing in this manner, I realized that to be consistent, I needed to approach ear training and dictation in the same way: to wit, intervals are the distances between scale degrees, not isolated events. Moreover, since the same interval can have very different tendencies, and even different meanings, depending on the context and the scale degrees involved, I wanted to develop a method of practicing scale degrees and intervals that would make the students aware of and sensitive to these different tendencies. Interval practice therefore begins in ear training almost immediately, but, similar to sight singing practice, without referring to intervals as such. It is only after intervals have been taught in written theory that they are referred to specifically in ear training. This approach therefore de-emphasizes intervals while stressing scale degrees, tendencies and tonic retention.
Scale degree exercises
Scale degree exercises are done as follows: A key is established, generally by playing a short cadential pattern. Students are asked to sing do. Some other scale degree is then played, and students are asked either to sing from that scale degree up or down to do, or to sing do and then sing up or down to that scale degree. They then identify that scale degree (by syllable). In between each exercise, students are asked to sing tonic. Thus, not only are they learning scale degrees in relation to tonic, but they are also learning, through constant repetition, to retain tonic and to use it as a constant reference point. Naturally, the students are hearing and singing intervals (from tonic to scale degree “x”) but they are not yet labeling them as such. This exercise also lays the foundation for the next one, another scale degree/interval exercise done without reference to intervals, at least at first. Now instead of hearing one scale degree in relation to tonic, students hear two scale degrees (one of which may still be tonic). For example, after establishing tonic, I may play ti-re. Students first sing do. They then sing the two pitches they heard, usually on a neutral syllable, and sing up or down from tonic until they find the two pitches, which they then sing on syllables. They are again singing an interval, but more importantly, they are learning the relationship of two scale degrees to each other, and to tonic. As students learn intervals in theory, we gradually add in the interval names for the scale degrees. At this point, when students sing “ti-re,” they also sing “minor third”:
Example 1
ti re minor third
Lastly, we listen for the interval’s tendency to resolve, if there is one. In this case, the resolution is to tonic, so I would have the students sing tonic after they sing the interval [7].
No random intervals
In addition, and I believe this to be a crucial point, I don’t randomly choose any two scale degrees. I choose scale degree pairs where the corresponding interval is most often, or very often encountered. And, I might add, I don’t choose them for the sake of the interval, but for the purpose of familiarizing students with important relationships in tonal music. So, for example, the ascending perfect fourth from sol to do is practiced frequently, because of its all important key-defining characteristic—the V to I relationship encountered so often in basslines (whether as a primary relationship or a secondary relationship: re -sol, for example, is still sol-do, but on a secondary level). Conversely, I rarely practice perfect fourths from ti to mi, because that fourth has little or no significance, at least in terms of frequency. (Mediant chords in a major key are rare as it is, and a tonicized mediant is rarer still. The most typical usage is of course in a circle-of-fifths progression.) The point I want to stress to my students is that even though most diatonic intervals can be found in several or many places in a key, certain intervals are encountered much more frequently in certain harmonies, or on certain scale degrees, than others. A minor seventh, for example, can be found diatonically in five different places, but its most important function is to define tonic, as part of a V7, from sol to fa (or acting as sol to fa on a secondary level).
Scale degree pairs I use most often
Here, then, is a chart to illustrate the scale degree pairs I use most often, along with their corresponding interval labels. (The scale degree pairs that are starred get a bit more attention than those that are not.)
I. Pairs used for ascending intervals found in major keys
m2: *Ti-do; mi-fa | M2: *Do-re; re-mi; *fa-sol; sol-la |
m3: *Ti-re; *re-fa; *mi-sol | M3: *Do-mi; *fa-la; *sol-ti |
P4: *Sol-do; do-fa; re-sol | |
P5: *Do-sol | |
TT: *Ti-fa; *fa-ti | |
m6: *Mi-do | M6: *Sol-mi |
m7: *Sol-fa | M7: Do-ti; fa-mi |
II. Pairs used for ascending intervals found in minor keys
m2: Re-me; *sol-le | M2: Me-fa |
m3: *Do-me; *fa-le | M3: *Me-sol |
m6: *Sol-me | M6: *Me-do |
d7: *Ti-le |
III. Pairs used for descending intervals found in major keys
m2: *Do-ti; *fa-mi | M2: *Re-do; *mi-re; sol-fa |
m3: *Re-ti; *fa-re; *sol-mi; do-la | M3: *Mi-do; *la-fa; *ti-sol |
PP4: *Do-sol; fa-do | |
P5: *Sol-do | |
m6: *Do-mi | M6: *Mi-sol |
m7: *Fa-sol | M7: Ti-do |
IV. Pairs used for descending intervals found in minor keys
m2: Me-re; *le-sol | M2: Fa-me |
m3: *Me-do; *le-fa | M3: *Sol-me |
m6: *Me-sol | M6: *Mi-sol |
d7: Le-ti |
Let me re-emphasize that I am not concerned with teaching intervals as discrete aural phenomena, but teaching important tonal patterns that can be given interval labels.
Triad types and harmonies
The next step in the process is to relate these intervals to triad types and harmonies. I begin this while students are learning about triads and harmonies in theory. Students first need to learn what syllables/scale degrees are in each diatonic triad. I in fact give them written quizzes on this, to encourage them to learn these quickly. We then start talking about what harmony a particular scale degree pair most likely belongs to. So, for instance, after establishing tonic, I may play ti-re. The students sing back “ti-re, minor third.” I then ask them what harmony this particular scale degree pair/interval most likely belongs to. The obvious answers are V and viio. We also consider the possibility that two different harmonies may be involved. If I instead played re-ti, then ii-V becomes a possibility as well. My hope is that with this type of approach, the skills learned in ear training will transfer more directly to a student’s work in theory and performance: if the student sees, for instance, an ascending perfect fourth in the bassline of a Baroque composition that he or she is studying, the student will think and hear sol-do, and recognize it as some kind of V-I relationship, whether primary or secondary.
I approach the study of triads in the same way. While I believe that a certain amount of “triad quality” recognition should be present, especially at the beginning stages, quality alone is not sufficient. Just as it is with intervals, so it is with triads. Like-quality triads, in the context of a key, may not sound the same at all. For example, take the case of the ii chord, a minor triad in a major key. In context, whether in root position or first inversion, this chord does not to my ear sound nearly as “minor” as either a minor tonic or a minor subdominant in a minor key. I believe Rameau’s assertion that a ii65 is really a subdominant chord with an added sixth, and by extension, that ii6 is a subdominant chord with a sixth replacing the fifth, and I really do experience it that way. Having the strong subdominant note in the bass with a major third above it gives this ii6 chord a quality somewhere between major and minor, and arguably closer to major. So very soon after introducing triads in theory and introducing the sound of major, minor, augmented, and diminished triads, we start identifying single triads in the context of a major or minor key. I again establish a key with a short cadence and have the students sing do frequently to keep that focal point in their inner ear. Then as a triad is arpeggiated, the students sing the syllables and identify the triad by Roman numeral, inversion and type. Their beginning chord repertoire includes the following chords and inversions:
Sight singing
I coordinate this harmony identification practice with their sight singing in several ways. One is by analyzing the melodies in their sight singing text before singing them, looking for any harmonies that they recognize. Another is by singing harmonic progressions, in the form of exercises I have given to the students on handouts [8]. One handout is a melodically written out circle-of-fifths progression in both a major and a minor key. A Roman numeral is placed under each harmony:
The purpose is to make explicit the harmony that the student is singing when he or she is singing a particular group of syllables. The emphasis is again on scale degrees and harmonies in context, not just quality of the triad or where the root is. An added benefit of this particular exercise is that it presents a good opportunity for students to see melodic lines as written-out chord progressions, or, stated another way, to see harmonies horizontalized. It helps reinforce the notion that, Bach chorales notwithstanding, most tonal melodies do not present a series of pitch classes to be harmonized individually, but rather successive groups of notes which belong to a single harmony. This type of harmonic practice has another benefit as well: it feeds right into the practice of taking harmonic dictation.
Harmonic dictation
Given the type of scale degree practice and triad/harmony practice that a student would have had up to this point, it makes sense to begin harmonic dictation as a logical extension of that practice. I introduce it when students are learning about harmonic progressions in their written theory; i.e., what chord successions are likely or unlikely in this style. We begin with basslines. Students hear a short bassline, sing it back on syllables, and determine the most likely chord progression or progressions that would fit that bassline. We discuss the possibilities of, say, a do-fa-sol bassline [9]. The most likely progressions are of course I-IV-V, or I-ii6-V. Without any further information, the second chord remains in question, but the first and third chords can reasonably be assumed to be I and V, respectively. The next step is to play the soprano. If do is the second note, then the second chord must be IV. If re is the second note, the second chord must be ii6.
Bass and soprano lines
Students work for a while exclusively with bass and soprano lines as a way of learning typical tonal patterns or cliches that suggest certain chord progressions, and also as a way of learning to distinguish bass and soprano lines from each other, and from the other notes that will be present later in three- and four-voice chords. When only two voices are played, triad quality cannot be ascertained with 100% accuracy (i.e., is it IV or ii6 if la is in the soprano?), but students learn instead to concentrate on what the possible chord choices are, given a particular bass and soprano line and a tonal context, and they realize that the choices are limited indeed. Naturally, when other voices are added, chord quality becomes a factor, and provides additional information, but it is neither the only information nor the most important information to be processed. I discourage listening primarily for chord quality and location of the root since that encourages vertical listening and thinking, almost to the exclusion, it would seem, of hearing the individual lines in a musical texture. Rather, by combining the various pieces of information that the student has (bass and soprano notes, knowledge of typical tonal chord progressions, and the chord qualities), the student can make an informed decision that will be at worst, reasonably accurate. Obviously, the more information the student has, and the more the student knows about what to expect given this information, the more the student will be able to recognize and understand what he or she has heard. Working in stages by playing just the bass line first, adding in the soprano next, adding in the complete chord last, and stressing the need to integrate this aural information with the information being taught in written theory helps students tremendously in working with chord dictation.
Without a systematic approach, such as the one I am proposing, students often feel frustrated and lost when trying to do chord dictation. They do not know where to begin or how to determine what any particular chord or chord progression might be. They often do not realize that in a typical, tonal 5-chord progression (e.g., I – ii6 – I6sus4 – V – I), not every possible chord is a choice for each of the five chords. But knowing the bass and soprano note for every chord eliminates many of the possibilities, and knowing something about musical grammar and syntax narrows down the choices even further. What remains are reasonable choices. If fa is in the bass and la is in the soprano and the student still can not discern whether the chord is ii6 or IV, the error is not so egregious. Both would be equally likely and possible, and distinguishing between them is not nearly as crucial as between, say, a ii6 and a vi6sus4, chords that are neither interchangeable nor equally likely to appear in any given context.
Integrate aural information with theoretical information
As a cautionary note, I would like to paraphrase Karpinski and point out that my goal when teaching harmonic dictation is not to produce musical shorthand takers [11]. It is to produce students who can hear a passage of music and recognize and understand what is happening harmonically. There is much to be gained by a student through the working out of the solution to the harmonic progression he or she heard. It is vitally important that students learn to use all the information they have when listening to music in order to know and understand what they are hearing. Probably more than any other exercise normally practiced in an ear training class, taking harmonic dictation forces students to integrate the aural information they are receiving with the theoretical information they already have in order to come up with the correct answer (or at least a very plausible answer). To illustrate, let’s assume a student hears a harmonic progression and is unsure of the fourth chord. The student must first remember what was heard and then sing it back in his or her head. Say in chord #4, the student heard le in the bass and fi in the soprano, and heard both notes move outward by half step to a dominant octave. Further, let’s assume that the student knows that the defining characteristic of an augmented sixth chord is the interval of an augmented sixth, which typically resolves outward to the dominant octave. Our student has a lot of information, both aural and theoretical, on which to base his or her decision. Bringing all that information to bear on this progression, our student can reasonably assume that the chord in question was an augmented sixth. On a subsequent playing the student will be expecting an augmented sixth chord and will listen with that expectation and perhaps now hear at least two chords as a musical unit: the augmented sixth and the V to which it resolves, and which, ipso facto, helps him or her to hear the augmented sixth as an augmented sixth.
Summary
I have attempted to offer suggestions for some new strategies that I have found useful and successful in teaching ear training, especially regarding the thorny issue of how to teach intervals in ear training, if at all. In my own teaching, I have been particularly encouraged in my approach by students who have already had some high school or college level ear training, and who therefore have a basis for comparison. They comment that this type of systematic, contextual approach has been much more beneficial to them than the less methodical non-contextual approach they were exposed to previously. I continue to experiment and try out new strategies, and hope that the foregoing discussion will persuade others to do likewise.
References
[1] See, for example, William Thomson’s “What is an Interval?”, JMTP 2/2 (Fall 1988), pp. 321-324; and Michael Rogers’ “Beyond Intervals: the Teaching of Tonal Hearing,” Indiana Theory Review, 6/3 (Spring 1983), pp. 18-34.
[2] I believe the label “ear training” is a term that encompasses all the activities that take place in an ear training, or aural skills class, including sight singing. However, for the sake of clarity, my use of that term here will include all activities but sight-singing. I will always refer to sight-singing separately as such.
[3] Berkowitz, Sol, Frontrier, Gabriel and Kraft, Leo. A New Approach to Sight Singing, 3rd ed. (New York: W. W. Norton, 1986).
[4] For a review of recent sight singing and ear training texts, see Gary Karpinski’s “Five Recent Sight Singing Texts,” JMTP 2/2 (Fall 1988), pp. 275-296; and “Ear Training and Integrated Aural Skills: Three Recent Texts,” JMTP 3/1 (Spring 1989), pp. 127-149.
[5] This assertion would seem to be substantiated by, among other things, Gary Potter’s experiment, detailed in his article “Identifying Successful Dictation Strategies,”JMTP 4/1 (Spring 1990), pp. 63-71.
[6] Michael Rogers, op. cit.
[7] This sense of instability/resolution is not only an aural phenomenon, but is of course dependent in part on a knowledge of tonal chord progression. At this point, however, since students will not yet have encountered that topic, the discussion of tendencies is limited for the most part to the 1/2 step phenomenon: ti has a strong tendency to go to do, fa has a tendency to go to mi, and together they have a very strong tendency to go to the do-mi major third.
[8] I would like to thank Michael Rogers for this idea.
[9] The students must at this point have learned what syllables are in each diatonic triad.
[10] I say “must” because I’m working at an elementary level. In doing this type of practice, I do limit the students’ chord choices. I begin with a small repertoire of chords and inversions, and gradually add in new chords and inversions. I want students to be informed listeners, and so to that end I encourage students to make educated guesses, in keeping with the choices they have.Thus, neither ii56 nor viio46, for example, would be a possibility at this point. In addition, at these beginning stages, and even at a more advanced level, any time a new chord is introduced, I keep what I call a “determining” note in the soprano. By this I mean simply a note that makes clear what the chord is. If fa is in the bass, I will play do in the soprano for a IV chord, and re for a ii chord, making the chord choice unequivocal. La or fa in the soprano would not be an unequivocal choice, and so would not be used initially.
[11] Karpinski, Gary, “A Model for Music Perception and its Implication in Melodic Dictation,” JMTP 4/2 (Fall 1990), pp. 191-229.
Paula Telesco is Associate Professor at the University of Massachusetts Lowell. Her areas of specialty include the history of music theory, analysis of 18th century music, and the pedagogy of music theory and aural skills acquisition.
You can find more reading material from Paula in the Article and Resource section of Music Theory Pedagogy Online